fixed all test and pmd or checkstyle errors

buildSrc/src/main/kotlin/com.iqser.red.service.java-conventions.gradle.kts

@@ -1,24 +1,16 @@
 plugins {
-    java
+    `java-library`
+    `maven-publish`
     pmd
     checkstyle
     jacoco
 }
 
 group = "com.iqser.red"
-version = "4.0-SNAPSHOT"
 
 java.sourceCompatibility = JavaVersion.VERSION_17
 java.targetCompatibility = JavaVersion.VERSION_17
 
-tasks.jacocoTestReport {
-    reports {
-        xml.required.set(false)
-        csv.required.set(false)
-        html.outputLocation.set(layout.buildDirectory.dir("jacocoHtml"))
-    }
-}
-
 tasks.pmdMain {
     pmd.ruleSetFiles = files("${rootDir}/config/pmd/pmd.xml")
 }
@@ -29,6 +21,9 @@ tasks.pmdTest {
 
 tasks.named<Test>("test") {
     useJUnitPlatform()
+    reports {
+        junitXml.outputLocation.set(layout.buildDirectory.dir("reports/junit"))
+    }
 }
 
 tasks.test {
@@ -40,6 +35,26 @@ tasks.jacocoTestReport {
     reports {
         xml.required.set(true)
         csv.required.set(false)
+        html.outputLocation.set(layout.buildDirectory.dir("jacocoHtml"))
+    }
+}
+
+allprojects {
+    publishing {
+        publications {
+            create<MavenPublication>(name) {
+                from(components["java"])
+            }
+        }
+        repositories {
+            maven {
+                url = uri("https://nexus.knecon.com/repository/red-platform-releases/")
+                credentials {
+                    username = providers.gradleProperty("mavenUser").getOrNull();
+                    password = providers.gradleProperty("mavenPassword").getOrNull();
+                }
+            }
+        }
     }
 }
 

config/pmd/pmd.xml

@@ -12,7 +12,6 @@
         <exclude name="DataflowAnomalyAnalysis"/>
         <exclude name="MissingSerialVersionUID"/>
         <exclude name="NullAssignment"/>
-        <exclude name="BeanMembersShouldSerialize"/>
         <exclude name="AvoidLiteralsInIfCondition"/>
         <exclude name="AvoidDuplicateLiterals"/>
         <exclude name="AvoidFieldNameMatchingMethodName"/>

config/pmd/test_pmd.xml

@@ -13,12 +13,12 @@
         <exclude name="DataflowAnomalyAnalysis"/>
         <exclude name="MissingSerialVersionUID"/>
         <exclude name="NullAssignment"/>
-        <exclude name="BeanMembersShouldSerialize"/>
         <exclude name="AvoidLiteralsInIfCondition"/>
         <exclude name="AvoidDuplicateLiterals"/>
         <exclude name="AvoidFieldNameMatchingMethodName"/>
         <exclude name="AvoidFieldNameMatchingTypeName"/>
         <exclude name="AssignmentInOperand"/>
+        <exclude name="TestClassWithoutTestCases"/>
     </rule>
 
 </ruleset>

gradle.properties.kts

@@ -0,0 +1 @@
+version = 4.0-SNAPSHOT

redaction-service-v1/redaction-service-api-v1/build.gradle.kts

@@ -1,13 +1,14 @@
 
 plugins {
     id("com.iqser.red.service.java-conventions")
-    id("io.freefair.lombok") version "8.0.1"
+    id("io.freefair.lombok") version "8.1.0"
-}
-
-dependencies {
-    implementation("org.springframework:spring-web:6.0.6")
-    implementation("com.iqser.red.service:persistence-service-internal-api-v1:2.79.0")
 }
 
 description = "redaction-service-api-v1"
 
+dependencies {
+    implementation("org.springframework:spring-web:6.0.6")
+    implementation("com.iqser.red.service:persistence-service-internal-api-v1:RED-6725")
+}
+
+

redaction-service-v1/redaction-service-server-v1/build.gradle.kts

@@ -6,15 +6,21 @@ plugins {
     id("org.springframework.boot") version "3.0.6"
     id("io.spring.dependency-management") version "1.1.0"
     id("org.sonarqube") version "4.2.1.3168"
-    id("io.freefair.lombok") version "8.0.1"
+    id("io.freefair.lombok") version "8.1.0"
 }
 
+description = "redaction-service-server-v1"
+
+extra["layoutparserVersion"] = "0.8.0"
+
 dependencies {
     implementation(project(":redaction-service-api-v1"))
-    implementation("com.iqser.red.service:persistence-service-internal-api-v1:2.79.0")
+    implementation("com.iqser.red.service:persistence-service-internal-api-v1:RED-6725")
+    implementation("com.knecon.fforesight:layoutparser-service-internal-api:0.12.0")
 
     implementation("com.iqser.red.commons:spring-commons:2.1.0") { exclude(group = "org.springframework.boot", module = "spring-boot-starter-log4j2") }
     implementation("com.iqser.red.commons:metric-commons:2.1.0") { exclude(group = "org.springframework.boot", module = "spring-boot-starter-log4j2") }
+
     implementation("com.iqser.red.commons:dictionary-merge-commons:1.3.0")
     implementation("com.iqser.red.commons:storage-commons:2.1.0")
     implementation("com.knecon.fforesight:tenant-commons:0.10.0")
@@ -31,15 +37,15 @@ dependencies {
 
     implementation("org.locationtech.jts:jts-core:1.19.0")
 
-    implementation("org.apache.pdfbox:pdfbox:2.0.24")
+    testImplementation("org.apache.pdfbox:pdfbox:3.0.0-alpha2")
-    implementation("org.apache.pdfbox:pdfbox-tools:2.0.24")
+    testImplementation("org.apache.pdfbox:pdfbox-tools:3.0.0-alpha2")
 
     implementation("org.springframework.cloud:spring-cloud-starter-openfeign:4.0.1")
     implementation("org.springframework.boot:spring-boot-starter-amqp:3.0.4")
     testImplementation("org.springframework.boot:spring-boot-starter-test:3.0.4")
+    testImplementation("com.knecon.fforesight:layoutparser-service-processor:0.12.0") { exclude(group = "com.iqser.red.service", module = "persistence-service-shared-api-v1") }
 }
 
-description = "redaction-service-server-v1"
 
 
 tasks.named<BootBuildImage>("bootBuildImage") {

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/data/AtomicPositionBlockData.java

@@ -1,50 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.data;
-
-import java.awt.geom.Rectangle2D;
-import java.util.List;
-
-import com.iqser.red.service.redaction.v1.server.document.graph.textblock.AtomicTextBlock;
-
-import lombok.AccessLevel;
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.experimental.FieldDefaults;
-
-@Data
-@Builder
-@AllArgsConstructor
-@FieldDefaults(makeFinal = true, level = AccessLevel.PRIVATE)
-public class AtomicPositionBlockData {
-
-    Long id;
-    int[] stringIdxToPositionIdx;
-    float[][] positions;
-
-
-    public static AtomicPositionBlockData fromAtomicTextBlock(AtomicTextBlock atomicTextBlock) {
-
-        return AtomicPositionBlockData.builder()
-                .id(atomicTextBlock.getId())
-                .positions(toPrimitiveFloatMatrix(atomicTextBlock.getPositions()))
-                .stringIdxToPositionIdx(atomicTextBlock.getStringIdxToPositionIdx().stream().mapToInt(Integer::intValue).toArray())
-                .build();
-    }
-
-
-    private static float[][] toPrimitiveFloatMatrix(List<Rectangle2D> positions) {
-
-        float[][] positionMatrix = new float[positions.size()][];
-        for (int i = 0; i < positions.size(); i++) {
-            positionMatrix[i] = toArray(positions.get(i));
-        }
-        return positionMatrix;
-    }
-
-
-    private static float[] toArray(Rectangle2D positions) {
-
-        return new float[]{(float) positions.getMinX(), (float) positions.getMinY(), (float) positions.getWidth(), (float) positions.getHeight()};
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/data/AtomicTextBlockData.java

@@ -1,39 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.data;
-
-import com.iqser.red.service.redaction.v1.server.document.graph.textblock.AtomicTextBlock;
-
-import lombok.AccessLevel;
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.experimental.FieldDefaults;
-
-@Data
-@Builder
-@AllArgsConstructor
-@FieldDefaults(makeFinal = true, level = AccessLevel.PRIVATE)
-public class AtomicTextBlockData {
-
-    Long id;
-    Long page;
-    String searchText;
-    int numberOnPage;
-    int start;
-    int end;
-    int[] lineBreaks;
-
-
-    public static AtomicTextBlockData fromAtomicTextBlock(AtomicTextBlock atomicTextBlock) {
-
-        return AtomicTextBlockData.builder()
-                .id(atomicTextBlock.getId())
-                .page(atomicTextBlock.getPage().getNumber().longValue())
-                .searchText(atomicTextBlock.getSearchText())
-                .numberOnPage(atomicTextBlock.getNumberOnPage())
-                .start(atomicTextBlock.getBoundary().start())
-                .end(atomicTextBlock.getBoundary().end())
-                .lineBreaks(atomicTextBlock.getLineBreaks().stream().mapToInt(Integer::intValue).toArray())
-                .build();
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/data/DocumentData.java

@@ -1,6 +1,9 @@
 package com.iqser.red.service.redaction.v1.server.document.data;
 
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentPage;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentPositionData;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentStructure;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentTextData;
 
 import lombok.AccessLevel;
 import lombok.AllArgsConstructor;
@@ -14,30 +17,9 @@ import lombok.experimental.FieldDefaults;
 @FieldDefaults(makeFinal = true, level = AccessLevel.PRIVATE)
 public class DocumentData {
 
-    PageData[] pages;
+    DocumentPage[] documentPages;
-    AtomicTextBlockData[] atomicTextBlocks;
+    DocumentTextData[] documentTextData;
-    AtomicPositionBlockData[] atomicPositionBlocks;
+    DocumentPositionData[] documentPositionData;
-    DocumentTreeData documentTreeData;
+    DocumentStructure documentStructure;
-
-
-    public static DocumentData fromDocument(Document document) {
-
-        var atomicPositionBlocks = document.streamTerminalTextBlocksInOrder()
-                .flatMap(textBlock -> textBlock.getAtomicTextBlocks().stream())
-                .distinct()
-                .map(AtomicPositionBlockData::fromAtomicTextBlock)
-                .toArray(AtomicPositionBlockData[]::new);
-
-        var atomicTextBlocks = document.streamTerminalTextBlocksInOrder()
-                .flatMap(textBlock -> textBlock.getAtomicTextBlocks().stream())
-                .distinct()
-                .map(AtomicTextBlockData::fromAtomicTextBlock)
-                .toArray(AtomicTextBlockData[]::new);
-
-        var pages = document.getPages().stream().map(PageData::fromPage).toArray(PageData[]::new);
-
-        var documentTreeData = new DocumentTreeData(DocumentTreeData.EntryData.fromEntry(document.getDocumentTree().getRoot()));
-        return new DocumentData(pages, atomicTextBlocks, atomicPositionBlocks, documentTreeData);
-    }
 
 }

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/data/DocumentTreeData.java

@@ -1,128 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.data;
-
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-import java.util.stream.Stream;
-
-import com.iqser.red.service.redaction.v1.server.document.data.mapper.PropertiesMapper;
-import com.iqser.red.service.redaction.v1.server.document.graph.DocumentTree;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Image;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.TableCell;
-import com.iqser.red.service.redaction.v1.server.document.graph.textblock.AtomicTextBlock;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Table;
-
-import lombok.AccessLevel;
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.Getter;
-import lombok.NoArgsConstructor;
-import lombok.experimental.FieldDefaults;
-
-@Data
-@Builder
-@AllArgsConstructor
-@NoArgsConstructor
-@FieldDefaults(level = AccessLevel.PRIVATE)
-public class DocumentTreeData {
-
-    EntryData root;
-
-
-    public EntryData get(List<Integer> tocId) {
-
-        if (tocId.isEmpty()) {
-            return root;
-        }
-        EntryData entry = root.children.get(tocId.get(0));
-        for (int id : tocId.subList(1, tocId.size())) {
-            entry = entry.children.get(id);
-        }
-        return entry;
-    }
-
-
-    public Stream<EntryData> streamAllEntries() {
-
-        return Stream.concat(Stream.of(root), root.children.stream()).flatMap(DocumentTreeData::flatten);
-    }
-
-
-    public String toString() {
-
-        return String.join("\n", streamAllEntries().map(EntryData::toString).toList());
-    }
-
-
-    private static Stream<EntryData> flatten(EntryData entry) {
-
-        return Stream.concat(Stream.of(entry), entry.children.stream().flatMap(DocumentTreeData::flatten));
-    }
-
-
-    @Builder
-    @Getter
-    @AllArgsConstructor
-    @FieldDefaults(makeFinal = true, level = AccessLevel.PRIVATE)
-    public static class EntryData {
-
-        NodeType type;
-        int[] treeId;
-        Long[] atomicBlockIds;
-        Long[] pageNumbers;
-        Map<String, String> properties;
-        List<EntryData> children;
-
-
-        public static EntryData fromEntry(DocumentTree.Entry entry) {
-
-            Long[] atomicBlockIds = toAtomicTextBlockIds(entry);
-
-            Map<String, String> properties = switch (entry.getType()) {
-                case TABLE -> PropertiesMapper.buildTableProperties((Table) entry.getNode());
-                case TABLE_CELL -> PropertiesMapper.buildTableCellProperties((TableCell) entry.getNode());
-                case IMAGE -> PropertiesMapper.buildImageProperties((Image) entry.getNode());
-                default -> new HashMap<>();
-            };
-            var treeId = entry.getTreeId().stream().mapToInt(Integer::intValue).toArray();
-            var pageNumbers = entry.getNode().getPages().stream().map(Page::getNumber).map(Integer::longValue).toArray(Long[]::new);
-            var subEntries = entry.getChildren().stream().map(EntryData::fromEntry).toList();
-            return new EntryData(entry.getType(), treeId, atomicBlockIds, pageNumbers, properties, subEntries);
-        }
-
-
-        private static Long[] toAtomicTextBlockIds(DocumentTree.Entry entry) {
-
-            if (entry.getNode().isLeaf()) {
-                return entry.getNode().getLeafTextBlock().getAtomicTextBlocks().stream().map(AtomicTextBlock::getId).toArray(Long[]::new);
-            } else {
-                return new Long[]{};
-            }
-        }
-
-
-        @Override
-        public String toString() {
-
-            StringBuilder sb = new StringBuilder();
-            sb.append("[");
-            for (int i : treeId) {
-                sb.append(i);
-                sb.append(",");
-            }
-            sb.delete(sb.length() - 1, sb.length());
-            sb.append("]: ");
-
-            sb.append(type);
-            sb.append(" atbs = ");
-            sb.append(atomicBlockIds.length);
-
-            return sb.toString();
-        }
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/data/PageData.java

@@ -1,28 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.data;
-
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Page;
-
-import lombok.AccessLevel;
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.experimental.FieldDefaults;
-
-@Data
-@Builder
-@AllArgsConstructor
-@FieldDefaults(makeFinal = true, level = AccessLevel.PRIVATE)
-public class PageData {
-
-    int number;
-    int height;
-    int width;
-    int rotation;
-
-
-    public static PageData fromPage(Page page) {
-
-        return new PageData(page.getNumber(), page.getHeight(), page.getWidth(), page.getRotation());
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/data/mapper/DocumentGraphMapper.java

@@ -19,14 +19,14 @@ import com.iqser.red.service.redaction.v1.server.document.graph.nodes.TableCell;
 import com.iqser.red.service.redaction.v1.server.document.graph.textblock.AtomicTextBlock;
 import com.iqser.red.service.redaction.v1.server.document.graph.textblock.TextBlock;
 import com.iqser.red.service.redaction.v1.server.document.graph.textblock.TextBlockCollector;
-import com.iqser.red.service.redaction.v1.server.document.data.AtomicPositionBlockData;
-import com.iqser.red.service.redaction.v1.server.document.data.AtomicTextBlockData;
 import com.iqser.red.service.redaction.v1.server.document.data.DocumentData;
-import com.iqser.red.service.redaction.v1.server.document.data.DocumentTreeData;
-import com.iqser.red.service.redaction.v1.server.document.data.PageData;
 import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Document;
 import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Headline;
 import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentPage;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentPositionData;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentStructure;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentTextData;
 
 import lombok.experimental.UtilityClass;
 
@@ -39,23 +39,23 @@ public class DocumentGraphMapper {
         DocumentTree documentTree = new DocumentTree(document);
         Context context = new Context(documentData, documentTree);
 
-        context.pages.addAll(Arrays.stream(documentData.getPages()).map(DocumentGraphMapper::buildPage).toList());
+        context.pageData.addAll(Arrays.stream(documentData.getDocumentPages()).map(DocumentGraphMapper::buildPage).toList());
 
-        context.documentTree.getRoot().getChildren().addAll(buildEntries(documentData.getDocumentTreeData().getRoot().getChildren(), context));
+        context.documentTree.getRoot().getChildren().addAll(buildEntries(documentData.getDocumentStructure().getRoot().getChildren(), context));
 
         document.setDocumentTree(context.documentTree);
-        document.setPages(new HashSet<>(context.pages));
+        document.setPages(new HashSet<>(context.pageData));
-        document.setNumberOfPages(documentData.getPages().length);
+        document.setNumberOfPages(documentData.getDocumentPages().length);
 
         document.setTextBlock(document.getTextBlock());
         return document;
     }
 
 
-    private List<DocumentTree.Entry> buildEntries(List<DocumentTreeData.EntryData> entries, Context context) {
+    private List<DocumentTree.Entry> buildEntries(List<DocumentStructure.EntryData> entries, Context context) {
 
         List<DocumentTree.Entry> newEntries = new LinkedList<>();
-        for (DocumentTreeData.EntryData entryData : entries) {
+        for (DocumentStructure.EntryData entryData : entries) {
 
             List<Page> pages = Arrays.stream(entryData.getPageNumbers()).map(pageNumber -> getPage(pageNumber, context)).toList();
 
@@ -154,14 +154,14 @@ public class DocumentGraphMapper {
 
     private AtomicTextBlock getAtomicTextBlock(Context context, SemanticNode parent, Long atomicTextBlockId) {
 
-        return AtomicTextBlock.fromAtomicTextBlockData(context.atomicTextBlockData.get(Math.toIntExact(atomicTextBlockId)),
+        return AtomicTextBlock.fromAtomicTextBlockData(context.documentTextData.get(Math.toIntExact(atomicTextBlockId)),
-                context.atomicPositionBlockData.get(Math.toIntExact(atomicTextBlockId)),
+                context.documentPositionData.get(Math.toIntExact(atomicTextBlockId)),
                 parent,
-                getPage(context.atomicTextBlockData.get(Math.toIntExact(atomicTextBlockId)).getPage(), context));
+                getPage(context.documentTextData.get(Math.toIntExact(atomicTextBlockId)).getPage(), context));
     }
 
 
-    private Page buildPage(PageData p) {
+    private Page buildPage(DocumentPage p) {
 
         return Page.builder().rotation(p.getRotation()).height(p.getHeight()).width(p.getWidth()).number(p.getNumber()).mainBody(new LinkedList<>()).build();
     }
@@ -169,7 +169,7 @@ public class DocumentGraphMapper {
 
     private Page getPage(Long pageIndex, Context context) {
 
-        return context.pages.stream()
+        return context.pageData.stream()
                 .filter(page -> page.getNumber() == Math.toIntExact(pageIndex))
                 .findFirst()
                 .orElseThrow(() -> new NoSuchElementException(String.format("ClassificationPage with number %d not found", pageIndex)));
@@ -179,17 +179,17 @@ public class DocumentGraphMapper {
     static final class Context {
 
         private final DocumentTree documentTree;
-        private final List<Page> pages;
+        private final List<Page> pageData;
-        private final List<AtomicTextBlockData> atomicTextBlockData;
+        private final List<DocumentTextData> documentTextData;
-        private final List<AtomicPositionBlockData> atomicPositionBlockData;
+        private final List<DocumentPositionData> documentPositionData;
 
 
         Context(DocumentData documentData, DocumentTree documentTree) {
 
             this.documentTree = documentTree;
-            this.pages = new LinkedList<>();
+            this.pageData = new LinkedList<>();
-            this.atomicTextBlockData = Arrays.stream(documentData.getAtomicTextBlocks()).toList();
+            this.documentTextData = Arrays.stream(documentData.getDocumentTextData()).toList();
-            this.atomicPositionBlockData = Arrays.stream(documentData.getAtomicPositionBlocks()).toList();
+            this.documentPositionData = Arrays.stream(documentData.getDocumentPositionData()).toList();
 
         }
 

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/factory/DocumentGraphFactory.java

@@ -1,242 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.factory;
-
-import static java.lang.String.format;
-import static java.util.stream.Collectors.groupingBy;
-import static java.util.stream.Collectors.toList;
-
-import java.awt.geom.Rectangle2D;
-import java.util.ArrayList;
-import java.util.HashMap;
-import java.util.LinkedList;
-import java.util.List;
-import java.util.Map;
-import java.util.NoSuchElementException;
-import java.util.Set;
-
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Footer;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.GenericSemanticNode;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Paragraph;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Section;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationFooter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationHeader;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.document.graph.DocumentTree;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Header;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Headline;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Image;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.document.graph.textblock.AtomicTextBlock;
-import com.iqser.red.service.redaction.v1.server.document.utils.TextPositionOperations;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.IdBuilder;
-
-import lombok.AccessLevel;
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Getter;
-import lombok.experimental.FieldDefaults;
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public class DocumentGraphFactory {
-
-    public Document buildDocumentGraph(ClassificationDocument document) {
-
-        Document documentGraph = new Document();
-        Context context = new Context(documentGraph);
-
-        document.getPages().forEach(context::buildAndAddPageWithCounter);
-        document.getSections().stream().flatMap(section -> section.getImages().stream()).forEach(image -> context.getImages().add(image));
-        addSections(document, context);
-        addHeaderAndFooterToEachPage(document, context);
-
-        documentGraph.setNumberOfPages(context.pages.size());
-        documentGraph.setPages(context.pages.keySet());
-        documentGraph.setDocumentTree(context.documentTree);
-        documentGraph.setTextBlock(documentGraph.getTextBlock());
-        return documentGraph;
-    }
-
-
-    private void addSections(ClassificationDocument document, Context context) {
-
-        document.getSections().forEach(section -> SectionNodeFactory.addSection(null, section.getPageBlocks(), section.getImages(), context));
-    }
-
-
-    public void addParagraphOrHeadline(GenericSemanticNode parentNode, TextPageBlock originalTextBlock, Context context, List<TextPageBlock> textBlocksToMerge) {
-
-        Page page = context.getPage(originalTextBlock.getPage());
-
-        GenericSemanticNode node;
-        if (originalTextBlock.isHeadline()) {
-            node = Headline.builder().documentTree(context.getDocumentTree()).build();
-        } else {
-            node = Paragraph.builder().documentTree(context.getDocumentTree()).build();
-        }
-
-        page.getMainBody().add(node);
-
-        List<TextPageBlock> textBlocks = new ArrayList<>(textBlocksToMerge);
-        AtomicTextBlock textBlock = context.textBlockFactory.fromContext(TextPositionOperations.mergeAndSortTextPositionSequenceByYThenX(textBlocks), node, context, page);
-        List<Integer> treeId = context.documentTree.createNewChildEntryAndReturnId(parentNode, node);
-        node.setLeafTextBlock(textBlock);
-        node.setTreeId(treeId);
-    }
-
-
-    public void addImage(Section section, ClassifiedImage image, Context context) {
-
-        Rectangle2D position = image.getPosition();
-        Page page = context.getPage(image.getPage());
-        Image imageNode = Image.builder()
-                .id(IdBuilder.buildId(Set.of(page), List.of(position), image.getImageType().name(), "image"))
-                .imageType(image.getImageType())
-                .position(position)
-                .transparent(image.isHasTransparency())
-                .page(page)
-                .documentTree(context.getDocumentTree())
-                .build();
-        page.getMainBody().add(imageNode);
-
-        List<Integer> tocId = context.getDocumentTree().createNewChildEntryAndReturnId(section, imageNode);
-        imageNode.setTreeId(tocId);
-    }
-
-
-    private void addHeaderAndFooterToEachPage(ClassificationDocument document, Context context) {
-
-        Map<Integer, List<TextPageBlock>> headers = document.getHeaders()
-                .stream()
-                .map(ClassificationHeader::getTextBlocks)
-                .flatMap(List::stream)
-                .collect(groupingBy(AbstractPageBlock::getPage, toList()));
-
-        Map<Integer, List<TextPageBlock>> footers = document.getFooters()
-                .stream()
-                .map(ClassificationFooter::getTextBlocks)
-                .flatMap(List::stream)
-                .collect(groupingBy(AbstractPageBlock::getPage, toList()));
-
-        for (int pageIndex = 1; pageIndex <= document.getPages().size(); pageIndex++) {
-            if (headers.containsKey(pageIndex)) {
-                addHeader(headers.get(pageIndex), context);
-            } else {
-                addEmptyHeader(pageIndex, context);
-            }
-        }
-
-        for (int pageIndex = 1; pageIndex <= document.getPages().size(); pageIndex++) {
-            if (footers.containsKey(pageIndex)) {
-                addFooter(footers.get(pageIndex), context);
-            } else {
-                addEmptyFooter(pageIndex, context);
-            }
-        }
-    }
-
-
-    private void addFooter(List<TextPageBlock> textBlocks, Context context) {
-
-        Page page = context.getPage(textBlocks.get(0).getPage());
-        Footer footer = Footer.builder().documentTree(context.getDocumentTree()).build();
-        AtomicTextBlock textBlock = context.textBlockFactory.fromContext(TextPositionOperations.mergeAndSortTextPositionSequenceByYThenX(textBlocks), footer, context, page);
-        List<Integer> tocId = context.getDocumentTree().createNewMainEntryAndReturnId(footer);
-        footer.setTreeId(tocId);
-        footer.setLeafTextBlock(textBlock);
-        page.setFooter(footer);
-    }
-
-
-    public void addHeader(List<TextPageBlock> textBlocks, Context context) {
-
-        Page page = context.getPage(textBlocks.get(0).getPage());
-        Header header = Header.builder().documentTree(context.getDocumentTree()).build();
-        AtomicTextBlock textBlock = context.textBlockFactory.fromNumberOnPage(TextPositionOperations.mergeAndSortTextPositionSequenceByYThenX(textBlocks), header, 0, page);
-        List<Integer> tocId = context.getDocumentTree().createNewMainEntryAndReturnId(header);
-        header.setTreeId(tocId);
-        header.setLeafTextBlock(textBlock);
-        page.setHeader(header);
-    }
-
-
-    private void addEmptyFooter(int pageIndex, Context context) {
-
-        Page page = context.getPage(pageIndex);
-        Footer footer = Footer.builder().documentTree(context.getDocumentTree()).build();
-        AtomicTextBlock textBlock = context.textBlockFactory.emptyTextBlock(footer, context, page);
-        List<Integer> tocId = context.getDocumentTree().createNewMainEntryAndReturnId(footer);
-        footer.setTreeId(tocId);
-        footer.setLeafTextBlock(textBlock);
-        page.setFooter(footer);
-    }
-
-
-    private void addEmptyHeader(int pageIndex, Context context) {
-
-        Page page = context.getPage(pageIndex);
-        Header header = Header.builder().documentTree(context.getDocumentTree()).build();
-        AtomicTextBlock textBlock = context.textBlockFactory.emptyTextBlockFromInteger(header, 0, page);
-        List<Integer> tocId = context.getDocumentTree().createNewMainEntryAndReturnId(header);
-        header.setTreeId(tocId);
-        header.setLeafTextBlock(textBlock);
-        page.setHeader(header);
-    }
-
-
-    @Getter
-    @Builder
-    @AllArgsConstructor
-    @FieldDefaults(makeFinal = true, level = AccessLevel.PRIVATE)
-    public final class Context {
-
-        DocumentTree documentTree;
-        Map<Page, Integer> pages;
-        List<Section> sections;
-        List<ClassifiedImage> images;
-        TextBlockFactory textBlockFactory;
-
-
-        public Context(Document document) {
-
-            documentTree = new DocumentTree(document);
-            pages = new HashMap<>();
-            sections = new LinkedList<>();
-            images = new LinkedList<>();
-            textBlockFactory = new TextBlockFactory();
-        }
-
-
-        public void buildAndAddPageWithCounter(ClassificationPage classificationPage) {
-
-            Page page = Page.fromClassificationPage(classificationPage);
-            //this counter counts the TextBlocks per page
-            //initial value is set to 1, because 0 is reserved for Header
-            pages.put(page, 1);
-        }
-
-
-        public int getAndIncrementTextBlockNumberOnPage(Page page) {
-
-            Integer textBlockNumberOnPage = pages.get(page);
-            pages.merge(page, 1, Integer::sum);
-            return textBlockNumberOnPage;
-        }
-
-
-        public Page getPage(int pageIndex) {
-
-            return pages.keySet()
-                    .stream()
-                    .filter(page -> page.getNumber() == pageIndex)
-                    .findFirst()
-                    .orElseThrow(() -> new NoSuchElementException(format("ClassificationPage with number %d not found", pageIndex)));
-        }
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/factory/SearchTextWithTextPositionDto.java

@@ -1,33 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.factory;
-
-import java.awt.geom.Rectangle2D;
-import java.util.Collections;
-import java.util.List;
-
-import lombok.AccessLevel;
-import lombok.Builder;
-import lombok.Getter;
-import lombok.experimental.FieldDefaults;
-
-@Builder
-@Getter
-@FieldDefaults(makeFinal = true, level = AccessLevel.PRIVATE)
-public class SearchTextWithTextPositionDto {
-
-    String searchText;
-    List<Integer> lineBreaks;
-    List<Integer> stringCoordsToPositionCoords;
-    List<Rectangle2D> positions;
-
-
-    public static SearchTextWithTextPositionDto empty() {
-
-        return SearchTextWithTextPositionDto.builder()
-                .searchText("")
-                .lineBreaks(Collections.emptyList())
-                .positions(Collections.emptyList())
-                .stringCoordsToPositionCoords(Collections.emptyList())
-                .build();
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/factory/SearchTextWithTextPositionFactory.java

@@ -1,185 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.factory;
-
-import java.awt.geom.AffineTransform;
-import java.awt.geom.Rectangle2D;
-import java.util.LinkedList;
-import java.util.List;
-import java.util.Objects;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.RedTextPosition;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextDirection;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public class SearchTextWithTextPositionFactory {
-
-    public final int HEIGHT_PADDING = 2;
-    // when checking for a hyphen linebreak, we need to check after a linebreak if the last hyphen was less than three symbols away.
-    // We detect a linebreak as either a "\n" character or if two adjacent symbol's position differ in y-coordinates by at least one character height.
-    // If there is a hyphen linebreak, the hyphen will be 1 position in front of a "\n" or 2 positions in front of the character which has a lower y-coordinate
-    // This is why, we need to initialize this to < -2, otherwise, if the very first symbol is a \n we would detect a hyphen linebreak that isn't there.
-    // Also, Integer.MIN_VALUE is a bad idea due to potential overflow during arithmetic operations. This is why the default should be -3.
-    public final int MAX_HYPHEN_LINEBREAK_DISTANCE = 3;
-
-
-    public SearchTextWithTextPositionDto buildSearchTextToTextPositionModel(List<TextPositionSequence> sequences) {
-
-        if (sequences.isEmpty() || sequences.stream().allMatch(sequence -> sequence.getTextPositions().isEmpty())) {
-            return SearchTextWithTextPositionDto.empty();
-        }
-
-        Context context = new Context();
-
-        RedTextPosition currentTextPosition = sequences.get(0).getTextPositions().get(0);
-        RedTextPosition previousTextPosition = RedTextPosition.builder().unicode(" ").position(currentTextPosition.getPosition()).build();
-
-        for (TextPositionSequence word : sequences) {
-            for (int i = 0; i < word.getTextPositions().size(); ++i) {
-
-                currentTextPosition = word.getTextPositions().get(i);
-                if (isLineBreak(currentTextPosition, previousTextPosition)) {
-                    removeHyphenLinebreaks(context);
-                    context.lineBreaksStringIdx.add(context.stringIdx);
-                }
-                if (!isRepeatedWhitespace(currentTextPosition.getUnicode(), previousTextPosition.getUnicode())) {
-                    if (isHyphen(currentTextPosition.getUnicode())) {
-                        context.lastHyphenIdx = context.stringIdx;
-                    }
-                    appendCurrentTextPosition(context, currentTextPosition);
-                }
-
-                previousTextPosition = currentTextPosition;
-                ++context.positionIdx;
-            }
-
-            previousTextPosition = RedTextPosition.builder().unicode(" ").position(previousTextPosition.getPosition()).build();
-            context.stringBuilder.append(" ");
-            context.stringIdxToPositionIdx.add(context.positionIdx);
-            ++context.stringIdx;
-        }
-
-        assert context.stringBuilder.length() == context.stringIdxToPositionIdx.size();
-
-        List<Rectangle2D> positions = sequences.stream()
-                .flatMap(sequence -> sequence.getTextPositions().stream().map(textPosition -> mapRedTextPositionToInitialUserSpace(textPosition, sequence)))
-                .toList();
-
-        return SearchTextWithTextPositionDto.builder()
-                .searchText(context.stringBuilder.toString())
-                .lineBreaks(context.lineBreaksStringIdx)
-                .stringCoordsToPositionCoords(context.stringIdxToPositionIdx)
-                .positions(positions)
-                .build();
-    }
-
-
-    private void appendCurrentTextPosition(Context context, RedTextPosition currentTextPosition) {
-
-        context.stringBuilder.append(currentTextPosition.getUnicode());
-
-        // unicode characters with more than 16-bit encoding have a length > 1 in java strings
-        for (int j = 0; j < currentTextPosition.getUnicode().length(); j++) {
-            context.stringIdxToPositionIdx.add(context.positionIdx);
-        }
-        context.stringIdx += currentTextPosition.getUnicode().length();
-    }
-
-
-    private void removeHyphenLinebreaks(Context context) {
-
-        if (lastHyphenDirectlyBeforeLineBreak(context)) {
-            context.stringBuilder.delete(context.lastHyphenIdx, context.stringBuilder.length());
-            context.stringIdxToPositionIdx = context.stringIdxToPositionIdx.subList(0, context.lastHyphenIdx);
-            context.stringIdx = context.lastHyphenIdx;
-            context.lastHyphenIdx = -MAX_HYPHEN_LINEBREAK_DISTANCE;
-        }
-    }
-
-
-    private boolean lastHyphenDirectlyBeforeLineBreak(Context context) {
-
-        return context.stringIdx - context.lastHyphenIdx < MAX_HYPHEN_LINEBREAK_DISTANCE;
-    }
-
-
-    private boolean isLineBreak(RedTextPosition currentTextPosition, RedTextPosition previousTextPosition) {
-
-        return Objects.equals(currentTextPosition.getUnicode(), "\n") || isDeltaYLargerThanTextHeight(currentTextPosition, previousTextPosition);
-    }
-
-
-    private boolean isDeltaYLargerThanTextHeight(RedTextPosition currentPosition, RedTextPosition previousPosition) {
-
-        if (previousPosition == null) {
-            return false;
-        }
-
-        float deltaY = Math.abs(currentPosition.getYDirAdj() - previousPosition.getYDirAdj());
-        return deltaY >= currentPosition.getHeightDir();
-    }
-
-
-    private boolean isRepeatedWhitespace(String currentUnicode, String previousUnicode) {
-
-        return Objects.equals(previousUnicode, " ") && Objects.equals(currentUnicode, " ");
-    }
-
-
-    private boolean isHyphen(String unicodeCharacter) {
-
-        return Objects.equals(unicodeCharacter, "-") || //
-                Objects.equals(unicodeCharacter, "~") || //
-                Objects.equals(unicodeCharacter, "‐") || //
-                Objects.equals(unicodeCharacter, "‒") || //
-                Objects.equals(unicodeCharacter, "⁻") || //
-                Objects.equals(unicodeCharacter, "−") || //
-                Objects.equals(unicodeCharacter, "﹣") || //
-                Objects.equals(unicodeCharacter, "゠") || //
-                Objects.equals(unicodeCharacter, "⁓") || //
-                Objects.equals(unicodeCharacter, "‑") || //
-                Objects.equals(unicodeCharacter, "\u00AD");
-    }
-
-
-    private Rectangle2D mapRedTextPositionToInitialUserSpace(RedTextPosition textPosition, TextPositionSequence sequence) {
-
-        float textHeight = sequence.getTextHeight() + HEIGHT_PADDING;
-        Rectangle2D rectangle2D = new Rectangle2D.Double(textPosition.getXDirAdj(),
-                textPosition.getYDirAdj() - textHeight,
-                textPosition.getWidthDirAdj(),
-                textHeight + HEIGHT_PADDING);
-
-        AffineTransform transform = new AffineTransform();
-
-        if (sequence.getDir() == TextDirection.ZERO || sequence.getDir() == TextDirection.HALF_CIRCLE) {
-            transform.rotate(sequence.getDir().getRadians(), sequence.getPageWidth() / 2f, sequence.getPageHeight() / 2f);
-            transform.translate(0f, sequence.getPageHeight());
-        } else if (sequence.getDir() == TextDirection.QUARTER_CIRCLE) {
-            transform.rotate(sequence.getDir().getRadians(), sequence.getPageWidth() / 2f, sequence.getPageWidth() / 2f);
-            transform.translate(0f, sequence.getPageWidth());
-        } else {
-            transform.rotate(sequence.getDir().getRadians(), sequence.getPageHeight() / 2f, sequence.getPageHeight() / 2f);
-            transform.translate(0f, sequence.getPageWidth());
-        }
-        transform.scale(1., -1.);
-
-        return transform.createTransformedShape(rectangle2D).getBounds2D();
-    }
-
-
-    private class Context {
-
-        List<Integer> stringIdxToPositionIdx = new LinkedList<>();
-        List<Integer> lineBreaksStringIdx = new LinkedList<>();
-        StringBuilder stringBuilder = new StringBuilder();
-
-        int stringIdx;
-        int positionIdx;
-
-        int lastHyphenIdx = -MAX_HYPHEN_LINEBREAK_DISTANCE;
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/factory/SectionNodeFactory.java

@@ -1,204 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.factory;
-
-import static java.lang.String.format;
-import static java.util.Collections.emptyList;
-import static java.util.stream.Collectors.groupingBy;
-
-import java.util.HashSet;
-import java.util.LinkedList;
-import java.util.List;
-import java.util.Map;
-import java.util.Set;
-import java.util.stream.Stream;
-
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.GenericSemanticNode;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Section;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.document.utils.TableMergingUtility;
-
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public class SectionNodeFactory {
-
-    public void addSection(GenericSemanticNode parentNode, List<AbstractPageBlock> pageBlocks, List<ClassifiedImage> images, DocumentGraphFactory.Context context) {
-
-        if (pageBlocks.isEmpty()) {
-            return;
-        }
-        Map<Integer, List<AbstractPageBlock>> blocksPerPage = pageBlocks.stream().collect(groupingBy(AbstractPageBlock::getPage));
-        Section section = Section.builder().documentTree(context.getDocumentTree()).build();
-
-        context.getSections().add(section);
-        blocksPerPage.keySet().forEach(pageNumber -> addSectionNodeToPageNode(context, section, pageNumber));
-
-        section.setTreeId(getTreeId(parentNode, context, section));
-
-        addFirstHeadlineDirectlyToSection(pageBlocks, context, section);
-        if (containsTablesAndTextBlocks(pageBlocks)) {
-            splitPageBlocksIntoSubSections(pageBlocks).forEach(subSectionPageBlocks -> addSection(section, subSectionPageBlocks, emptyList(), context));
-        } else {
-            addTablesAndParagraphsAndHeadlinesToSection(pageBlocks, context, section);
-        }
-
-        images.stream().distinct().forEach(image -> DocumentGraphFactory.addImage(section, image, context));
-    }
-
-
-    private List<Integer> getTreeId(GenericSemanticNode parentNode, DocumentGraphFactory.Context context, Section section) {
-
-        if (parentNode == null) {
-            return context.getDocumentTree().createNewMainEntryAndReturnId(section);
-        } else {
-            return context.getDocumentTree().createNewChildEntryAndReturnId(parentNode, section);
-        }
-    }
-
-
-    private void addFirstHeadlineDirectlyToSection(List<AbstractPageBlock> pageBlocks, DocumentGraphFactory.Context context, Section section) {
-
-        if (pageBlocks.get(0).isHeadline()) {
-            addTablesAndParagraphsAndHeadlinesToSection(List.of(pageBlocks.get(0)), context, section);
-            pageBlocks.remove(0);
-        }
-    }
-
-
-    private void addTablesAndParagraphsAndHeadlinesToSection(List<AbstractPageBlock> pageBlocks, DocumentGraphFactory.Context context, Section section) {
-
-        Set<AbstractPageBlock> alreadyMerged = new HashSet<>();
-        List<AbstractPageBlock> remainingBlocks = new LinkedList<>(pageBlocks);
-        for (AbstractPageBlock abstractPageBlock : pageBlocks) {
-
-            if (alreadyMerged.contains(abstractPageBlock)) {
-                continue;
-            }
-
-            remainingBlocks.removeAll(alreadyMerged);
-
-            if (abstractPageBlock instanceof TextPageBlock) {
-//                List<TextPageBlock> textBlocksToMerge = findTextBlocksWithSameClassificationAndAlignsYAndSameOrientationUntilConvergence((TextPageBlock) abstractPageBlock, remainingBlocks);
-                List<TextPageBlock> textBlocksToMerge = findTextBlocksWithSameClassificationAndAlignsYAndSameOrientation(List.of((TextPageBlock) abstractPageBlock),
-                        remainingBlocks);
-                alreadyMerged.addAll(textBlocksToMerge);
-                DocumentGraphFactory.addParagraphOrHeadline(section, (TextPageBlock) abstractPageBlock, context, textBlocksToMerge);
-            } else if (abstractPageBlock instanceof TablePageBlock tablePageBlock) {
-                List<TablePageBlock> tablesToMerge = TableMergingUtility.findConsecutiveTablesWithSameColCountAndSameHeaders(tablePageBlock, remainingBlocks);
-                alreadyMerged.addAll(tablesToMerge);
-                TableNodeFactory.addTable(section, tablesToMerge, context);
-            } else {
-                throw new RuntimeException(format("Unhandled AbstractPageBlockType %s!", abstractPageBlock.getClass()));
-            }
-        }
-    }
-
-
-    private boolean containsTablesAndTextBlocks(List<AbstractPageBlock> pageBlocks) {
-
-        return pageBlocks.stream().anyMatch(pageBlock -> pageBlock instanceof TablePageBlock) && pageBlocks.stream().anyMatch(pageBlock -> pageBlock instanceof TextPageBlock);
-    }
-
-
-    /**
-     * This function splits the list of PageBlocks around TablePageBlocks, such that SubSections can be created, that don't include tables.
-     * This is needed so we can execute rules on sections, that do not contain tables.
-     * See: <a href="https://knecon.atlassian.net/wiki/spaces/RED/pages/14765218/Document+Structure">document structure wiki</a>
-     *
-     * @param pageBlocks a List of AbstractPageBlocks, which have at least one TablePageBlock and one ClassificationTextBlock
-     * @return List of Lists of AbstractPageBlocks, which include either a single Headline ClassificationTextBlock and a TablePageBlock or only ClassificationTextBlocks.
-     */
-    private List<List<AbstractPageBlock>> splitPageBlocksIntoSubSections(List<AbstractPageBlock> pageBlocks) {
-
-        List<List<AbstractPageBlock>> splitList = splitIntoCoherentList(pageBlocks);
-        movePrecedingHeadlineToTableList(splitList);
-        return splitList.stream().filter(list -> !list.isEmpty()).toList();
-    }
-
-
-    private void movePrecedingHeadlineToTableList(List<List<AbstractPageBlock>> splitList) {
-
-        for (int i = 0; i < splitList.size(); i++) {
-            if (listIsTablesOnly(splitList.get(i)) && i > 0) {
-                List<AbstractPageBlock> previousList = splitList.get(i - 1);
-                AbstractPageBlock lastPageBlockInPreviousList = previousList.get(previousList.size() - 1);
-                if (lastPageBlockInPreviousList.isHeadline()) {
-                    previousList.remove(previousList.size() - 1);
-                    splitList.get(i).add(0, lastPageBlockInPreviousList);
-                }
-            }
-        }
-    }
-
-
-    private boolean listIsTablesOnly(List<AbstractPageBlock> abstractPageBlocks) {
-
-        return abstractPageBlocks.stream().allMatch(abstractPageBlock -> abstractPageBlock instanceof TablePageBlock);
-    }
-
-
-    /**
-     * @param pageBlocks a List of AbstractPageBlocks, which have at least one TablePageBlock and one ClassificationTextBlock
-     * @return List of Lists of AbstractPageBlocks, which are exclusively of type ClassificationTextBlock or TablePageBlock
-     */
-    private List<List<AbstractPageBlock>> splitIntoCoherentList(List<AbstractPageBlock> pageBlocks) {
-
-        List<List<AbstractPageBlock>> splitList = new LinkedList<>();
-        List<AbstractPageBlock> currentList = new LinkedList<>();
-        splitList.add(currentList);
-
-        Class<? extends AbstractPageBlock> lastPageBlockClass = pageBlocks.get(0).getClass();
-        for (AbstractPageBlock pageBlock : pageBlocks) {
-            if (lastPageBlockClass.isInstance(pageBlock)) {
-                currentList.add(pageBlock);
-            } else {
-                currentList = new LinkedList<>();
-                currentList.add(pageBlock);
-                splitList.add(currentList);
-                lastPageBlockClass = pageBlock.getClass();
-            }
-        }
-        return splitList;
-    }
-
-
-    @SuppressWarnings("PMD")
-    // experimental feature to be used later
-    private List<TextPageBlock> findTextBlocksWithSameClassificationAndAlignsYAndSameOrientationUntilConvergence(TextPageBlock originalTextBlocks,
-                                                                                                                 List<AbstractPageBlock> pageBlocks) {
-
-        int previousCount = 1;
-        List<TextPageBlock> alignedBlocks = findTextBlocksWithSameClassificationAndAlignsYAndSameOrientation(List.of(originalTextBlocks), pageBlocks);
-        while (previousCount < alignedBlocks.size()) {
-            alignedBlocks = findTextBlocksWithSameClassificationAndAlignsYAndSameOrientation(alignedBlocks, pageBlocks);
-            previousCount = alignedBlocks.size();
-        }
-        return alignedBlocks;
-    }
-
-
-    private static List<TextPageBlock> findTextBlocksWithSameClassificationAndAlignsYAndSameOrientation(List<TextPageBlock> textBlocksToMerge, List<AbstractPageBlock> pageBlocks) {
-
-        return Stream.concat(pageBlocks.stream()
-                                .filter(abstractPageBlock -> !textBlocksToMerge.contains(abstractPageBlock))
-                                .filter(abstractPageBlock -> textBlocksToMerge.stream().allMatch(textBlockToMerge -> abstractPageBlock.getPage() == textBlockToMerge.getPage()))
-                                .filter(abstractPageBlock -> textBlocksToMerge.stream().allMatch(textBlockToMerge -> abstractPageBlock.getOrientation().equals(textBlockToMerge.getOrientation())))
-                                .filter(abstractPageBlock -> textBlocksToMerge.stream().anyMatch(abstractPageBlock::intersectsY))
-                                //.filter(abstractPageBlock -> textBlocksToMerge.stream().anyMatch(abstractPageBlock::intersectsX))
-                                .filter(abstractPageBlock -> abstractPageBlock instanceof TextPageBlock)
-                                .map(abstractPageBlock -> (TextPageBlock) abstractPageBlock), //
-                        textBlocksToMerge.stream())//
-                .toList();
-    }
-
-
-    private void addSectionNodeToPageNode(DocumentGraphFactory.Context context, Section section, Integer pageNumber) {
-
-        Page page = context.getPage(pageNumber);
-        page.getMainBody().add(section);
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/factory/TableNodeFactory.java

@@ -1,141 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.factory;
-
-import static java.util.Collections.emptyList;
-
-import java.util.Collection;
-import java.util.List;
-import java.util.Set;
-import java.util.stream.Collectors;
-
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.GenericSemanticNode;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Cell;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Table;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.TableCell;
-import com.iqser.red.service.redaction.v1.server.document.graph.textblock.TextBlock;
-import com.iqser.red.service.redaction.v1.server.document.utils.TextPositionOperations;
-
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public class TableNodeFactory {
-
-    public final double TABLE_CELL_MERGE_CONTENTS_SIZE_THRESHOLD = 0.05;
-
-
-    public void addTable(GenericSemanticNode parentNode, List<TablePageBlock> tablesToMerge, DocumentGraphFactory.Context context) {
-
-        setPageNumberInCells(tablesToMerge);
-        Set<Page> pages = tablesToMerge.stream().map(AbstractPageBlock::getPage).map(context::getPage).collect(Collectors.toSet());
-        List<List<Cell>> mergedRows = tablesToMerge.stream().map(TablePageBlock::getRows).flatMap(Collection::stream).toList();
-
-        Table table = Table.builder()
-                .documentTree(context.getDocumentTree())
-                .numberOfCols(mergedRows.isEmpty() ? 0 : mergedRows.get(0).size())
-                .numberOfRows(mergedRows.size())
-                .build();
-
-        pages.forEach(page -> addTableToPage(page, parentNode, table));
-
-        List<Integer> treeId = context.getDocumentTree().createNewChildEntryAndReturnId(parentNode, table);
-        table.setTreeId(treeId);
-        addTableCells(mergedRows, table, context);
-
-        ifTableHasNoHeadersSetFirstRowAsHeaders(table);
-    }
-
-
-    private void setPageNumberInCells(List<TablePageBlock> tablesToMerge) {
-
-        // For some reason I can't figure out, in some table cells, the ClassificationTextBlocks have 0 as page number
-        // So I am fixing this here, but this should actually be fixed upstream.
-        tablesToMerge.forEach(table -> table.getRows()
-                .stream()
-                .flatMap(Collection::stream)
-                .peek(cell -> cell.setPageNumber(table.getPage()))
-                .forEach(cell -> setPageNumberInTextBlocksWithPageNumberSetTo0(table, cell)));
-    }
-
-
-    private void setPageNumberInTextBlocksWithPageNumberSetTo0(TablePageBlock table, Cell cell) {
-
-        cell.getTextBlocks().stream()//
-                .filter(tb -> tb.getPage() == 0)//
-                .forEach(tb -> tb.setPage(table.getPage()));
-    }
-
-
-    @SuppressWarnings("PMD.UnusedPrivateMethod") // PMD actually flags this wrong
-    private void addTableToPage(Page page, SemanticNode parentNode, Table table) {
-
-        if (!page.getMainBody().contains(parentNode)) {
-            parentNode.getPages().add(page);
-        }
-
-        page.getMainBody().add(table);
-    }
-
-
-    private void ifTableHasNoHeadersSetFirstRowAsHeaders(Table table) {
-
-        if (table.streamHeaders().findAny().isEmpty()) {
-            table.streamRow(0).forEach(tableCellNode -> tableCellNode.setHeader(true));
-        }
-    }
-
-
-    private void addTableCells(List<List<Cell>> rows, Table table, DocumentGraphFactory.Context context) {
-
-        for (int rowIndex = 0; rowIndex < rows.size(); rowIndex++) {
-            for (int colIndex = 0; colIndex < rows.get(rowIndex).size(); colIndex++) {
-                addTableCell(rows.get(rowIndex).get(colIndex), rowIndex, colIndex, table, context);
-            }
-        }
-    }
-
-
-    @SuppressWarnings("PMD.UnusedPrivateMethod") // PMD actually flags this wrong
-    private void addTableCell(Cell cell, int rowIndex, int colIndex, Table tableNode, DocumentGraphFactory.Context context) {
-
-        Page page = context.getPage(cell.getPageNumber());
-
-        TableCell tableCell = TableCell.builder().documentTree(context.getDocumentTree()).row(rowIndex).col(colIndex).header(cell.isHeaderCell()).bBox(cell.getBounds2D()).build();
-        page.getMainBody().add(tableCell);
-
-        List<Integer> treeId = context.getDocumentTree().createNewTableChildEntryAndReturnId(tableNode, tableCell);
-        tableCell.setTreeId(treeId);
-
-        TextBlock textBlock;
-        if (cell.getTextBlocks().isEmpty()) {
-            tableCell.setLeafTextBlock(context.getTextBlockFactory().emptyTextBlock(tableNode, context, page));
-        } else if (cell.getTextBlocks().size() == 1) {
-            textBlock = context.getTextBlockFactory().fromContext(cell.getTextBlocks().get(0).getSequences(), tableCell, context, page);
-            tableCell.setLeafTextBlock(textBlock);
-        } else if (firstTextBlockIsHeadline(cell)) {
-            SectionNodeFactory.addSection(tableCell, cell.getTextBlocks().stream().map(tb -> (AbstractPageBlock) tb).toList(), emptyList(), context);
-        } else if (cellAreaIsSmallerThanPageAreaTimesThreshold(cell, page)) {
-            List<TextPositionSequence> sequences = TextPositionOperations.mergeAndSortTextPositionSequenceByYThenX(cell.getTextBlocks());
-            textBlock = context.getTextBlockFactory().fromContext(sequences, tableCell, context, page);
-            tableCell.setLeafTextBlock(textBlock);
-        } else {
-            cell.getTextBlocks().forEach(tb -> DocumentGraphFactory.addParagraphOrHeadline(tableCell, tb, context, List.of(tb)));
-        }
-    }
-
-
-    private boolean cellAreaIsSmallerThanPageAreaTimesThreshold(Cell cell, Page page) {
-
-        return cell.getArea() < TABLE_CELL_MERGE_CONTENTS_SIZE_THRESHOLD * page.getHeight() * page.getWidth();
-    }
-
-
-    private boolean firstTextBlockIsHeadline(Cell cell) {
-
-        return cell.getTextBlocks().get(0).isHeadline();
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/factory/TextBlockFactory.java

@@ -1,53 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.factory;
-
-import java.util.List;
-
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.document.graph.textblock.AtomicTextBlock;
-
-import lombok.AccessLevel;
-import lombok.experimental.FieldDefaults;
-
-@FieldDefaults(level = AccessLevel.PRIVATE)
-public class TextBlockFactory {
-
-    int stringOffset;
-    long textBlockIdx;
-
-
-    public AtomicTextBlock fromContext(List<TextPositionSequence> sequences, SemanticNode parent, DocumentGraphFactory.Context context, Page page) {
-
-        Integer numberOnPage = context.getAndIncrementTextBlockNumberOnPage(page);
-        return fromNumberOnPage(sequences, parent, numberOnPage, page);
-    }
-
-
-    public AtomicTextBlock fromNumberOnPage(List<TextPositionSequence> sequences, SemanticNode parent, Integer numberOnPage, Page page) {
-
-        SearchTextWithTextPositionDto searchTextWithTextPositionDto = SearchTextWithTextPositionFactory.buildSearchTextToTextPositionModel(sequences);
-        int offset = stringOffset;
-        stringOffset += searchTextWithTextPositionDto.getSearchText().length();
-        long idx = textBlockIdx;
-        textBlockIdx++;
-        return AtomicTextBlock.fromSearchTextWithTextPositionDto(searchTextWithTextPositionDto, parent, offset, idx, numberOnPage, page);
-    }
-
-
-    public AtomicTextBlock emptyTextBlock(SemanticNode parent, DocumentGraphFactory.Context context, Page page) {
-
-        long idx = textBlockIdx;
-        textBlockIdx++;
-        return AtomicTextBlock.empty(idx, stringOffset, page, context.getAndIncrementTextBlockNumberOnPage(page), parent);
-    }
-
-
-    public AtomicTextBlock emptyTextBlockFromInteger(SemanticNode parent, Integer numberOnPage, Page page) {
-
-        long idx = textBlockIdx;
-        textBlockIdx++;
-        return AtomicTextBlock.empty(idx, stringOffset, page, numberOnPage, parent);
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/graph/nodes/Page.java

@@ -6,7 +6,6 @@ import java.util.List;
 import java.util.Set;
 
 import com.iqser.red.service.redaction.v1.server.document.graph.entity.RedactionEntity;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
 import com.iqser.red.service.redaction.v1.server.document.graph.textblock.TextBlock;
 import com.iqser.red.service.redaction.v1.server.document.graph.textblock.TextBlockCollector;
 
@@ -48,18 +47,6 @@ public class Page {
     Set<Image> images = new HashSet<>();
 
 
-    public static Page fromClassificationPage(ClassificationPage classificationPage) {
-
-        return Page.builder()
-                .height((int) classificationPage.getPageHeight())
-                .width((int) classificationPage.getPageWidth())
-                .number(classificationPage.getPageNumber())
-                .rotation(classificationPage.getRotation())
-                .mainBody(new LinkedList<>())
-                .build();
-    }
-
-
     public TextBlock getMainBodyTextBlock() {
 
         return mainBody.stream().filter(SemanticNode::isLeaf).map(SemanticNode::getLeafTextBlock).collect(new TextBlockCollector());

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/graph/textblock/AtomicTextBlock.java

@@ -12,13 +12,12 @@ import java.util.Map;
 import java.util.Set;
 import java.util.stream.Collectors;
 
-import com.iqser.red.service.redaction.v1.server.document.factory.SearchTextWithTextPositionDto;
 import com.iqser.red.service.redaction.v1.server.document.graph.Boundary;
 import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Page;
 import com.iqser.red.service.redaction.v1.server.document.graph.nodes.SemanticNode;
 import com.iqser.red.service.redaction.v1.server.document.utils.RectangleTransformations;
-import com.iqser.red.service.redaction.v1.server.document.data.AtomicPositionBlockData;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentPositionData;
-import com.iqser.red.service.redaction.v1.server.document.data.AtomicTextBlockData;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentTextData;
 
 import lombok.AccessLevel;
 import lombok.AllArgsConstructor;
@@ -58,28 +57,6 @@ public class AtomicTextBlock implements TextBlock {
         return lineBreaks.size() + 1;
     }
 
-
-    public static AtomicTextBlock fromSearchTextWithTextPositionDto(SearchTextWithTextPositionDto searchTextWithTextPositionDto,
-                                                                    SemanticNode parent,
-                                                                    int stringOffset,
-                                                                    Long textBlockIdx,
-                                                                    Integer numberOnPage,
-                                                                    Page page) {
-
-        return AtomicTextBlock.builder()
-                .id(textBlockIdx)
-                .parent(parent)
-                .searchText(searchTextWithTextPositionDto.getSearchText())
-                .numberOnPage(numberOnPage)
-                .page(page)
-                .lineBreaks(searchTextWithTextPositionDto.getLineBreaks())
-                .positions(searchTextWithTextPositionDto.getPositions())
-                .stringIdxToPositionIdx(searchTextWithTextPositionDto.getStringCoordsToPositionCoords())
-                .boundary(new Boundary(stringOffset, stringOffset + searchTextWithTextPositionDto.getSearchText().length()))
-                .build();
-    }
-
-
     public static AtomicTextBlock empty(Long textBlockIdx, int stringOffset, Page page, int numberOnPage, SemanticNode parent) {
 
         return AtomicTextBlock.builder()
@@ -96,8 +73,8 @@ public class AtomicTextBlock implements TextBlock {
     }
 
 
-    public static AtomicTextBlock fromAtomicTextBlockData(AtomicTextBlockData atomicTextBlockData,
+    public static AtomicTextBlock fromAtomicTextBlockData(DocumentTextData atomicTextBlockData,
-                                                          AtomicPositionBlockData atomicPositionBlockData,
+                                                          DocumentPositionData atomicPositionBlockData,
                                                           SemanticNode parent,
                                                           Page page) {
 

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/utils/RectangleTransformations.java

@@ -12,11 +12,9 @@ import java.util.function.Function;
 import java.util.function.Supplier;
 import java.util.stream.Collector;
 
-import org.apache.pdfbox.pdmodel.common.PDRectangle;
 
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Point;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
 import com.iqser.red.service.redaction.v1.server.document.graph.textblock.AtomicTextBlock;
 
 import lombok.AllArgsConstructor;
@@ -24,27 +22,6 @@ import lombok.NoArgsConstructor;
 
 public class RectangleTransformations {
 
-    public static PDRectangle toPDRectangleBBox(List<Rectangle> rectangles) {
-
-        Rectangle2D rectangle2D = RectangleTransformations.rectangleBBox(rectangles);
-
-        PDRectangle annotationPosition = new PDRectangle();
-        annotationPosition.setLowerLeftX((float) rectangle2D.getMinX());
-        annotationPosition.setLowerLeftY((float) rectangle2D.getMinY());
-        annotationPosition.setUpperRightX((float) rectangle2D.getMaxX());
-        annotationPosition.setUpperRightY((float) rectangle2D.getMaxY());
-        return annotationPosition;
-    }
-
-    public static Rectangle2D abstractPageBlockBBox(List<AbstractPageBlock> abstractPageBlocks) {
-
-        return abstractPageBlocks.stream()
-                .map(abstractPageBlock -> new Rectangle2D.Double(abstractPageBlock.getMinX(),
-                        abstractPageBlock.getMinY(),
-                        abstractPageBlock.getWidth(),
-                        abstractPageBlock.getHeight())).collect(new Rectangle2DBBoxCollector());
-    }
-
 
     public static Rectangle2D atomicTextBlockBBox(List<AtomicTextBlock> atomicTextBlocks) {
 
@@ -114,12 +91,6 @@ public class RectangleTransformations {
     }
 
 
-    public static Rectangle2D toRectangle2D(PDRectangle cropBox) {
-
-        return new Rectangle2D.Double(cropBox.getLowerLeftX(), cropBox.getLowerLeftY(), cropBox.getWidth(), cropBox.getHeight());
-    }
-
-
     private static class Rectangle2DBBoxCollector implements Collector<Rectangle2D, Rectangle2DBBoxCollector.BBox, Rectangle2D> {
 
         @Override

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/utils/TableMergingUtility.java

@@ -1,52 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.utils;
-
-import java.util.Collection;
-import java.util.LinkedList;
-import java.util.List;
-import java.util.stream.Stream;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Cell;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
-
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public class TableMergingUtility {
-
-    private static final double TABLE_ALIGNMENT_THRESHOLD = 2d;
-
-
-    public List<TablePageBlock> findConsecutiveTablesWithSameColCountAndSameHeaders(TablePageBlock originalTablePageBlock, List<AbstractPageBlock> pageBlocks) {
-
-        List<TablePageBlock> consecutiveTables = pageBlocks.stream()
-                .map(abstractPageBlock -> (TablePageBlock) abstractPageBlock)
-                .filter(tablePageBlock -> !tablePageBlock.equals(originalTablePageBlock))
-                .toList();
-        assert consecutiveTables.size() == pageBlocks.size() - 1;
-
-        List<TablePageBlock> consecutiveTablesWithSameColCountAndHeaders = new LinkedList<>();
-        for (TablePageBlock consecutiveTable : consecutiveTables) {
-            if (consecutiveTable.getColCount() == originalTablePageBlock.getColCount() && !hasTableHeader(consecutiveTable) && outerBoundaryAlignsX(originalTablePageBlock,
-                    consecutiveTable)) {
-                consecutiveTablesWithSameColCountAndHeaders.add(consecutiveTable);
-            } else {
-                break;
-            }
-        }
-        return Stream.concat(Stream.of(originalTablePageBlock), consecutiveTablesWithSameColCountAndHeaders.stream()).toList();
-    }
-
-
-    private static boolean outerBoundaryAlignsX(TablePageBlock originalTablePageBlock, TablePageBlock consecutiveTable) {
-
-        return Math.abs(consecutiveTable.getMinX() - originalTablePageBlock.getMinX()) < TABLE_ALIGNMENT_THRESHOLD && Math.abs(consecutiveTable.getMaxX() - originalTablePageBlock.getMaxX()) < TABLE_ALIGNMENT_THRESHOLD;
-    }
-
-
-    private boolean hasTableHeader(TablePageBlock table) {
-
-        return table.getRows().stream().flatMap(Collection::stream).anyMatch(Cell::isHeaderCell);
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/utils/TextPositionOperations.java

@@ -1,27 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.document.utils;
-
-import java.util.List;
-import java.util.stream.Collectors;
-
-import org.apache.pdfbox.util.QuickSort;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-
-public class TextPositionOperations {
-
-    private static final TextPositionSequenceComparator comparator = new TextPositionSequenceComparator();
-
-
-    public static List<TextPositionSequence> mergeAndSortTextPositionSequenceByYThenX(List<TextPageBlock> textBlocks) {
-
-        var sequence = textBlocks.stream().flatMap(tb -> tb.getSequences().stream()).collect(Collectors.toList());
-
-        // because the TextPositionSequenceComparator is not transitive, but
-        // JDK7+ enforces transitivity on comparators, we need to use
-        // a custom quicksort implementation (which is slower, unfortunately).
-        QuickSort.sort(sequence, comparator);
-        return sequence;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/document/utils/TextPositionSequenceComparator.java

@@ -1,74 +0,0 @@
-/*
- * Licensed to the Apache Software Foundation (ASF) under one or more
- * contributor license agreements.  See the NOTICE file distributed with
- * this work for additional information regarding copyright ownership.
- * The ASF licenses this file to You under the Apache License, Version 2.0
- * (the "License"); you may not use this file except in compliance with
- * the License.  You may obtain a copy of the License at
- *
- *      http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-package com.iqser.red.service.redaction.v1.server.document.utils;
-
-import java.util.Comparator;
-
-import org.apache.pdfbox.text.TextPosition;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-
-/**
- * This class is a comparator for TextPosition operators.  It handles
- * pages with text in different directions by grouping the text based
- * on direction and sorting in that direction. This allows continuous text
- * in a given direction to be more easily grouped together.  
- *
- * @author Ben Litchfield
- */
-public class TextPositionSequenceComparator implements Comparator<TextPositionSequence>
-{
-    @Override
-    public int compare(TextPositionSequence pos1, TextPositionSequence pos2)
-    {
-        // only compare text that is in the same direction
-        int cmp1 = Float.compare(pos1.getDir().getDegrees(), pos2.getDir().getDegrees());
-        if (cmp1 != 0)
-        {
-            return cmp1;
-        }
-        
-        // get the text direction adjusted coordinates
-        float x1 = pos1.getMinXDirAdj();
-        float x2 = pos2.getMinXDirAdj();
-        
-        float pos1YBottom = pos1.getMaxYDirAdj();
-        float pos2YBottom = pos2.getMaxYDirAdj();
-
-        // note that the coordinates have been adjusted so 0,0 is in upper left
-        float pos1YTop = pos1YBottom - pos1.getTextHeight();
-        float pos2YTop = pos2YBottom - pos2.getTextHeight();
-
-        float yDifference = Math.abs(pos1YBottom - pos2YBottom);
-
-        // we will do a simple tolerance comparison
-        if (yDifference < .1 ||
-            pos2YBottom >= pos1YTop && pos2YBottom <= pos1YBottom ||
-            pos1YBottom >= pos2YTop && pos1YBottom <= pos2YBottom)
-        {
-            return Float.compare(x1, x2);
-        }
-        else if (pos1YBottom < pos2YBottom)
-        {
-            return -1;
-        }
-        else
-        {
-            return 1;
-        }
-    }
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/ImageServiceResponseAdapter.java

@@ -1,80 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter;
-
-import java.awt.geom.Rectangle2D;
-import java.util.ArrayList;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Locale;
-import java.util.Map;
-
-import org.springframework.stereotype.Service;
-
-import com.fasterxml.jackson.databind.ObjectMapper;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.dossier.file.FileType;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image.ImageServiceResponse;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.ImageType;
-import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
-
-import lombok.RequiredArgsConstructor;
-import lombok.SneakyThrows;
-
-@Service
-@RequiredArgsConstructor
-public class ImageServiceResponseAdapter {
-
-    private final ObjectMapper objectMapper;
-    private final RedactionStorageService redactionStorageService;
-
-
-    @SneakyThrows
-    public Map<Integer, List<ClassifiedImage>> convertImages(String dossierId, String fileId) {
-
-        var imageClassificationStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(dossierId, fileId, FileType.IMAGE_INFO));
-
-        ImageServiceResponse imageServiceResponse = objectMapper.readValue(imageClassificationStream, ImageServiceResponse.class);
-
-        Map<Integer, List<ClassifiedImage>> images = new HashMap<>();
-        imageServiceResponse.getData().forEach(imageMetadata -> {
-            var classification = imageMetadata.getFilters().isAllPassed() ? ImageType.valueOf(imageMetadata.getClassification()
-                    .getLabel()
-                    .toUpperCase(Locale.ROOT)) : ImageType.OTHER;
-            images.computeIfAbsent(imageMetadata.getPosition().getPageNumber(), x -> new ArrayList<>())
-                    .add(new ClassifiedImage(new Rectangle2D.Double(imageMetadata.getPosition().getX1(),
-                            imageMetadata.getPosition().getY1(),
-                            imageMetadata.getGeometry().getWidth(),
-                            imageMetadata.getGeometry().getHeight()), classification, imageMetadata.isAlpha(), imageMetadata.getPosition().getPageNumber()));
-        });
-
-        // Currently This is a copy but, it will be changed later because i don' t think that we should unclassified images.
-        imageServiceResponse.getDataCV().forEach(imageMetadata -> {
-            var classification = imageMetadata.getFilters().isAllPassed() ? ImageType.valueOf(imageMetadata.getClassification()
-                    .getLabel()
-                    .toUpperCase(Locale.ROOT)) : ImageType.OTHER;
-            images.computeIfAbsent(imageMetadata.getPosition().getPageNumber(), x -> new ArrayList<>())
-                    .add(new ClassifiedImage(new Rectangle2D.Double(imageMetadata.getPosition().getX1(),
-                            imageMetadata.getPosition().getY1(),
-                            imageMetadata.getGeometry().getWidth(),
-                            imageMetadata.getGeometry().getHeight()), classification, imageMetadata.isAlpha(), imageMetadata.getPosition().getPageNumber()));
-        });
-
-        return images;
-    }
-
-
-    public void findOcr(ClassificationPage page) {
-
-        page.getImages().forEach(image -> {
-            if (image.getImageType().equals(ImageType.OTHER)) {
-                page.getTextBlocks().forEach(textblock -> {
-                    if (image.getPosition().contains(textblock.getMinX(), textblock.getMinY(), textblock.getWidth(), textblock.getHeight())) {
-                        image.setImageType(ImageType.OCR);
-                    }
-                });
-            }
-        });
-    }
-
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/TableServiceResponseAdapter.java

@@ -1,63 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter;
-
-import java.util.ArrayList;
-import java.util.Collection;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-
-import org.springframework.stereotype.Service;
-
-import com.fasterxml.jackson.databind.ObjectMapper;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.dossier.file.FileType;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.PdfTableCell;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.TableCells;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.TableServiceResponse;
-import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
-
-import lombok.RequiredArgsConstructor;
-import lombok.SneakyThrows;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@Service
-@RequiredArgsConstructor
-public class TableServiceResponseAdapter {
-
-    private final ObjectMapper objectMapper;
-    private final RedactionStorageService redactionStorageService;
-
-
-    @SneakyThrows
-    public Map<Integer, List<PdfTableCell>> convertTables(String dossierId, String fileId) {
-
-        var tableClassificationStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(dossierId, fileId, FileType.TABLES));
-
-        TableServiceResponse tableServiceResponse = objectMapper.readValue(tableClassificationStream, TableServiceResponse.class);
-
-        Map<Integer, List<PdfTableCell>> tableCells = new HashMap<>();
-        tableServiceResponse.getData()
-                .forEach(tableData -> tableCells.computeIfAbsent(tableData.getPageInfo().getNumber(), tableCell -> new ArrayList<>())
-                        .addAll(convertTableCells(tableData.getTableCells())));
-
-        return tableCells;
-    }
-
-
-    private Collection<? extends PdfTableCell> convertTableCells(List<TableCells> tableCells) {
-
-        List<PdfTableCell> pdfTableCells = new ArrayList<>();
-
-        tableCells.forEach(t -> pdfTableCells.add(PdfTableCell.builder()
-                .y0(t.getY0())
-                .x1(t.getX1())
-                .y1(t.getY1())
-                .x0(t.getX0())
-                .width(t.getWidth())
-                .height(t.getHeight())
-                .build()));
-
-        return pdfTableCells;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/Classification.java

@@ -1,14 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import java.util.HashMap;
-import java.util.Map;
-
-import lombok.Data;
-
-@Data
-public class Classification {
-
-    private Map<String, Float> probabilities = new HashMap<>();
-    private String label;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/FilterGeometry.java

@@ -1,11 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import lombok.Data;
-
-@Data
-public class FilterGeometry {
-
-    private ImageSize imageSize;
-    private ImageFormat imageFormat;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/Filters.java

@@ -1,12 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import lombok.Data;
-
-@Data
-public class Filters {
-
-    private FilterGeometry geometry;
-    private Probability probability;
-    private boolean allPassed;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/Geometry.java

@@ -1,11 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import lombok.Data;
-
-@Data
-public class Geometry {
-
-    private float width;
-    private float height;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/ImageFormat.java

@@ -1,12 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import lombok.Data;
-
-@Data
-public class ImageFormat {
-
-    private float quotient;
-    private boolean tooTall;
-    private boolean tooWide;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/ImageMetadata.java

@@ -1,14 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import lombok.Data;
-
-@Data
-public class ImageMetadata {
-
-    private Classification classification;
-    private Position position;
-    private Geometry geometry;
-    private Filters filters;
-    private boolean alpha;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/ImageServiceResponse.java

@@ -1,46 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import com.fasterxml.jackson.annotation.JsonAlias;
-import com.fasterxml.jackson.annotation.JsonProperty;
-
-import lombok.Data;
-
-@Data
-public class ImageServiceResponse {
-
-    private String dossierId;
-    private String fileId;
-
-    @JsonProperty(value = "imageMetadata")
-    @JsonAlias("data")
-    private List<ImageMetadata> data = new ArrayList<>();
-
-    private List<ImageMetadata> dataCV = new ArrayList<>();
-
-
-    @JsonProperty(value = "imageMetadata")
-    @JsonAlias("data")
-    public void setData(List<ImageMetadata> data) {this.data = data;}
-
-
-    public List<ImageMetadata> getData() {
-
-        if (this.data == null) {
-            this.data = new ArrayList<>();
-        }
-        return data;
-    }
-
-
-    public List<ImageMetadata> getDataCV() {
-
-        if (this.dataCV == null) {
-            this.dataCV = new ArrayList<>();
-        }
-        return dataCV;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/ImageSize.java

@@ -1,12 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import lombok.Data;
-
-@Data
-public class ImageSize {
-
-    private float quotient;
-    private boolean tooLarge;
-    private boolean tooSmall;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/Position.java

@@ -1,14 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import lombok.Data;
-
-@Data
-public class Position {
-
-    private float x1;
-    private float x2;
-    private float y1;
-    private float y2;
-    private int pageNumber;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/image/Probability.java

@@ -1,10 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.image;
-
-import lombok.Data;
-
-@Data
-public class Probability {
-
-    private boolean unconfident;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/table/PageInfo.java

@@ -1,13 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
-
-import lombok.Data;
-
-@Data
-public class PageInfo {
-
-    private int number;
-    private int rotation;
-    private float width;
-    private float height;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/table/PdfTableCell.java

@@ -1,21 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
-
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.RequiredArgsConstructor;
-
-@Data
-@Builder
-@AllArgsConstructor
-@RequiredArgsConstructor
-public class PdfTableCell {
-
-    private float x0;
-    private float y0;
-    private float x1;
-    private float y1;
-    private float width;
-    private float height;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/table/TableCells.java

@@ -1,15 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
-
-import lombok.Data;
-
-@Data
-public class TableCells {
-
-    private float x0;
-    private float y0;
-    private float x1;
-    private float y1;
-    private float width;
-    private float height;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/table/TableData.java

@@ -1,14 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import lombok.Data;
-
-@Data
-public class TableData {
-
-    private PageInfo pageInfo;
-    private List<TableCells> tableCells = new ArrayList<>();
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/adapter/table/TableServiceResponse.java

@@ -1,19 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import lombok.Data;
-
-@Data
-public class TableServiceResponse {
-
-    private String dossierId;
-    private String fileId;
-    private String operation;
-    private String targetFileExtension;
-    private String responseFileExtension;
-
-    private List<TableData> data = new ArrayList<>();
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/AbstractPageBlock.java

@@ -1,88 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-import com.fasterxml.jackson.annotation.JsonIgnore;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-
-import lombok.AllArgsConstructor;
-import lombok.Data;
-import lombok.NoArgsConstructor;
-
-@Data
-@AllArgsConstructor
-@NoArgsConstructor
-public abstract class AbstractPageBlock {
-
-    @JsonIgnore
-    protected float minX;
-    @JsonIgnore
-    protected float maxX;
-    @JsonIgnore
-    protected float minY;
-    @JsonIgnore
-    protected float maxY;
-    @JsonIgnore
-    protected PageBlockType classification;
-    @JsonIgnore
-    protected int page;
-
-    int columnIndex;
-
-    @JsonIgnore
-    private Orientation orientation = Orientation.NONE;
-
-
-    public abstract String getText();
-
-
-    public boolean isHeadline() {
-
-        return this instanceof TextPageBlock && this.getClassification() != null && this.getClassification().isHeadline();
-    }
-
-
-    public boolean containsBlock(TextPageBlock other) {
-
-        return this.minX <= other.getPdfMinX() && this.maxX >= other.getPdfMaxX() && this.minY >= other.getPdfMinY() && this.maxY <= other.getPdfMaxY();
-    }
-
-
-    public boolean contains(AbstractPageBlock other) {
-
-        return this.minX <= other.minX && this.maxX >= other.maxX && this.minY >= other.minY && this.maxY <= other.maxY;
-    }
-
-
-    public boolean contains(Rectangle other) {
-
-        return page == other.getPage() && this.minX <= other.getTopLeft().getX() && this.maxX >= other.getTopLeft().getX() + other.getWidth() && this.minY <= other.getTopLeft()
-                .getY() && this.maxY >= other.getTopLeft().getY() + other.getHeight();
-    }
-
-
-    @JsonIgnore
-    public float getHeight() {
-
-        return maxY - minY;
-    }
-
-
-    @JsonIgnore
-    public float getWidth() {
-
-        return maxX - minX;
-    }
-
-
-    public boolean intersectsY(AbstractPageBlock atc) {
-
-        return this.minY <= atc.getMaxY() && this.maxY >= atc.getMinY();
-    }
-
-
-    public boolean intersectsX(AbstractPageBlock atc) {
-
-        return this.minX <= atc.getMaxX() && this.maxX >= atc.getMinX();
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/ClassificationDocument.java

@@ -1,33 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.section.SectionGrid;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.UnclassifiedText;
-import com.iqser.red.service.redaction.v1.server.redaction.model.dictionary.DictionaryVersion;
-
-import lombok.Data;
-import lombok.NoArgsConstructor;
-
-@Data
-@NoArgsConstructor
-public class ClassificationDocument {
-
-    private List<ClassificationPage> pages = new ArrayList<>();
-    private List<ClassificationSection> sections = new ArrayList<>();
-    private List<ClassificationHeader> headers = new ArrayList<>();
-    private List<ClassificationFooter> footers = new ArrayList<>();
-    private List<UnclassifiedText> unclassifiedTexts = new ArrayList<>();
-    private FloatFrequencyCounter textHeightCounter = new FloatFrequencyCounter();
-    private FloatFrequencyCounter fontSizeCounter = new FloatFrequencyCounter();
-    private StringFrequencyCounter fontCounter = new StringFrequencyCounter();
-    private StringFrequencyCounter fontStyleCounter = new StringFrequencyCounter();
-    private boolean headlines;
-
-    private SectionGrid sectionGrid = new SectionGrid();
-    private DictionaryVersion dictionaryVersion;
-    private long rulesVersion;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/ClassificationFooter.java

@@ -1,16 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-import java.util.List;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-
-import lombok.AllArgsConstructor;
-import lombok.Data;
-
-@Data
-@AllArgsConstructor
-public class ClassificationFooter {
-
-    private List<TextPageBlock> textBlocks;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/ClassificationHeader.java

@@ -1,16 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-import java.util.List;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-
-import lombok.AllArgsConstructor;
-import lombok.Data;
-
-@Data
-@AllArgsConstructor
-public class ClassificationHeader {
-
-    private List<TextPageBlock> textBlocks;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/ClassificationPage.java

@@ -1,38 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
-
-import lombok.Data;
-import lombok.NonNull;
-import lombok.RequiredArgsConstructor;
-
-@Data
-@RequiredArgsConstructor
-public class ClassificationPage {
-
-    @NonNull
-    private List<AbstractPageBlock> textBlocks;
-
-    private List<ClassifiedImage> images = new ArrayList<>();
-
-    private Rectangle bodyTextFrame;
-
-    private boolean landscape;
-    private int rotation;
-
-    private int pageNumber;
-
-    private FloatFrequencyCounter textHeightCounter = new FloatFrequencyCounter();
-    private FloatFrequencyCounter fontSizeCounter = new FloatFrequencyCounter();
-    private StringFrequencyCounter fontCounter = new StringFrequencyCounter();
-    private StringFrequencyCounter fontStyleCounter = new StringFrequencyCounter();
-
-    private float pageWidth;
-    private float pageHeight;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/ClassificationSection.java

@@ -1,32 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
-
-import lombok.Data;
-import lombok.NoArgsConstructor;
-
-@Data
-@NoArgsConstructor
-public class ClassificationSection {
-
-    private List<AbstractPageBlock> pageBlocks = new ArrayList<>();
-    private List<ClassifiedImage> images = new ArrayList<>();
-    private String headline;
-
-
-    public List<TablePageBlock> getTables() {
-
-        List<TablePageBlock> tables = new ArrayList<>();
-        pageBlocks.forEach(block -> {
-            if (block instanceof TablePageBlock) {
-                tables.add((TablePageBlock) block);
-            }
-        });
-        return tables;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/Column.java

@@ -1,14 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-import java.awt.geom.Rectangle2D;
-
-import lombok.AllArgsConstructor;
-
-@AllArgsConstructor
-public class Column {
-
-    int index;
-    ColumnType columnType;
-    Rectangle2D bBox;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/ColumnType.java

@@ -1,6 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-public enum ColumnType {
-    RULING,
-    DISTANCE
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/FloatFrequencyCounter.java

@@ -1,77 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-import java.util.ArrayList;
-import java.util.Collections;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-import java.util.stream.Collectors;
-
-import lombok.Getter;
-
-public class FloatFrequencyCounter {
-
-    @Getter
-    Map<Float, Integer> countPerValue = new HashMap<>();
-
-
-    public void add(float value) {
-
-        if (!countPerValue.containsKey(value)) {
-            countPerValue.put(value, 1);
-        } else {
-            countPerValue.put(value, countPerValue.get(value) + 1);
-        }
-    }
-
-
-    public void addAll(Map<Float, Integer> otherCounter) {
-
-        for (Map.Entry<Float, Integer> entry : otherCounter.entrySet()) {
-            if (countPerValue.containsKey(entry.getKey())) {
-                countPerValue.put(entry.getKey(), countPerValue.get(entry.getKey()) + entry.getValue());
-            } else {
-                countPerValue.put(entry.getKey(), entry.getValue());
-            }
-        }
-    }
-
-
-    public Float getMostPopular() {
-
-        Map.Entry<Float, Integer> mostPopular = null;
-        for (Map.Entry<Float, Integer> entry : countPerValue.entrySet()) {
-            if (mostPopular == null || entry.getValue() >= mostPopular.getValue()) {
-                mostPopular = entry;
-            }
-        }
-        return mostPopular != null ? mostPopular.getKey() : null;
-    }
-
-
-    public List<Float> getHighterThanMostPopular() {
-
-        Float mostPopular = getMostPopular();
-        List<Float> higher = new ArrayList<>();
-        for (Float value : countPerValue.keySet()) {
-            if (value > mostPopular) {
-                higher.add(value);
-            }
-        }
-
-        return higher.stream().sorted(Collections.reverseOrder()).collect(Collectors.toList());
-    }
-
-
-    public Float getHighest() {
-
-        Float highest = null;
-        for (Float value : countPerValue.keySet()) {
-            if (highest == null || value > highest) {
-                highest = value;
-            }
-        }
-        return highest;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/Orientation.java

@@ -1,8 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-public enum Orientation {
-
-    NONE,
-    LEFT,
-    RIGHT
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/PageBlockType.java

@@ -1,38 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model;
-
-public enum PageBlockType {
-    H1,
-    H2,
-    H3,
-    H4,
-    H5,
-    H6,
-    HEADER,
-    FOOTER,
-    TITLE,
-    PARAGRAPH,
-    PARAGRAPH_BOLD,
-    PARAGRAPH_ITALIC,
-    PARAGRAPH_UNKNOWN,
-    OTHER,
-    TABLE;
-
-
-    public static PageBlockType getHeadlineType(int i) {
-
-        return switch (i) {
-            case 1 -> PageBlockType.H1;
-            case 2 -> PageBlockType.H2;
-            case 3 -> PageBlockType.H3;
-            case 4 -> PageBlockType.H4;
-            case 5 -> PageBlockType.H5;
-            default -> PageBlockType.H6;
-        };
-    }
-
-
-    public boolean isHeadline() {
-
-        return this.equals(H1) || this.equals(H2) || this.equals(H3) || this.equals(H4) || this.equals(H5) || this.equals(H6);
-    }
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/image/ClassifiedImage.java

@@ -1,25 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image;
-
-import java.awt.geom.Rectangle2D;
-
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.ImageType;
-
-import lombok.Data;
-import lombok.NonNull;
-import lombok.RequiredArgsConstructor;
-
-@Data
-@RequiredArgsConstructor
-public class ClassifiedImage {
-
-    @NonNull
-    private Rectangle2D position;
-    @NonNull
-    private ImageType imageType;
-    private boolean isAppendedToSection;
-    @NonNull
-    private boolean hasTransparency;
-    @NonNull
-    private int page;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/table/Cell.java

@@ -1,79 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
-
-import java.awt.geom.Point2D;
-import java.util.ArrayList;
-import java.util.Iterator;
-import java.util.List;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
-
-import lombok.Data;
-import lombok.EqualsAndHashCode;
-import lombok.NoArgsConstructor;
-
-@SuppressWarnings("serial")
-@Data
-@EqualsAndHashCode(callSuper = true)
-@NoArgsConstructor
-public class Cell extends Rectangle {
-
-    private List<TextPageBlock> textBlocks = new ArrayList<>();
-
-    private List<Cell> headerCells = new ArrayList<>();
-
-    private boolean isHeaderCell;
-
-    private static final int MIN_SIZE = 1;
-
-    private int pageNumber;
-
-
-    public Cell(Point2D topLeft, Point2D bottomRight) {
-
-        super((float) topLeft.getY(), (float) topLeft.getX(), (float) (bottomRight.getX() - topLeft.getX()), (float) (bottomRight.getY() - topLeft.getY()));
-    }
-
-
-    public void addTextBlock(TextPageBlock textBlock) {
-
-        textBlocks.add(textBlock);
-    }
-
-
-    @Override
-    public String toString() {
-
-        StringBuilder sb = new StringBuilder();
-
-        Iterator<TextPageBlock> itty = textBlocks.iterator();
-        TextPositionSequence previous = null;
-        while (itty.hasNext()) {
-
-            TextPageBlock textBlock = itty.next();
-
-            for (TextPositionSequence word : textBlock.getSequences()) {
-                if (previous != null) {
-                    if (Math.abs(previous.getMaxYDirAdj() - word.getMaxYDirAdj()) > word.getTextHeight()) {
-                        sb.append('\n');
-                    } else {
-                        sb.append(' ');
-                    }
-                }
-                sb.append(word.toString());
-                previous = word;
-            }
-
-        }
-
-        return TextNormalizationUtilities.removeHyphenLineBreaks(sb.toString()).replaceAll("\n", " ").replaceAll(" {2}", " ");
-    }
-
-
-    public boolean hasMinimumSize() {
-
-        return this.getHeight() >= MIN_SIZE && this.getWidth() >= MIN_SIZE;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/table/CellPosition.java

@@ -1,22 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
-
-import lombok.RequiredArgsConstructor;
-import lombok.Value;
-
-@Value
-@RequiredArgsConstructor
-public class CellPosition implements Comparable<CellPosition> {
-
-    int row;
-
-    int col;
-
-
-    @Override
-    public int compareTo(CellPosition other) {
-
-        int rowDiff = row - other.row;
-        return rowDiff != 0 ? rowDiff : col - other.col;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/table/CleanRulings.java

@@ -1,15 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
-
-import java.util.List;
-
-import lombok.Builder;
-import lombok.Data;
-
-@Data
-@Builder
-public class CleanRulings {
-
-    List<Ruling> horizontal;
-    List<Ruling> vertical;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/table/Rectangle.java

@@ -1,218 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
-
-import java.awt.geom.Point2D;
-import java.awt.geom.Rectangle2D;
-import java.util.Comparator;
-import java.util.List;
-
-@SuppressWarnings("all")
-public class Rectangle extends Rectangle2D.Float {
-
-    protected static final float VERTICAL_COMPARISON_THRESHOLD = 0.4f;
-    /**
-     * Ill-defined comparator, from when Rectangle was Comparable.
-     * <p>
-     * see https://github.com/tabulapdf/tabula-java/issues/116
-     *
-     * @deprecated with no replacement
-     */
-    @Deprecated
-    public static final Comparator<Rectangle> ILL_DEFINED_ORDER = new Comparator<Rectangle>() {
-        @Override
-        public int compare(Rectangle o1, Rectangle o2) {
-
-            if (o1.equals(o2)) {
-                return 0;
-            }
-            if (o1.verticalOverlap(o2) > VERTICAL_COMPARISON_THRESHOLD) {
-                return o1.isLtrDominant() == -1 && o2.isLtrDominant() == -1 ? -java.lang.Double.compare(o1.getX(), o2.getX()) : java.lang.Double.compare(o1.getX(), o2.getX());
-            } else {
-                return java.lang.Float.compare(o1.getBottom(), o2.getBottom());
-            }
-        }
-    };
-
-
-    public Rectangle() {
-
-        super();
-    }
-
-
-    public Rectangle(float top, float left, float width, float height) {
-
-        super();
-        this.setRect(left, top, width, height);
-    }
-
-
-    /**
-     * @param rectangles
-     * @return minimum bounding box that contains all the rectangles
-     */
-    public static Rectangle boundingBoxOf(List<? extends Rectangle> rectangles) {
-
-        float minx = java.lang.Float.MAX_VALUE;
-        float miny = java.lang.Float.MAX_VALUE;
-        float maxx = java.lang.Float.MIN_VALUE;
-        float maxy = java.lang.Float.MIN_VALUE;
-
-        for (Rectangle r : rectangles) {
-            minx = (float) Math.min(r.getMinX(), minx);
-            miny = (float) Math.min(r.getMinY(), miny);
-            maxx = (float) Math.max(r.getMaxX(), maxx);
-            maxy = (float) Math.max(r.getMaxY(), maxy);
-        }
-        return new Rectangle(miny, minx, maxx - minx, maxy - miny);
-    }
-
-
-    public int compareTo(Rectangle other) {
-
-        return ILL_DEFINED_ORDER.compare(this, other);
-    }
-
-
-    // I'm bad at Java and need this for fancy sorting in
-    // technology.tabula.TextChunk.
-    public int isLtrDominant() {
-
-        return 0;
-    }
-
-
-    public float getArea() {
-
-        return this.width * this.height;
-    }
-
-
-    public float verticalOverlap(Rectangle other) {
-
-        return Math.max(0, Math.min(this.getBottom(), other.getBottom()) - Math.max(this.getTop(), other.getTop()));
-    }
-
-
-    public boolean verticallyOverlaps(Rectangle other) {
-
-        return verticalOverlap(other) > 0;
-    }
-
-
-    public float horizontalOverlap(Rectangle other) {
-
-        return Math.max(0, Math.min(this.getRight(), other.getRight()) - Math.max(this.getLeft(), other.getLeft()));
-    }
-
-
-    public boolean horizontallyOverlaps(Rectangle other) {
-
-        return horizontalOverlap(other) > 0;
-    }
-
-
-    public float verticalOverlapRatio(Rectangle other) {
-
-        float rv = 0, delta = Math.min(this.getBottom() - this.getTop(), other.getBottom() - other.getTop());
-
-        if (other.getTop() <= this.getTop() && this.getTop() <= other.getBottom() && other.getBottom() <= this.getBottom()) {
-            rv = (other.getBottom() - this.getTop()) / delta;
-        } else if (this.getTop() <= other.getTop() && other.getTop() <= this.getBottom() && this.getBottom() <= other.getBottom()) {
-            rv = (this.getBottom() - other.getTop()) / delta;
-        } else if (this.getTop() <= other.getTop() && other.getTop() <= other.getBottom() && other.getBottom() <= this.getBottom()) {
-            rv = (other.getBottom() - other.getTop()) / delta;
-        } else if (other.getTop() <= this.getTop() && this.getTop() <= this.getBottom() && this.getBottom() <= other.getBottom()) {
-            rv = (this.getBottom() - this.getTop()) / delta;
-        }
-
-        return rv;
-
-    }
-
-
-    public float overlapRatio(Rectangle other) {
-
-        double intersectionWidth = Math.max(0, Math.min(this.getRight(), other.getRight()) - Math.max(this.getLeft(), other.getLeft()));
-        double intersectionHeight = Math.max(0, Math.min(this.getBottom(), other.getBottom()) - Math.max(this.getTop(), other.getTop()));
-        double intersectionArea = Math.max(0, intersectionWidth * intersectionHeight);
-        double unionArea = this.getArea() + other.getArea() - intersectionArea;
-
-        return (float) (intersectionArea / unionArea);
-    }
-
-
-    public Rectangle merge(Rectangle other) {
-
-        this.setRect(this.createUnion(other));
-        return this;
-    }
-
-
-    public float getTop() {
-
-        return (float) this.getMinY();
-    }
-
-
-    public void setTop(float top) {
-
-        float deltaHeight = top - this.y;
-        this.setRect(this.x, top, this.width, this.height - deltaHeight);
-    }
-
-
-    public float getRight() {
-
-        return (float) this.getMaxX();
-    }
-
-
-    public void setRight(float right) {
-
-        this.setRect(this.x, this.y, right - this.x, this.height);
-    }
-
-
-    public float getLeft() {
-
-        return (float) this.getMinX();
-    }
-
-
-    public void setLeft(float left) {
-
-        float deltaWidth = left - this.x;
-        this.setRect(left, this.y, this.width - deltaWidth, this.height);
-    }
-
-
-    public float getBottom() {
-
-        return (float) this.getMaxY();
-    }
-
-
-    public void setBottom(float bottom) {
-
-        this.setRect(this.x, this.y, this.width, bottom - this.y);
-    }
-
-
-    public Point2D[] getPoints() {
-
-        return new Point2D[]{new Point2D.Float(this.getLeft(), this.getTop()), new Point2D.Float(this.getRight(), this.getTop()), new Point2D.Float(this.getRight(),
-                this.getBottom()), new Point2D.Float(this.getLeft(), this.getBottom())};
-    }
-
-
-    @Override
-    public String toString() {
-
-        StringBuilder sb = new StringBuilder();
-        String s = super.toString();
-        sb.append(s.substring(0, s.length() - 1));
-        sb.append(String.format(",bottom=%f,right=%f]", this.getBottom(), this.getRight()));
-        return sb.toString();
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/table/Ruling.java

@@ -1,437 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
-
-import java.awt.geom.Line2D;
-import java.awt.geom.Point2D;
-import java.awt.geom.Rectangle2D;
-import java.util.ArrayList;
-import java.util.Collections;
-import java.util.Comparator;
-import java.util.Formatter;
-import java.util.List;
-import java.util.Map;
-import java.util.TreeMap;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.CohenSutherlandClipping;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.Utils;
-
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@SuppressWarnings("all")
-public class Ruling extends Line2D.Float {
-
-    private static int PERPENDICULAR_PIXEL_EXPAND_AMOUNT = 2;
-
-
-    public Ruling(Point2D p1, Point2D p2) {
-
-        super(p1, p2);
-    }
-
-
-    public static List<Ruling> cropRulingsToArea(List<Ruling> rulings, Rectangle2D area) {
-
-        ArrayList<Ruling> rv = new ArrayList<>();
-        for (Ruling r : rulings) {
-            if (r.intersects(area)) {
-                rv.add(r.intersect(area));
-            }
-        }
-        return rv;
-    }
-
-
-    // log(n) implementation of find_intersections
-    // based on http://people.csail.mit.edu/indyk/6.838-old/handouts/lec2.pdf
-    public static Map<Point2D, Ruling[]> findIntersections(List<Ruling> horizontals, List<Ruling> verticals) {
-
-        class SortObject {
-
-            protected SOType type;
-            protected float position;
-            protected Ruling ruling;
-
-
-            public SortObject(SOType type, float position, Ruling ruling) {
-
-                this.type = type;
-                this.position = position;
-                this.ruling = ruling;
-            }
-
-        }
-
-        List<SortObject> sos = new ArrayList<>();
-
-        TreeMap<Ruling, Boolean> tree = new TreeMap<>(new Comparator<Ruling>() {
-            @Override
-            public int compare(Ruling o1, Ruling o2) {
-
-                return java.lang.Double.compare(o1.getTop(), o2.getTop());
-            }
-        });
-
-        TreeMap<Point2D, Ruling[]> rv = new TreeMap<>(new Comparator<Point2D>() {
-            @Override
-            public int compare(Point2D o1, Point2D o2) {
-
-                if (o1.getY() > o2.getY()) {
-                    return 1;
-                }
-                if (o1.getY() < o2.getY()) {
-                    return -1;
-                }
-                if (o1.getX() > o2.getX()) {
-                    return 1;
-                }
-                if (o1.getX() < o2.getX()) {
-                    return -1;
-                }
-                return 0;
-            }
-        });
-
-        for (Ruling h : horizontals) {
-            sos.add(new SortObject(SOType.HLEFT, h.getLeft() - PERPENDICULAR_PIXEL_EXPAND_AMOUNT, h));
-            sos.add(new SortObject(SOType.HRIGHT, h.getRight() + PERPENDICULAR_PIXEL_EXPAND_AMOUNT, h));
-        }
-
-        for (Ruling v : verticals) {
-            sos.add(new SortObject(SOType.VERTICAL, v.getLeft(), v));
-        }
-
-        Collections.sort(sos, new Comparator<SortObject>() {
-            @Override
-            public int compare(SortObject a, SortObject b) {
-
-                int rv;
-                if (Utils.feq(a.position, b.position)) {
-                    if (a.type == SOType.VERTICAL && b.type == SOType.HLEFT) {
-                        rv = 1;
-                    } else if (a.type == SOType.VERTICAL && b.type == SOType.HRIGHT) {
-                        rv = -1;
-                    } else if (a.type == SOType.HLEFT && b.type == SOType.VERTICAL) {
-                        rv = -1;
-                    } else if (a.type == SOType.HRIGHT && b.type == SOType.VERTICAL) {
-                        rv = 1;
-                    } else {
-                        rv = java.lang.Double.compare(a.position, b.position);
-                    }
-                } else {
-                    return java.lang.Double.compare(a.position, b.position);
-                }
-                return rv;
-            }
-        });
-
-        for (SortObject so : sos) {
-            switch (so.type) {
-                case VERTICAL:
-                    for (Map.Entry<Ruling, Boolean> h : tree.entrySet()) {
-                        try {
-                            Point2D i = h.getKey().intersectionPoint(so.ruling);
-                            if (i == null) {
-                                continue;
-                            }
-                            rv.put(i, new Ruling[]{h.getKey().expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT), so.ruling.expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT)});
-                        } catch (UnsupportedOperationException e) {
-                            log.info("Some line are oblique, ignoring...");
-                            continue;
-                        }
-                    }
-                    break;
-                case HRIGHT:
-                    tree.remove(so.ruling);
-                    break;
-                case HLEFT:
-                    tree.put(so.ruling, true);
-                    break;
-            }
-        }
-
-        return rv;
-
-    }
-
-
-    public boolean vertical() {
-
-        return this.length() > 0 && Utils.feq(this.x1, this.x2); //diff < ORIENTATION_CHECK_THRESHOLD;
-    }
-
-
-    public boolean horizontal() {
-
-        return this.length() > 0 && Utils.feq(this.y1, this.y2); //diff < ORIENTATION_CHECK_THRESHOLD;
-    }
-
-    // attributes that make sense only for non-oblique lines
-    // these are used to have a single collapse method (in page, currently)
-
-
-    public boolean oblique() {
-
-        return !(this.vertical() || this.horizontal());
-    }
-
-
-    public float getPosition() {
-
-        if (this.oblique()) {
-            throw new UnsupportedOperationException();
-        }
-        return this.vertical() ? this.getLeft() : this.getTop();
-    }
-
-
-    public float getStart() {
-
-        if (this.oblique()) {
-            throw new UnsupportedOperationException();
-        }
-        return this.vertical() ? this.getTop() : this.getLeft();
-    }
-
-
-    public void setStart(float v) {
-
-        if (this.oblique()) {
-            throw new UnsupportedOperationException();
-        }
-        if (this.vertical()) {
-            this.setTop(v);
-        } else {
-            this.setLeft(v);
-        }
-    }
-
-
-    public float getEnd() {
-
-        if (this.oblique()) {
-            throw new UnsupportedOperationException();
-        }
-        return this.vertical() ? this.getBottom() : this.getRight();
-    }
-
-
-    public void setEnd(float v) {
-
-        if (this.oblique()) {
-            throw new UnsupportedOperationException();
-        }
-        if (this.vertical()) {
-            this.setBottom(v);
-        } else {
-            this.setRight(v);
-        }
-    }
-
-
-    public void setStartEnd(float start, float end) {
-
-        if (this.oblique()) {
-            throw new UnsupportedOperationException();
-        }
-        if (this.vertical()) {
-            this.setTop(start);
-            this.setBottom(end);
-        } else {
-            this.setLeft(start);
-            this.setRight(end);
-        }
-    }
-
-
-    public boolean perpendicularTo(Ruling other) {
-
-        return this.vertical() == other.horizontal();
-    }
-
-
-    public boolean nearlyIntersects(Ruling another, int colinearOrParallelExpandAmount) {
-
-        if (this.intersectsLine(another)) {
-            return true;
-        }
-
-        boolean rv = false;
-
-        if (this.perpendicularTo(another)) {
-            rv = this.expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT).intersectsLine(another);
-        } else {
-            rv = this.expand(colinearOrParallelExpandAmount).intersectsLine(another.expand(colinearOrParallelExpandAmount));
-        }
-
-        return rv;
-    }
-
-
-    public double length() {
-
-        return Math.sqrt(Math.pow(this.x1 - this.x2, 2) + Math.pow(this.y1 - this.y2, 2));
-    }
-
-
-    public Ruling intersect(Rectangle2D clip) {
-
-        Float clipee = (Float) this.clone();
-        boolean clipped = new CohenSutherlandClipping(clip).clip(clipee);
-
-        if (clipped) {
-            return new Ruling(clipee.getP1(), clipee.getP2());
-        } else {
-            return this;
-        }
-    }
-
-
-    public Ruling expand(float amount) {
-
-        Ruling r = (Ruling) this.clone();
-        try {
-            r.setStart(this.getStart() - amount);
-            r.setEnd(this.getEnd() + amount);
-        } catch (UnsupportedOperationException e) {
-            log.warn("Could not expand ruling!");
-        }
-        return r;
-    }
-
-
-    public Point2D intersectionPoint(Ruling other) {
-
-        Ruling this_l = this.expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT);
-        Ruling other_l = other.expand(PERPENDICULAR_PIXEL_EXPAND_AMOUNT);
-        Ruling horizontal, vertical;
-
-        if (!this_l.intersectsLine(other_l)) {
-            return null;
-        }
-
-        if (this_l.horizontal() && other_l.vertical()) {
-            horizontal = this_l;
-            vertical = other_l;
-        } else if (this_l.vertical() && other_l.horizontal()) {
-            vertical = this_l;
-            horizontal = other_l;
-        } else {
-            log.warn("lines must be orthogonal, vertical and horizontal");
-            return null;
-        }
-        return new Point2D.Float(vertical.getLeft(), horizontal.getTop());
-    }
-
-
-    @Override
-    public boolean equals(Object other) {
-
-        if (this == other) {
-            return true;
-        }
-
-        if (!(other instanceof Ruling)) {
-            return false;
-        }
-
-        Ruling o = (Ruling) other;
-        return this.getP1().equals(o.getP1()) && this.getP2().equals(o.getP2());
-    }
-
-
-    @Override
-    public int hashCode() {
-
-        return super.hashCode();
-    }
-
-
-    public float getTop() {
-
-        return this.y1;
-    }
-
-
-    public void setTop(float v) {
-
-        setLine(this.getLeft(), v, this.getRight(), this.getBottom());
-    }
-
-
-    public float getLeft() {
-
-        return this.x1;
-    }
-
-
-    public void setLeft(float v) {
-
-        setLine(v, this.getTop(), this.getRight(), this.getBottom());
-    }
-
-
-    public float getBottom() {
-
-        return this.y2;
-    }
-
-
-    public void setBottom(float v) {
-
-        setLine(this.getLeft(), this.getTop(), this.getRight(), v);
-    }
-
-
-    public float getRight() {
-
-        return this.x2;
-    }
-
-
-    public void setRight(float v) {
-
-        setLine(this.getLeft(), this.getTop(), v, this.getBottom());
-    }
-
-
-    public float getWidth() {
-
-        return this.getRight() - this.getLeft();
-    }
-
-
-    public float getHeight() {
-
-        return this.getBottom() - this.getTop();
-    }
-
-
-    public double getAngle() {
-
-        double angle = Math.toDegrees(Math.atan2(this.getP2().getY() - this.getP1().getY(), this.getP2().getX() - this.getP1().getX()));
-
-        if (angle < 0) {
-            angle += 360;
-        }
-        return angle;
-    }
-
-
-    @Override
-    public String toString() {
-
-        StringBuilder sb = new StringBuilder();
-        Formatter formatter = new Formatter(sb);
-        String rv = formatter.format("%s[minX=%f minY=%f maxX=%f maxY=%f]", this.getClass().toString(), this.x1, this.y1, this.x2, this.y2).toString();
-        formatter.close();
-        return rv;
-    }
-
-
-    private enum SOType {
-        VERTICAL,
-        HRIGHT,
-        HLEFT
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/table/TablePageBlock.java

@@ -1,342 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table;
-
-import java.awt.geom.Point2D;
-import java.util.ArrayList;
-import java.util.Collections;
-import java.util.HashSet;
-import java.util.List;
-import java.util.Set;
-import java.util.TreeMap;
-
-import org.apache.commons.collections4.CollectionUtils;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-
-import lombok.Getter;
-import lombok.Setter;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-public class TablePageBlock extends AbstractPageBlock {
-
-    private final TreeMap<CellPosition, Cell> cells = new TreeMap<>();
-
-    private final int rotation;
-    @Getter
-    @Setter
-    private String headline;
-    private int unrotatedRowCount;
-    private int unrotatedColCount;
-    private List<List<Cell>> rows;
-
-
-    public TablePageBlock(List<Cell> cells, Rectangle area, int rotation) {
-
-        addCells(cells);
-        minX = area.getLeft();
-        minY = area.getBottom();
-        maxX = area.getRight();
-        maxY = area.getTop();
-        classification = PageBlockType.TABLE;
-        this.rotation = rotation;
-    }
-
-
-    public List<List<Cell>> getRows() {
-
-        if (rows == null) {
-            rows = computeRows();
-
-            // Ignore rows that does not contain any cells and values.
-            List<List<Cell>> rowsToRemove = new ArrayList<>();
-            for (List<Cell> row : rows) {
-                if (row.size() == 1 && row.get(0).getTextBlocks().isEmpty()) {
-                    rowsToRemove.add(row);
-                }
-            }
-            rows.removeAll(rowsToRemove);
-
-            computeHeaders();
-        }
-
-        return rows;
-
-    }
-
-
-    public int getRowCount() {
-
-        return getRows().size();
-    }
-
-
-    public int getColCount() {
-
-        return getRows().stream().mapToInt(List::size).max().orElse(0);
-
-    }
-
-
-    /**
-     * Detect header cells (either first row or first column):
-     * Column is marked as header if cell text is bold and row cell text is not bold.
-     * Defaults to row.
-     */
-    private void computeHeaders() {
-
-        if (rows == null) {
-            rows = computeRows();
-        }
-        // A bold cell is a header cell as long as every cell to the left/top is bold, too
-        // we move from left to right and top to bottom
-        for (int rowIndex = 0; rowIndex < rows.size(); rowIndex++) {
-            List<Cell> rowCells = rows.get(rowIndex);
-            if (rowCells.size() == 1) {
-                continue;
-            }
-
-            for (int colIndex = 0; colIndex < rowCells.size(); colIndex++) {
-                Cell cell = rowCells.get(colIndex);
-                List<Cell> cellsToTheLeft = rowCells.subList(0, colIndex);
-                Cell lastHeaderCell = null;
-                for (Cell leftCell : cellsToTheLeft) {
-                    if (leftCell.isHeaderCell()) {
-                        lastHeaderCell = leftCell;
-                    } else {
-                        break;
-                    }
-                }
-                if (lastHeaderCell != null) {
-                    cell.getHeaderCells().add(lastHeaderCell);
-                }
-                List<Cell> cellsToTheTop = new ArrayList<>();
-                for (int i = 0; i < rowIndex; i++) {
-                    try {
-                        cellsToTheTop.add(rows.get(i).get(colIndex));
-                    } catch (IndexOutOfBoundsException e) {
-                        log.debug("No cell {} in row {}, ignoring.", colIndex, rowIndex);
-                    }
-                }
-                for (Cell topCell : cellsToTheTop) {
-                    if (topCell.isHeaderCell()) {
-                        lastHeaderCell = topCell;
-                    } else {
-                        break;
-                    }
-                }
-                if (lastHeaderCell != null) {
-                    cell.getHeaderCells().add(lastHeaderCell);
-                }
-                if (CollectionUtils.isNotEmpty(cell.getTextBlocks()) && cell.getTextBlocks().get(0).getMostPopularWordStyle().equals("bold")) {
-                    cell.setHeaderCell(true);
-                }
-            }
-        }
-
-    }
-
-
-    private List<List<Cell>> computeRows() {
-
-        List<List<Cell>> rows = new ArrayList<>();
-        if (rotation == 90) {
-            for (int i = 0; i < unrotatedColCount; i++) { // rows
-                List<Cell> lastRow = new ArrayList<>();
-                for (int j = unrotatedRowCount - 1; j >= 0; j--) { // cols
-                    Cell cell = cells.get(new CellPosition(j, i));
-                    if (cell != null) {
-                        lastRow.add(cell);
-                    }
-                }
-                rows.add(lastRow);
-            }
-        } else if (rotation == 270) {
-            for (int i = unrotatedColCount - 1; i >= 0; i--) { // rows
-                List<Cell> lastRow = new ArrayList<>();
-                for (int j = 0; j < unrotatedRowCount; j++) { // cols
-                    Cell cell = cells.get(new CellPosition(j, i));
-                    if (cell != null) {
-                        lastRow.add(cell);
-                    }
-                }
-                rows.add(lastRow);
-            }
-        } else {
-            for (int i = 0; i < unrotatedRowCount; i++) {
-                List<Cell> lastRow = new ArrayList<>();
-                for (int j = 0; j < unrotatedColCount; j++) {
-                    Cell cell = cells.get(new CellPosition(i, j)); // JAVA_8 use getOrDefault()
-                    if (cell != null) {
-                        lastRow.add(cell);
-                    }
-                }
-                rows.add(lastRow);
-            }
-        }
-
-        return rows;
-
-    }
-
-
-    private void add(Cell chunk, int row, int col) {
-
-        unrotatedRowCount = Math.max(unrotatedRowCount, row + 1);
-        unrotatedColCount = Math.max(unrotatedColCount, col + 1);
-
-        CellPosition cp = new CellPosition(row, col);
-        cells.put(cp, chunk);
-
-    }
-
-
-    private void addCells(List<Cell> cells) {
-
-        if (cells.isEmpty()) {
-            return;
-        }
-
-        cells.removeIf(cell -> cell.getWidth() < 1.1 || cell.getHeight() < 1.1);
-
-        List<List<Cell>> rowsOfCells = calculateStructure(cells);
-
-        for (int i = 0; i < rowsOfCells.size(); i++) {
-            for (int j = 0; j < rowsOfCells.get(i).size(); j++) {
-                add(rowsOfCells.get(i).get(j), i, j);
-            }
-        }
-
-    }
-
-
-    /**
-     * Calculates the structure of the table. For spanning rows and columns multiple cells with the same values will be inserted.
-     *
-     * @param cells The found cells
-     * @return TablePageBlock Structure
-     */
-    private List<List<Cell>> calculateStructure(List<Cell> cells) {
-
-        List<List<Cell>> matrix = new ArrayList<>();
-
-        if (cells.isEmpty()) {
-            return matrix;
-        }
-
-        Set<Float> uniqueX = new HashSet<>();
-        Set<Float> uniqueY = new HashSet<>();
-        cells.stream().filter(c -> !c.getTextBlocks().isEmpty() || c.getHeight() > 3 && c.getWidth() > 3).forEach(c -> {
-            uniqueX.add(c.getLeft());
-            uniqueX.add(c.getRight());
-            uniqueY.add(c.getBottom());
-            uniqueY.add(c.getTop());
-        });
-
-        var sortedUniqueX = uniqueX.stream().sorted().toList();
-        var sortedUniqueY = uniqueY.stream().sorted().toList();
-
-        Float prevY = null;
-        for (Float y : sortedUniqueY) {
-
-            List<Cell> row = new ArrayList<>();
-
-            Float prevX = null;
-            for (Float x : sortedUniqueX) {
-
-                if (prevY != null && prevX != null) {
-                    var cell = new Cell(new Point2D.Float(prevX, prevY), new Point2D.Float(x, y));
-
-                    var intersectionCell = cells.stream().filter(c -> cell.intersects(c) && cell.overlapRatio(c) > 0.1f).findFirst();
-                    intersectionCell.ifPresent(value -> cell.getTextBlocks().addAll(value.getTextBlocks()));
-                    if (cell.hasMinimumSize()) {
-                        row.add(cell);
-                    }
-                }
-                prevX = x;
-            }
-
-            if (prevY != null && prevX != null && !row.isEmpty()) {
-                matrix.add(row);
-            }
-            prevY = y;
-        }
-
-        Collections.reverse(matrix);
-
-        return matrix;
-    }
-
-
-    @Override
-    public String getText() {
-
-        StringBuilder sb = new StringBuilder();
-        List<List<Cell>> rows = getRows();
-
-        int i = 0;
-        for (List<Cell> row : rows) {
-            if (i != 0) {
-                sb.append("\n");
-            }
-            if (!row.isEmpty()) {
-                boolean firstColumn = true;
-                for (Cell column : row) {
-                    if (!firstColumn) {
-                        sb.append(",");
-                    }
-                    if (column != null && column.getTextBlocks() != null) {
-                        boolean first = true;
-                        for (TextPageBlock textBlock : column.getTextBlocks()) {
-                            if (!first) {
-                                sb.append("\n");
-                            }
-                            sb.append('\"').append(textBlock.getText().replaceAll("\"", "\\\"")).append('\"');
-                            first = false;
-                        }
-                    }
-                    firstColumn = false;
-                }
-            }
-            i++;
-        }
-
-        return sb.toString();
-    }
-
-
-    public String getTextAsHtml() {
-
-        StringBuilder sb = new StringBuilder();
-        List<List<Cell>> rows = getRows();
-
-        sb.append("<table border=\"1\">");
-        int i = 0;
-        for (List<Cell> row : rows) {
-            sb.append("\n<tr>");
-            if (!row.isEmpty()) {
-                for (Cell column : row) {
-                    sb.append(i == 0 ? "\n<th>" : "\n<td>");
-                    if (column != null && column.getTextBlocks() != null) {
-                        boolean first = true;
-                        for (TextPageBlock textBlock : column.getTextBlocks()) {
-                            if (!first) {
-                                sb.append("<br />");
-                            }
-                            sb.append(textBlock.getText().replaceAll("\\n", "<br />"));
-                            first = false;
-                        }
-                    }
-                    sb.append(i == 0 ? "</th>" : "</td>");
-                }
-            }
-            sb.append("</tr>");
-            i++;
-        }
-        sb.append("</table>");
-
-        return sb.toString();
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/RedTextPosition.java

@@ -1,100 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import org.apache.pdfbox.text.TextPosition;
-import org.springframework.beans.BeanUtils;
-
-import com.fasterxml.jackson.annotation.JsonIgnore;
-
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.NoArgsConstructor;
-import lombok.SneakyThrows;
-
-@Data
-@Builder
-@NoArgsConstructor
-@AllArgsConstructor
-public class RedTextPosition {
-
-    private String textMatrix;
-    private float[] position;
-
-    @JsonIgnore
-    private int rotation;
-
-    @JsonIgnore
-    private float pageHeight;
-
-    @JsonIgnore
-    private float pageWidth;
-
-    private String unicode;
-
-    @JsonIgnore
-    private float dir;
-
-    // not used in reanalysis
-    @JsonIgnore
-    private float widthOfSpace;
-
-    // not used in reanalysis
-    @JsonIgnore
-    private float fontSizeInPt;
-
-    // not used in reanalysis
-    @JsonIgnore
-    private String fontName;
-
-
-    @SneakyThrows
-    public static RedTextPosition fromTextPosition(TextPosition textPosition) {
-
-        var pos = new RedTextPosition();
-        BeanUtils.copyProperties(textPosition, pos);
-        pos.setFontName(textPosition.getFont().getName());
-
-        pos.setFontSizeInPt(textPosition.getFontSizeInPt());
-
-        pos.setTextMatrix(textPosition.getTextMatrix().toString());
-
-        var position = new float[4];
-
-        position[0] = textPosition.getXDirAdj();
-        position[1] = textPosition.getYDirAdj();
-        position[2] = textPosition.getWidthDirAdj();
-        position[3] = textPosition.getHeightDir();
-
-        pos.setPosition(position);
-        return pos;
-    }
-
-
-    @JsonIgnore
-    public float getXDirAdj() {
-
-        return position[0];
-    }
-
-
-    @JsonIgnore
-    public float getYDirAdj() {
-
-        return position[1];
-    }
-
-
-    @JsonIgnore
-    public float getWidthDirAdj() {
-
-        return position[2];
-    }
-
-
-    @JsonIgnore
-    public float getHeightDir() {
-
-        return position[3];
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/SearchableText.java

@@ -1,49 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
-
-import lombok.Getter;
-
-public class SearchableText {
-
-    @Getter
-    private final List<TextPositionSequence> sequences = new ArrayList<>();
-
-
-    public void add(TextPositionSequence textPositionSequence) {
-
-        sequences.add(textPositionSequence);
-    }
-
-
-    public void addAll(List<TextPositionSequence> textPositionSequences) {
-
-        sequences.addAll(textPositionSequences);
-    }
-
-
-    @Override
-    public String toString() {
-
-        return buildString(sequences);
-    }
-
-
-    public static String buildString(List<TextPositionSequence> sequences) {
-
-        StringBuilder sb = new StringBuilder();
-        for (TextPositionSequence word : sequences) {
-            sb.append(word);
-            sb.append(' ');
-        }
-        String text = sb.toString();
-        text = TextNormalizationUtilities.removeHyphenLineBreaks(text);
-        text = TextNormalizationUtilities.removeLineBreaks(text);
-        text = TextNormalizationUtilities.removeRepeatingWhitespaces(text);
-        return text;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/SimplifiedSectionText.java

@@ -1,17 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.NoArgsConstructor;
-
-@Data
-@Builder
-@NoArgsConstructor
-@AllArgsConstructor
-public class SimplifiedSectionText {
-
-    private int sectionNumber;
-    private String text;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/SimplifiedText.java

@@ -1,20 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.NoArgsConstructor;
-
-@Data
-@Builder
-@NoArgsConstructor
-@AllArgsConstructor
-public class SimplifiedText {
-
-    private int numberOfPages;
-    private List<SimplifiedSectionText> sectionTexts = new ArrayList<>();
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/StringFrequencyCounter.java

@@ -1,47 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import java.util.HashMap;
-import java.util.Map;
-
-import lombok.Getter;
-
-public class StringFrequencyCounter {
-
-    @Getter
-    private final Map<String, Integer> countPerValue = new HashMap<>();
-
-
-    public void add(String value) {
-
-        if (!countPerValue.containsKey(value)) {
-            countPerValue.put(value, 1);
-        } else {
-            countPerValue.put(value, countPerValue.get(value) + 1);
-        }
-    }
-
-
-    public void addAll(Map<String, Integer> otherCounter) {
-
-        for (Map.Entry<String, Integer> entry : otherCounter.entrySet()) {
-            if (countPerValue.containsKey(entry.getKey())) {
-                countPerValue.put(entry.getKey(), countPerValue.get(entry.getKey()) + entry.getValue());
-            } else {
-                countPerValue.put(entry.getKey(), entry.getValue());
-            }
-        }
-    }
-
-
-    public String getMostPopular() {
-
-        Map.Entry<String, Integer> mostPopular = null;
-        for (Map.Entry<String, Integer> entry : countPerValue.entrySet()) {
-            if (mostPopular == null || entry.getValue() > mostPopular.getValue()) {
-                mostPopular = entry;
-            }
-        }
-        return mostPopular != null ? mostPopular.getKey() : null;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/TextDirection.java

@@ -1,47 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import com.fasterxml.jackson.annotation.JsonCreator;
-import com.fasterxml.jackson.annotation.JsonValue;
-
-import lombok.Getter;
-
-@Getter
-public enum TextDirection {
-    ZERO(0f),
-    QUARTER_CIRCLE(90f),
-    HALF_CIRCLE(180f),
-    THREE_QUARTER_CIRCLE(270f);
-
-    public static final String VALUE_STRING_SUFFIX = "°";
-
-    @JsonValue
-    private final float degrees;
-    private final float radians;
-
-
-    TextDirection(float degreeValue) {
-
-        degrees = degreeValue;
-        radians = (float) Math.toRadians(degreeValue);
-    }
-
-
-    @Override
-    public String toString() {
-
-        return degrees + VALUE_STRING_SUFFIX;
-    }
-
-
-    @JsonCreator(mode = JsonCreator.Mode.DELEGATING)
-    public static TextDirection fromDegrees(float degrees) {
-
-        for (var dir : TextDirection.values()) {
-            if (degrees == dir.degrees) {
-                return dir;
-            }
-        }
-
-        throw new IllegalArgumentException(String.format("A value of %f is not supported by TextDirection", degrees));
-    }
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/TextPageBlock.java

@@ -1,302 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import java.util.ArrayList;
-import java.util.List;
-
-import com.fasterxml.jackson.annotation.JsonIgnore;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
-
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.NoArgsConstructor;
-
-@AllArgsConstructor
-@Builder
-@Data
-@NoArgsConstructor
-public class TextPageBlock extends AbstractPageBlock {
-
-    @Builder.Default
-    private List<TextPositionSequence> sequences = new ArrayList<>();
-
-    @JsonIgnore
-    private int rotation;
-
-    @JsonIgnore
-    private String mostPopularWordFont;
-
-    @JsonIgnore
-    private String mostPopularWordStyle;
-
-    @JsonIgnore
-    private float mostPopularWordFontSize;
-
-    @JsonIgnore
-    private float mostPopularWordHeight;
-
-    @JsonIgnore
-    private float mostPopularWordSpaceWidth;
-
-    @JsonIgnore
-    private float highestFontSize;
-
-    @JsonIgnore
-    private PageBlockType classification;
-
-
-    @JsonIgnore
-    public TextDirection getDir() {
-
-        return sequences.get(0).getDir();
-    }
-
-
-    @JsonIgnore
-    private float getPageHeight() {
-
-        return sequences.get(0).getPageHeight();
-    }
-
-
-    @JsonIgnore
-    private float getPageWidth() {
-
-        return sequences.get(0).getPageWidth();
-    }
-
-
-    /**
-     * Returns the minX value in pdf coordinate system.
-     * Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
-     * 0 -> LowerLeft
-     * 90 -> UpperLeft
-     * 180 -> UpperRight
-     * 270 -> LowerRight
-     *
-     * @return the minX value in pdf coordinate system
-     */
-    @JsonIgnore
-    public float getPdfMinX() {
-
-        if (getDir().getDegrees() == 90) {
-            return minY;
-        } else if (getDir().getDegrees() == 180) {
-            return getPageWidth() - maxX;
-
-        } else if (getDir().getDegrees() == 270) {
-
-            return getPageWidth() - maxY;
-        } else {
-            return minX;
-        }
-    }
-
-
-    /**
-     * Returns the maxX value in pdf coordinate system.
-     * Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
-     * 0 -> LowerLeft
-     * 90 -> UpperLeft
-     * 180 -> UpperRight
-     * 270 -> LowerRight
-     *
-     * @return the maxX value in pdf coordinate system
-     */
-    @JsonIgnore
-    public float getPdfMaxX() {
-
-        if (getDir().getDegrees() == 90) {
-            return maxY;
-        } else if (getDir().getDegrees() == 180) {
-            return getPageWidth() - minX;
-        } else if (getDir().getDegrees() == 270) {
-            return getPageWidth() - minY;
-
-        } else {
-            return maxX;
-        }
-    }
-
-
-    /**
-     * Returns the minY value in pdf coordinate system.
-     * Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
-     * 0 -> LowerLeft
-     * 90 -> UpperLeft
-     * 180 -> UpperRight
-     * 270 -> LowerRight
-     *
-     * @return the minY value in pdf coordinate system
-     */
-    @JsonIgnore
-    public float getPdfMinY() {
-
-        if (getDir().getDegrees() == 90) {
-            return minX;
-        } else if (getDir().getDegrees() == 180) {
-            return maxY;
-
-        } else if (getDir().getDegrees() == 270) {
-            return getPageHeight() - maxX;
-
-        } else {
-            return getPageHeight() - maxY;
-        }
-    }
-
-
-    /**
-     * Returns the maxY value in pdf coordinate system.
-     * Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
-     * 0 -> LowerLeft
-     * 90 -> UpperLeft
-     * 180 -> UpperRight
-     * 270 -> LowerRight
-     *
-     * @return the maxY value in pdf coordinate system
-     */
-    @JsonIgnore
-    public float getPdfMaxY() {
-
-        if (getDir().getDegrees() == 90) {
-            return maxX;
-        } else if (getDir().getDegrees() == 180) {
-
-            return minY;
-        } else if (getDir().getDegrees() == 270) {
-            return getPageHeight() - minX;
-        } else {
-            return getPageHeight() - minY;
-        }
-    }
-
-
-    public TextPageBlock(float minX, float maxX, float minY, float maxY, List<TextPositionSequence> sequences, int rotation) {
-
-        this.minX = minX;
-        this.maxX = maxX;
-        this.minY = minY;
-        this.maxY = maxY;
-        this.sequences = sequences;
-        this.rotation = rotation;
-    }
-
-
-    public TextPageBlock union(TextPositionSequence r) {
-
-        TextPageBlock union = this.copy();
-        union.add(r);
-        return union;
-    }
-
-
-    public TextPageBlock union(TextPageBlock r) {
-
-        TextPageBlock union = this.copy();
-        union.add(r);
-        return union;
-    }
-
-
-    public void add(TextPageBlock r) {
-
-        if (r.getMinX() < minX) {
-            minX = r.getMinX();
-        }
-        if (r.getMaxX() > maxX) {
-            maxX = r.getMaxX();
-        }
-        if (r.getMinY() < minY) {
-            minY = r.getMinY();
-        }
-        if (r.getMaxY() > maxY) {
-            maxY = r.getMaxY();
-        }
-        sequences.addAll(r.getSequences());
-    }
-
-
-    public void add(TextPositionSequence r) {
-
-        if (r.getMinXDirAdj() < minX) {
-            minX = r.getMinXDirAdj();
-        }
-        if (r.getMaxXDirAdj() > maxX) {
-            maxX = r.getMaxXDirAdj();
-        }
-        if (r.getMinYDirAdj() < minY) {
-            minY = r.getMinYDirAdj();
-        }
-        if (r.getMaxYDirAdj() > maxY) {
-            maxY = r.getMaxYDirAdj();
-        }
-    }
-
-
-    public TextPageBlock copy() {
-
-        return new TextPageBlock(minX, maxX, minY, maxY, sequences, rotation);
-    }
-
-
-    public void resize(float x1, float y1, float width, float height) {
-
-        set(x1, y1, x1 + width, y1 + height);
-    }
-
-
-    public void set(float x1, float y1, float x2, float y2) {
-
-        this.minX = Math.min(x1, x2);
-        this.maxX = Math.max(x1, x2);
-        this.minY = Math.min(y1, y2);
-        this.maxY = Math.max(y1, y2);
-    }
-
-
-    @Override
-    public String toString() {
-
-        StringBuilder builder = new StringBuilder();
-
-        for (int i = 0; i < sequences.size(); i++) {
-            String sequenceAsString = sequences.get(i).toString();
-            // Fix for missing Whitespace. This is recognized in getSequences method. See PDFTextStripper Line 1730.
-            if (i != 0 && sequences.get(i - 1).charAt(sequences.get(i - 1).length() - 1) != ' ' && sequenceAsString.charAt(0) != ' ') {
-                builder.append(' ');
-            }
-            builder.append(sequenceAsString);
-        }
-
-        return builder.toString();
-
-    }
-
-
-    @Override
-    @JsonIgnore
-    public String getText() {
-
-        StringBuilder sb = new StringBuilder();
-
-        TextPositionSequence previous = null;
-        for (TextPositionSequence word : sequences) {
-            if (previous != null) {
-                if (Math.abs(previous.getMaxYDirAdj() - word.getMaxYDirAdj()) > word.getTextHeight()) {
-                    sb.append('\n');
-                } else {
-                    sb.append(' ');
-                }
-            }
-            sb.append(word.toString());
-            previous = word;
-        }
-
-        return TextNormalizationUtilities.removeHyphenLineBreaks(sb.toString());
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/TextPositionSequence.java

@@ -1,302 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import java.awt.geom.AffineTransform;
-import java.awt.geom.Point2D;
-import java.util.ArrayList;
-import java.util.List;
-import java.util.Locale;
-import java.util.stream.Collectors;
-
-import org.apache.pdfbox.text.TextPosition;
-
-import com.fasterxml.jackson.annotation.JsonIgnore;
-import com.fasterxml.jackson.annotation.JsonIgnoreProperties;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Point;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.NoArgsConstructor;
-import lombok.SneakyThrows;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@Data
-@Builder
-@NoArgsConstructor
-@AllArgsConstructor
-@JsonIgnoreProperties({"empty"})
-public class TextPositionSequence implements CharSequence {
-
-    public static final int HEIGHT_PADDING = 2;
-    private int page;
-    private List<RedTextPosition> textPositions = new ArrayList<>();
-
-    private TextDirection dir;
-    private int rotation;
-    private float pageHeight;
-    private float pageWidth;
-
-
-    public TextPositionSequence(int page) {
-
-        this.page = page;
-    }
-
-
-    public TextPositionSequence(List<TextPosition> textPositions, int page) {
-
-        this.textPositions = textPositions.stream().map(RedTextPosition::fromTextPosition).collect(Collectors.toList());
-        this.page = page;
-        this.dir = TextDirection.fromDegrees(textPositions.get(0).getDir());
-        this.rotation = textPositions.get(0).getRotation();
-        this.pageHeight = textPositions.get(0).getPageHeight();
-        this.pageWidth = textPositions.get(0).getPageWidth();
-    }
-
-
-    @Override
-    public int length() {
-
-        return textPositions.size();
-    }
-
-
-    @Override
-    public char charAt(int index) {
-
-        RedTextPosition textPosition = textPositionAt(index);
-        String text = textPosition.getUnicode();
-        return text.charAt(0);
-    }
-
-
-    public char charAt(int index, boolean caseInSensitive) {
-
-        RedTextPosition textPosition = textPositionAt(index);
-        String text = textPosition.getUnicode();
-        return caseInSensitive ? text.toLowerCase(Locale.ROOT).charAt(0) : text.charAt(0);
-    }
-
-
-    @Override
-    public TextPositionSequence subSequence(int start, int end) {
-
-        var textPositionSequence = new TextPositionSequence();
-        textPositionSequence.textPositions = textPositions.subList(start, end);
-        textPositionSequence.page = page;
-        textPositionSequence.dir = dir;
-        textPositionSequence.rotation = rotation;
-        textPositionSequence.pageHeight = pageHeight;
-        textPositionSequence.pageWidth = pageWidth;
-
-        return textPositionSequence;
-    }
-
-
-    @Override
-    public String toString() {
-
-        StringBuilder builder = new StringBuilder(length());
-        for (int i = 0; i < length(); i++) {
-            builder.append(charAt(i));
-        }
-        return builder.toString();
-    }
-
-
-    public RedTextPosition textPositionAt(int index) {
-
-        return textPositions.get(index);
-    }
-
-
-    public void add(TextPositionSequence textPositionSequence, RedTextPosition textPosition) {
-
-        this.textPositions.add(textPosition);
-        this.page = textPositionSequence.getPage();
-        this.dir = textPositionSequence.getDir();
-        this.rotation = textPositionSequence.getRotation();
-        this.pageHeight = textPositionSequence.getPageHeight();
-        this.pageWidth = textPositionSequence.getPageWidth();
-    }
-
-
-    public void add(TextPosition textPosition) {
-
-        this.textPositions.add(RedTextPosition.fromTextPosition(textPosition));
-
-        this.dir = TextDirection.fromDegrees(textPositions.get(0).getDir());
-        this.rotation = textPositions.get(0).getRotation();
-        this.pageHeight = textPositions.get(0).getPageHeight();
-        this.pageWidth = textPositions.get(0).getPageWidth();
-
-    }
-
-
-    /**
-     * This value is adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
-     * This method ignores the page rotation but takes the text rotation and adjusts the coordinates to awt.
-     *
-     * @return the text direction adjusted minX value
-     */
-    @JsonIgnore
-    public float getMinXDirAdj() {
-
-        return textPositions.get(0).getXDirAdj();
-
-    }
-
-
-    /**
-     * This value is adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
-     * This method ignores the page rotation but takes the text rotation and adjusts the coordinates to awt.
-     *
-     * @return the text direction adjusted maxX value
-     */
-    @JsonIgnore
-    public float getMaxXDirAdj() {
-
-        return textPositions.get(textPositions.size() - 1).getXDirAdj() + textPositions.get(textPositions.size() - 1).getWidthDirAdj() + HEIGHT_PADDING;
-
-    }
-
-
-    /**
-     * This value is adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
-     * This method ignores the page rotation but takes the text rotation and adjusts the coordinates to awt.
-     *
-     * @return the text direction adjusted minY value. The upper border of the bounding box of the word.
-     */
-    @JsonIgnore
-    public float getMinYDirAdj() {
-
-        return textPositions.get(0).getYDirAdj() - getTextHeight();
-
-    }
-
-
-    /**
-     * This value is adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
-     * This method ignores the page rotation but takes the text rotation and adjusts the coordinates to awt.
-     *
-     * @return the text direction adjusted maxY value. The lower border of the bounding box of the word.
-     */
-    @JsonIgnore
-    public float getMaxYDirAdj() {
-
-        return textPositions.get(0).getYDirAdj();
-
-    }
-
-
-    @JsonIgnore
-    public float getTextHeight() {
-
-        return textPositions.get(0).getHeightDir() + HEIGHT_PADDING;
-    }
-
-
-    @JsonIgnore
-    public float getHeight() {
-
-        return getMaxYDirAdj() - getMinYDirAdj();
-    }
-
-
-    @JsonIgnore
-    public float getWidth() {
-
-        return getMaxXDirAdj() - getMinXDirAdj();
-    }
-
-
-    @JsonIgnore
-    public String getFont() {
-
-        return textPositions.get(0).getFontName().toLowerCase(Locale.ROOT).replaceAll(",bold", "").replaceAll(",italic", "");
-    }
-
-
-    @JsonIgnore
-    public String getFontStyle() {
-
-        String lowercaseFontName = textPositions.get(0).getFontName().toLowerCase(Locale.ROOT);
-
-        if (lowercaseFontName.contains("bold") && lowercaseFontName.contains("italic")) {
-            return "bold, italic";
-        } else if (lowercaseFontName.contains("bold")) {
-            return "bold";
-        } else if (lowercaseFontName.contains("italic")) {
-            return "italic";
-        } else {
-            return "standard";
-        }
-
-    }
-
-
-    @JsonIgnore
-    public float getFontSize() {
-
-        return textPositions.get(0).getFontSizeInPt();
-    }
-
-
-    @JsonIgnore
-    public float getSpaceWidth() {
-
-        return textPositions.get(0).getWidthOfSpace();
-    }
-
-
-    /**
-     * This returns the bounding box of the word in Pdf Coordinate System where {0,0} rotated with the page rotation.
-     * 0 -> LowerLeft
-     * 90 -> UpperLeft
-     * 180 -> UpperRight
-     * 270 -> LowerRight
-     *
-     * @return bounding box of the word in Pdf Coordinate System
-     */
-    @JsonIgnore
-    @SneakyThrows
-    public Rectangle getRectangle() {
-
-        log.debug("Page: '{}', Word: '{}', Rotation: '{}', textRotation {}", page, this, rotation, dir);
-
-        float textHeight = getTextHeight();
-
-        RedTextPosition firstTextPos = textPositions.get(0);
-        RedTextPosition lastTextPos = textPositions.get(textPositions.size() - 1);
-
-        Point2D bottomLeft = new Point2D.Double(firstTextPos.getXDirAdj(), firstTextPos.getYDirAdj() - HEIGHT_PADDING);
-        Point2D topRight = new Point2D.Double(lastTextPos.getXDirAdj() + lastTextPos.getWidthDirAdj(), lastTextPos.getYDirAdj() + textHeight + HEIGHT_PADDING);
-
-        AffineTransform transform = new AffineTransform();
-        if (dir == TextDirection.ZERO || dir == TextDirection.HALF_CIRCLE) {
-            transform.rotate(dir.getRadians(), pageWidth / 2f, pageHeight / 2f);
-            transform.translate(0f, pageHeight + textHeight);
-            transform.scale(1., -1.);
-        } else if (dir == TextDirection.QUARTER_CIRCLE) {
-            transform.rotate(dir.getRadians(), pageWidth / 2f, pageWidth / 2f);
-            transform.translate(0f, pageWidth + textHeight);
-            transform.scale(1., -1.);
-        } else {
-            transform.rotate(dir.getRadians(), pageHeight / 2f, pageHeight / 2f);
-            transform.translate(0f, pageWidth + textHeight);
-            transform.scale(1., -1.);
-        }
-
-        bottomLeft = transform.transform(bottomLeft, null);
-        topRight = transform.transform(topRight, null);
-
-        return new Rectangle( //
-                new Point((float) bottomLeft.getX(), (float) bottomLeft.getY()),
-                (float) (topRight.getX() - bottomLeft.getX()),
-                (float) (topRight.getY() - bottomLeft.getY()),
-                page);
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/model/text/UnclassifiedText.java

@@ -1,14 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text;
-
-import java.util.List;
-
-import lombok.AllArgsConstructor;
-import lombok.Data;
-
-@Data
-@AllArgsConstructor
-public class UnclassifiedText {
-
-    private List<TextPageBlock> textBlocks;
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/parsing/LegacyPDFStreamEngine.java

@@ -1,384 +0,0 @@
-/*
- * Licensed to the Apache Software Foundation (ASF) under one or more
- * contributor license agreements.  See the NOTICE file distributed with
- * this work for additional information regarding copyright ownership.
- * The ASF licenses this file to You under the Apache License, Version 2.0
- * (the "License"); you may not use this file except in compliance with
- * the License.  You may obtain a copy of the License at
- *
- *      http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.parsing;
-
-import java.io.IOException;
-import java.io.InputStream;
-import java.util.Map;
-import java.util.WeakHashMap;
-
-import org.apache.commons.logging.Log;
-import org.apache.commons.logging.LogFactory;
-import org.apache.fontbox.ttf.TrueTypeFont;
-import org.apache.fontbox.util.BoundingBox;
-import org.apache.pdfbox.contentstream.PDFStreamEngine;
-import org.apache.pdfbox.contentstream.operator.DrawObject;
-import org.apache.pdfbox.contentstream.operator.state.Concatenate;
-import org.apache.pdfbox.contentstream.operator.state.Restore;
-import org.apache.pdfbox.contentstream.operator.state.Save;
-import org.apache.pdfbox.contentstream.operator.state.SetGraphicsStateParameters;
-import org.apache.pdfbox.contentstream.operator.state.SetMatrix;
-import org.apache.pdfbox.contentstream.operator.text.BeginText;
-import org.apache.pdfbox.contentstream.operator.text.EndText;
-import org.apache.pdfbox.contentstream.operator.text.MoveText;
-import org.apache.pdfbox.contentstream.operator.text.MoveTextSetLeading;
-import org.apache.pdfbox.contentstream.operator.text.NextLine;
-import org.apache.pdfbox.contentstream.operator.text.SetCharSpacing;
-import org.apache.pdfbox.contentstream.operator.text.SetFontAndSize;
-import org.apache.pdfbox.contentstream.operator.text.SetTextHorizontalScaling;
-import org.apache.pdfbox.contentstream.operator.text.SetTextLeading;
-import org.apache.pdfbox.contentstream.operator.text.SetTextRenderingMode;
-import org.apache.pdfbox.contentstream.operator.text.SetTextRise;
-import org.apache.pdfbox.contentstream.operator.text.SetWordSpacing;
-import org.apache.pdfbox.contentstream.operator.text.ShowText;
-import org.apache.pdfbox.contentstream.operator.text.ShowTextAdjusted;
-import org.apache.pdfbox.contentstream.operator.text.ShowTextLine;
-import org.apache.pdfbox.contentstream.operator.text.ShowTextLineAndSpace;
-import org.apache.pdfbox.cos.COSDictionary;
-import org.apache.pdfbox.pdmodel.PDPage;
-import org.apache.pdfbox.pdmodel.common.PDRectangle;
-import org.apache.pdfbox.pdmodel.font.PDCIDFont;
-import org.apache.pdfbox.pdmodel.font.PDCIDFontType2;
-import org.apache.pdfbox.pdmodel.font.PDFont;
-import org.apache.pdfbox.pdmodel.font.PDFontDescriptor;
-import org.apache.pdfbox.pdmodel.font.PDSimpleFont;
-import org.apache.pdfbox.pdmodel.font.PDTrueTypeFont;
-import org.apache.pdfbox.pdmodel.font.PDType0Font;
-import org.apache.pdfbox.pdmodel.font.PDType3Font;
-import org.apache.pdfbox.pdmodel.font.encoding.GlyphList;
-import org.apache.pdfbox.pdmodel.graphics.state.PDGraphicsState;
-import org.apache.pdfbox.text.TextPosition;
-import org.apache.pdfbox.util.Matrix;
-import org.apache.pdfbox.util.Vector;
-
-/**
- * LEGACY text calculations which are known to be incorrect but are depended on by PDFTextStripper.
- * <p>
- * This class exists only so that we don't break the code of users who have their own subclasses of
- * PDFTextStripper. It replaces the mostly empty implementation of showGlyph() in PDFStreamEngine
- * with a heuristic implementation which is backwards compatible.
- * <p>
- * DO NOT USE THIS CODE UNLESS YOU ARE WORKING WITH PDFTextStripper.
- * THIS CODE IS DELIBERATELY INCORRECT, USE PDFStreamEngine INSTEAD.
- */
-@SuppressWarnings({"PMD", "checkstyle:all"})
-class LegacyPDFStreamEngine extends PDFStreamEngine {
-
-    private static final Log LOG = LogFactory.getLog(LegacyPDFStreamEngine.class);
-
-    private int pageRotation;
-    private PDRectangle pageSize;
-    private Matrix translateMatrix;
-    private final GlyphList glyphList;
-    private final Map<COSDictionary, Float> fontHeightMap = new WeakHashMap<COSDictionary, Float>();
-
-
-    /**
-     * Constructor.
-     */
-    LegacyPDFStreamEngine() throws IOException {
-
-        addOperator(new BeginText());
-        addOperator(new Concatenate());
-        addOperator(new DrawObject()); // special text version
-        addOperator(new EndText());
-        addOperator(new SetGraphicsStateParameters());
-        addOperator(new Save());
-        addOperator(new Restore());
-        addOperator(new NextLine());
-        addOperator(new SetCharSpacing());
-        addOperator(new MoveText());
-        addOperator(new MoveTextSetLeading());
-        addOperator(new SetFontAndSize());
-        addOperator(new ShowText());
-        addOperator(new ShowTextAdjusted());
-        addOperator(new SetTextLeading());
-        addOperator(new SetMatrix());
-        addOperator(new SetTextRenderingMode());
-        addOperator(new SetTextRise());
-        addOperator(new SetWordSpacing());
-        addOperator(new SetTextHorizontalScaling());
-        addOperator(new ShowTextLine());
-        addOperator(new ShowTextLineAndSpace());
-
-        // load additional glyph list for Unicode mapping
-        String path = "/org/apache/pdfbox/resources/glyphlist/additional.txt";
-        InputStream input = GlyphList.class.getResourceAsStream(path);
-        glyphList = new GlyphList(GlyphList.getAdobeGlyphList(), input);
-    }
-
-
-    /**
-     * This will initialize and process the contents of the stream.
-     *
-     * @param page the page to process
-     * @throws java.io.IOException if there is an error accessing the stream.
-     */
-    @Override
-    public void processPage(PDPage page) throws IOException {
-
-        this.pageRotation = page.getRotation();
-        this.pageSize = page.getCropBox();
-
-        if (pageSize.getLowerLeftX() == 0 && pageSize.getLowerLeftY() == 0) {
-            translateMatrix = null;
-        } else {
-            // translation matrix for cropbox
-            translateMatrix = Matrix.getTranslateInstance(-pageSize.getLowerLeftX(), -pageSize.getLowerLeftY());
-        }
-        super.processPage(page);
-    }
-
-
-    /**
-     * Called when a glyph is to be processed. The heuristic calculations here were originally
-     * written by Ben Litchfield for PDFStreamEngine.
-     */
-    @Override
-    protected void showGlyph(Matrix textRenderingMatrix, PDFont font, int code, String unicode, Vector displacement) throws IOException {
-        //
-        // legacy calculations which were previously in PDFStreamEngine
-        //
-        //  DO NOT USE THIS CODE UNLESS YOU ARE WORKING WITH PDFTextStripper.
-        //  THIS CODE IS DELIBERATELY INCORRECT
-        //
-
-        PDGraphicsState state = getGraphicsState();
-        Matrix ctm = state.getCurrentTransformationMatrix();
-        float fontSize = state.getTextState().getFontSize();
-        float horizontalScaling = state.getTextState().getHorizontalScaling() / 100f;
-        Matrix textMatrix = getTextMatrix();
-
-        float displacementX = displacement.getX();
-        // the sorting algorithm is based on the width of the character. As the displacement
-        // for vertical characters doesn't provide any suitable value for it, we have to 
-        // calculate our own
-        if (font.isVertical()) {
-            displacementX = font.getWidth(code) / 1000;
-            // there may be an additional scaling factor for true type fonts
-            TrueTypeFont ttf = null;
-            if (font instanceof PDTrueTypeFont) {
-                ttf = ((PDTrueTypeFont) font).getTrueTypeFont();
-            } else if (font instanceof PDType0Font) {
-                PDCIDFont cidFont = ((PDType0Font) font).getDescendantFont();
-                if (cidFont instanceof PDCIDFontType2) {
-                    ttf = ((PDCIDFontType2) cidFont).getTrueTypeFont();
-                }
-            }
-            if (ttf != null && ttf.getUnitsPerEm() != 1000) {
-                displacementX *= 1000f / ttf.getUnitsPerEm();
-            }
-        }
-
-        //
-        // legacy calculations which were previously in PDFStreamEngine
-        //
-        //  DO NOT USE THIS CODE UNLESS YOU ARE WORKING WITH PDFTextStripper.
-        //  THIS CODE IS DELIBERATELY INCORRECT
-        //
-
-        // (modified) combined displacement, this is calculated *without* taking the character
-        // spacing and word spacing into account, due to legacy code in TextStripper
-        float tx = displacementX * fontSize * horizontalScaling;
-        float ty = displacement.getY() * fontSize;
-
-        // (modified) combined displacement matrix
-        Matrix td = Matrix.getTranslateInstance(tx, ty);
-
-        // (modified) text rendering matrix
-        Matrix nextTextRenderingMatrix = td.multiply(textMatrix).multiply(ctm); // text space -> device space
-        float nextX = nextTextRenderingMatrix.getTranslateX();
-        float nextY = nextTextRenderingMatrix.getTranslateY();
-
-        // (modified) width and height calculations
-        float dxDisplay = nextX - textRenderingMatrix.getTranslateX();
-        Float fontHeight = fontHeightMap.get(font.getCOSObject());
-        if (fontHeight == null) {
-            fontHeight = computeFontHeight(font);
-            fontHeightMap.put(font.getCOSObject(), fontHeight);
-        }
-        float dyDisplay = fontHeight * textRenderingMatrix.getScalingFactorY();
-
-        //
-        // start of the original method
-        //
-
-        // Note on variable names. There are three different units being used in this code.
-        // Character sizes are given in glyph units, text locations are initially given in text
-        // units, and we want to save the data in display units. The variable names should end with
-        // Text or Disp to represent if the values are in text or disp units (no glyph units are
-        // saved).
-
-        float glyphSpaceToTextSpaceFactor = 1 / 1000f;
-        if (font instanceof PDType3Font) {
-            glyphSpaceToTextSpaceFactor = font.getFontMatrix().getScaleX();
-        }
-
-        float spaceWidthText = 0;
-        try {
-            // to avoid crash as described in PDFBOX-614, see what the space displacement should be
-            spaceWidthText = font.getSpaceWidth() * glyphSpaceToTextSpaceFactor;
-        } catch (Throwable exception) {
-            LOG.warn(exception, exception);
-        }
-
-        if (spaceWidthText == 0) {
-            spaceWidthText = font.getAverageFontWidth() * glyphSpaceToTextSpaceFactor;
-            // the average space width appears to be higher than necessary so make it smaller
-            spaceWidthText *= .80f;
-        }
-        if (spaceWidthText == 0) {
-            spaceWidthText = 1.0f; // if could not find font, use a generic value
-        }
-
-        // the space width has to be transformed into display units
-        float spaceWidthDisplay = spaceWidthText * textRenderingMatrix.getScalingFactorX();
-
-        // use our additional glyph list for Unicode mapping
-        String unicodeMapping = font.toUnicode(code, glyphList);
-
-        // when there is no Unicode mapping available, Acrobat simply coerces the character code
-        // into Unicode, so we do the same. Subclasses of PDFStreamEngine don't necessarily want
-        // this, which is why we leave it until this point in PDFTextStreamEngine.
-        if (unicodeMapping == null) {
-            if (font instanceof PDSimpleFont) {
-                char c = (char) code;
-                unicodeMapping = new String(new char[]{c});
-            } else {
-                // Acrobat doesn't seem to coerce composite font's character codes, instead it
-                // skips them. See the "allah2.pdf" TestTextStripper file.
-                return;
-            }
-        }
-
-        // adjust for cropbox if needed
-        Matrix translatedTextRenderingMatrix;
-        if (translateMatrix == null) {
-            translatedTextRenderingMatrix = textRenderingMatrix;
-        } else {
-            translatedTextRenderingMatrix = Matrix.concatenate(translateMatrix, textRenderingMatrix);
-            nextX -= pageSize.getLowerLeftX();
-            nextY -= pageSize.getLowerLeftY();
-        }
-
-        // This is a hack for unicode letter with 2 chars e.g. RA see unicodeProblem.pdf
-        if (unicodeMapping.length() == 2) {
-            processTextPosition(new TextPosition(pageRotation,
-                    pageSize.getWidth(),
-                    pageSize.getHeight(),
-                    translatedTextRenderingMatrix,
-                    nextX,
-                    nextY,
-                    Math.abs(dyDisplay),
-                    dxDisplay,
-                    Math.abs(spaceWidthDisplay),
-                    Character.toString(unicodeMapping.charAt(0)),
-                    new int[]{code},
-                    font,
-                    fontSize,
-                    (int) (fontSize * textMatrix.getScalingFactorX())));
-            processTextPosition(new TextPosition(pageRotation,
-                    pageSize.getWidth(),
-                    pageSize.getHeight(),
-                    translatedTextRenderingMatrix,
-                    nextX,
-                    nextY,
-                    Math.abs(dyDisplay),
-                    dxDisplay,
-                    Math.abs(spaceWidthDisplay),
-                    Character.toString(unicodeMapping.charAt(1)),
-                    new int[]{code},
-                    font,
-                    fontSize,
-                    (int) (fontSize * textMatrix.getScalingFactorX())));
-        } else {
-
-            processTextPosition(new TextPosition(pageRotation,
-                    pageSize.getWidth(),
-                    pageSize.getHeight(),
-                    translatedTextRenderingMatrix,
-                    nextX,
-                    nextY,
-                    Math.abs(dyDisplay),
-                    dxDisplay,
-                    Math.abs(spaceWidthDisplay),
-                    unicodeMapping,
-                    new int[]{code},
-                    font,
-                    fontSize,
-                    (int) (fontSize * textMatrix.getScalingFactorX())));
-        }
-    }
-
-
-    /**
-     * Compute the font height. Override this if you want to use own calculations.
-     *
-     * @param font the font.
-     * @return the font height.
-     * @throws IOException if there is an error while getting the font bounding box.
-     */
-    protected float computeFontHeight(PDFont font) throws IOException {
-
-        BoundingBox bbox = font.getBoundingBox();
-        if (bbox.getLowerLeftY() < Short.MIN_VALUE) {
-            // PDFBOX-2158 and PDFBOX-3130
-            // files by Salmat eSolutions / ClibPDF Library
-            bbox.setLowerLeftY(-(bbox.getLowerLeftY() + 65536));
-        }
-        // 1/2 the bbox is used as the height todo: why?
-        float glyphHeight = bbox.getHeight() / 2;
-
-        // sometimes the bbox has very high values, but CapHeight is OK
-        PDFontDescriptor fontDescriptor = font.getFontDescriptor();
-        if (fontDescriptor != null) {
-            float capHeight = fontDescriptor.getCapHeight();
-            if (Float.compare(capHeight, 0) != 0 && (capHeight < glyphHeight || Float.compare(glyphHeight, 0) == 0)) {
-                glyphHeight = capHeight;
-            }
-            // PDFBOX-3464, PDFBOX-4480, PDFBOX-4553:
-            // sometimes even CapHeight has very high value, but Ascent and Descent are ok
-            float ascent = fontDescriptor.getAscent();
-            float descent = fontDescriptor.getDescent();
-            if (capHeight > ascent && ascent > 0 && descent < 0 && ((ascent - descent) / 2 < glyphHeight || Float.compare(glyphHeight, 0) == 0)) {
-                glyphHeight = (ascent - descent) / 2;
-            }
-        }
-
-        // transformPoint from glyph space -> text space
-        float height;
-        if (font instanceof PDType3Font) {
-            height = font.getFontMatrix().transformPoint(0, glyphHeight).y;
-        } else {
-            height = glyphHeight / 1000;
-        }
-
-        return height;
-    }
-
-
-    /**
-     * A method provided as an event interface to allow a subclass to perform some specific
-     * functionality when text needs to be processed.
-     *
-     * @param text The text to be processed.
-     */
-    protected void processTextPosition(TextPosition text) {
-        // subclasses can override to provide specific functionality
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/parsing/PDFLinesTextStripper.java

@@ -1,335 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.parsing;
-
-import java.awt.geom.Point2D;
-import java.io.IOException;
-import java.util.ArrayList;
-import java.util.Comparator;
-import java.util.List;
-
-import org.apache.pdfbox.contentstream.operator.Operator;
-import org.apache.pdfbox.contentstream.operator.OperatorName;
-import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingColor;
-import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingColorN;
-import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingColorSpace;
-import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingDeviceCMYKColor;
-import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingDeviceGrayColor;
-import org.apache.pdfbox.contentstream.operator.color.SetNonStrokingDeviceRGBColor;
-import org.apache.pdfbox.contentstream.operator.color.SetStrokingColor;
-import org.apache.pdfbox.contentstream.operator.color.SetStrokingColorN;
-import org.apache.pdfbox.contentstream.operator.color.SetStrokingColorSpace;
-import org.apache.pdfbox.contentstream.operator.color.SetStrokingDeviceCMYKColor;
-import org.apache.pdfbox.contentstream.operator.color.SetStrokingDeviceGrayColor;
-import org.apache.pdfbox.contentstream.operator.color.SetStrokingDeviceRGBColor;
-import org.apache.pdfbox.contentstream.operator.state.SetFlatness;
-import org.apache.pdfbox.contentstream.operator.state.SetLineCapStyle;
-import org.apache.pdfbox.contentstream.operator.state.SetLineDashPattern;
-import org.apache.pdfbox.contentstream.operator.state.SetLineJoinStyle;
-import org.apache.pdfbox.contentstream.operator.state.SetLineMiterLimit;
-import org.apache.pdfbox.contentstream.operator.state.SetLineWidth;
-import org.apache.pdfbox.contentstream.operator.state.SetRenderingIntent;
-import org.apache.pdfbox.contentstream.operator.text.SetFontAndSize;
-import org.apache.pdfbox.cos.COSBase;
-import org.apache.pdfbox.cos.COSNumber;
-import org.apache.pdfbox.pdmodel.PDDocument;
-import org.apache.pdfbox.pdmodel.PDPage;
-import org.apache.pdfbox.text.TextPosition;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.RedTextPosition;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-
-import lombok.Getter;
-import lombok.Setter;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-public class PDFLinesTextStripper extends PDFTextStripper {
-
-    @Getter
-    private final List<TextPositionSequence> textPositionSequences = new ArrayList<>();
-    @Getter
-    private final List<Ruling> rulings = new ArrayList<>();
-    private final List<Ruling> graphicsPath = new ArrayList<>();
-    @Setter
-    protected PDPage pdpage;
-    @Getter
-    private int minCharWidth;
-    @Getter
-    private int maxCharWidth;
-    @Getter
-    private int minCharHeight;
-    @Getter
-    private int maxCharHeight;
-
-    private float path_x;
-    private float path_y;
-
-    @Setter
-    private int pageNumber;
-
-
-    public PDFLinesTextStripper() throws IOException {
-
-        super();
-        this.addOperator(new SetStrokingColorSpace());
-        this.addOperator(new SetNonStrokingColorSpace());
-        this.addOperator(new SetLineDashPattern());
-        this.addOperator(new SetStrokingDeviceGrayColor());
-        this.addOperator(new SetNonStrokingDeviceGrayColor());
-        this.addOperator(new SetFlatness());
-        this.addOperator(new SetLineJoinStyle());
-        this.addOperator(new SetLineCapStyle());
-        this.addOperator(new SetStrokingDeviceCMYKColor());
-        this.addOperator(new SetNonStrokingDeviceCMYKColor());
-        this.addOperator(new SetLineMiterLimit());
-        this.addOperator(new SetStrokingDeviceRGBColor());
-        this.addOperator(new SetNonStrokingDeviceRGBColor());
-        this.addOperator(new SetRenderingIntent());
-        this.addOperator(new SetStrokingColor());
-        this.addOperator(new SetNonStrokingColor());
-        this.addOperator(new SetStrokingColorN());
-        this.addOperator(new SetNonStrokingColorN());
-        this.addOperator(new SetFontAndSize());
-        this.addOperator(new SetLineWidth());
-    }
-
-
-    @Override
-    protected void processOperator(Operator operator, List<COSBase> arguments) throws IOException {
-
-        String operation = operator.getName();
-
-        //move
-        switch (operation) {
-            case OperatorName.MOVE_TO:
-                if (arguments.size() == 2) {
-                    Point2D.Float pos = transformPosition(floatValue(arguments.get(0)), floatValue(arguments.get(1)));
-                    path_x = (float) pos.getX();
-                    path_y = (float) pos.getY();
-                }
-                break;
-
-            //line
-            case OperatorName.LINE_TO:
-                if (arguments.size() == 2) {
-                    Point2D.Float pos = transformPosition(floatValue(arguments.get(0)), floatValue(arguments.get(1)));
-
-                    // The direction of vertical lines must always be from bottom to top for the table extraction algorithm.
-                    if (pos.getY() > path_y) {
-                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, path_y), new Point2D.Float((float) pos.getX(), (float) pos.getY())));
-                    } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, (float) pos.getY()), new Point2D.Float((float) pos.getX(), path_y)));
-                    }
-
-                    path_x = (float) pos.getX();
-                    path_y = (float) pos.getY();
-                }
-                break;
-
-            //rectangle
-            case OperatorName.APPEND_RECT:
-
-                if (arguments.size() == 4) {
-                    float x = floatValue(arguments.get(0));
-                    float y = floatValue(arguments.get(1));
-                    float width = floatValue(arguments.get(2));
-                    float height = floatValue(arguments.get(3));
-
-                    Point2D p1 = transformPosition(x, y);
-                    Point2D p2 = transformPosition(x + width, y + height);
-
-                    // Horizontal lines
-                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p2.getX(), (float) p1.getY())));
-                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p2.getX(), (float) p2.getY())));
-
-                    // Vertical lines, direction must always be from bottom to top for the table extraction algorithm.
-                    if (p2.getY() > p1.getY()) {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p1.getY()), new Point2D.Float((float) p2.getX(), (float) p2.getY())));
-                    } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p2.getY()), new Point2D.Float((float) p2.getX(), (float) p1.getY())));
-                    }
-                    if (p2.getY() > p1.getY()) {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p1.getX(), (float) p2.getY())));
-                    } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p1.getX(), (float) p1.getY())));
-                    }
-                }
-                break;
-
-            //fill
-            case OperatorName.FILL_NON_ZERO:
-            case OperatorName.LEGACY_FILL_NON_ZERO:
-            case OperatorName.FILL_EVEN_ODD:
-                addVisibleRulings(graphicsPath, false);
-                graphicsPath.clear();
-                break;
-
-            //stroke
-            case OperatorName.STROKE_PATH:
-                addVisibleRulings(graphicsPath, true);
-                graphicsPath.clear();
-                break;
-
-            //cancel path
-            case OperatorName.ENDPATH:
-                graphicsPath.clear();
-                break;
-
-        }
-
-        super.processOperator(operator, arguments);
-    }
-
-
-    private float floatValue(COSBase value) {
-
-        if (value instanceof COSNumber) {
-            return ((COSNumber) value).floatValue();
-        } else {
-            return 0;
-        }
-    }
-
-
-    private Point2D.Float transformPosition(float x, float y) {
-
-        return super.transformedPoint(x, y);
-    }
-
-
-    private void addVisibleRulings(List<Ruling> path, boolean stroke) throws IOException {
-
-        try {
-            if (stroke && !getGraphicsState().getStrokingColor().isPattern() && getGraphicsState().getStrokingColor()
-                    .toRGB() == 0 || !stroke && !getGraphicsState().getNonStrokingColor().isPattern() && getGraphicsState().getNonStrokingColor().toRGB() == 0) {
-                rulings.addAll(path);
-            }
-        } catch (UnsupportedOperationException e) {
-            log.debug("UnsupportedOperationException: " + getGraphicsState().getStrokingColor().getColorSpace().getName() + " or " + getGraphicsState().getNonStrokingColor()
-                    .getColorSpace()
-                    .getName() + " does not support toRGB");
-        }
-    }
-
-
-    @Override
-    public void writeString(String text, List<TextPosition> textPositions) throws IOException {
-
-        int startIndex = 0;
-        RedTextPosition previous = null;
-
-        textPositions.sort(Comparator.comparing(TextPosition::getXDirAdj));
-
-        for (int i = 0; i <= textPositions.size() - 1; i++) {
-
-            if (!textPositionSequences.isEmpty()) {
-                previous = textPositionSequences.get(textPositionSequences.size() - 1)
-                        .getTextPositions()
-                        .get(textPositionSequences.get(textPositionSequences.size() - 1).getTextPositions().size() - 1);
-            }
-
-            int charWidth = (int) textPositions.get(i).getWidthDirAdj();
-            if (charWidth < minCharWidth) {
-                minCharWidth = charWidth;
-            }
-            if (charWidth > maxCharWidth) {
-                maxCharWidth = charWidth;
-            }
-
-            int charHeight = (int) textPositions.get(i).getHeightDir();
-            if (charHeight < minCharHeight) {
-                minCharHeight = charHeight;
-            }
-            if (charWidth > maxCharHeight) {
-                maxCharHeight = charHeight;
-            }
-
-            if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0") || textPositions.get(i).getUnicode().equals("\t"))) {
-                startIndex++;
-                continue;
-            }
-
-            // Strange but sometimes this is happening, for example: Metolachlor2.pdf
-            if (i > 0 && textPositions.get(i).getXDirAdj() < textPositions.get(i - 1).getXDirAdj()) {
-                List<TextPosition> sublist = textPositions.subList(startIndex, i);
-                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
-                        .getUnicode()
-                        .equals("\t")))) {
-                    textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
-                }
-                startIndex = i;
-            }
-
-            if (textPositions.get(i).getRotation() == 0 && i > 0 && textPositions.get(i).getX() > textPositions.get(i - 1).getEndX() + 1) {
-                List<TextPosition> sublist = textPositions.subList(startIndex, i);
-                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
-                        .getUnicode()
-                        .equals("\t")))) {
-                    textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
-                }
-                startIndex = i;
-            }
-
-            if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0") || textPositions.get(i)
-                    .getUnicode()
-                    .equals("\t")) && i <= textPositions.size() - 2) {
-                List<TextPosition> sublist = textPositions.subList(startIndex, i);
-                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
-                        .getUnicode()
-                        .equals("\t")))) {
-
-                    // Remove false sequence ends (whitespaces)
-                    if (previous != null && sublist.get(0).getYDirAdj() == previous.getYDirAdj() && sublist.get(0)
-                            .getXDirAdj() - (previous.getXDirAdj() + previous.getWidthDirAdj()) < 0.01) {
-                        for (TextPosition textPosition : sublist) {
-                            textPositionSequences.get(textPositionSequences.size() - 1).add(textPosition);
-                        }
-                    } else {
-                        textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
-                    }
-                }
-                startIndex = i + 1;
-            }
-        }
-
-        List<TextPosition> sublist = textPositions.subList(startIndex, textPositions.size());
-        if (!sublist.isEmpty() && (sublist.get(sublist.size() - 1).getUnicode().equals(" ") || sublist.get(sublist.size() - 1)
-                .getUnicode()
-                .equals("\u00A0") || sublist.get(sublist.size() - 1).getUnicode().equals("\t"))) {
-            sublist = sublist.subList(0, sublist.size() - 1);
-        }
-
-        if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
-                .getUnicode()
-                .equals("\t")))) {
-            if (previous != null && sublist.get(0).getYDirAdj() == previous.getYDirAdj() && sublist.get(0)
-                    .getXDirAdj() - (previous.getXDirAdj() + previous.getWidthDirAdj()) < 0.01) {
-                for (TextPosition t : sublist) {
-                    textPositionSequences.get(textPositionSequences.size() - 1).add(t);
-                }
-            } else {
-                textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
-            }
-        }
-        super.writeString(text);
-    }
-
-
-    @Override
-    public String getText(PDDocument doc) throws IOException {
-
-        minCharWidth = Integer.MAX_VALUE;
-        maxCharWidth = 0;
-        minCharHeight = Integer.MAX_VALUE;
-        maxCharHeight = 0;
-        textPositionSequences.clear();
-        rulings.clear();
-        graphicsPath.clear();
-        path_x = 0.0f;
-        path_y = 0.0f;
-
-        return super.getText(doc);
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/BlockificationService.java

@@ -1,25 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import static java.util.stream.Collectors.toSet;
-
-import java.util.ArrayList;
-import java.util.Comparator;
-import java.util.Iterator;
-import java.util.List;
-
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.FloatFrequencyCounter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.Orientation;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.RulingTextDirAdjustUtil;
-
-public interface BlockificationService {
-
-    ClassificationPage blockify(List<TextPositionSequence> textPositions, List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines);
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/BodyTextFrameService.java

@@ -1,169 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.util.List;
-
-import org.springframework.beans.factory.annotation.Value;
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Point;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.FloatFrequencyCounter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Cell;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.PositionUtils;
-
-@Service
-public class BodyTextFrameService {
-
-    @Value("${application.type:RedactManager}")
-    private String applicationType;
-
-    private static final float APPROXIMATE_HEADER_LINE_COUNT = 2.9f;
-
-
-    /**
-     * Adjusts and sets the body text frame to a page.
-     * Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
-     * 0 -> LowerLeft
-     * 90 -> UpperLeft
-     * 180 -> UpperRight
-     * 270 -> LowerRight
-     * The aspect ratio of the page is also regarded.
-     *
-     * @param page                   The page
-     * @param bodyTextFrame          frame that contains the main text on portrait pages
-     * @param landscapeBodyTextFrame frame that contains the main text on landscape pages
-     */
-    public void setBodyTextFrameAdjustedToPage(ClassificationPage page, Rectangle bodyTextFrame, Rectangle landscapeBodyTextFrame) {
-
-        Rectangle textFrame = page.isLandscape() ? landscapeBodyTextFrame : bodyTextFrame;
-
-        if (page.getPageWidth() > page.getPageHeight() && page.getRotation() == 270) {
-            textFrame = new Rectangle(new Point(textFrame.getTopLeft().getY(), page.getPageHeight() - textFrame.getTopLeft().getX() - textFrame.getWidth()),
-                    textFrame.getHeight(),
-                    textFrame.getWidth(),
-                    0);
-        } else if (page.getPageWidth() > page.getPageHeight() && page.getRotation() != 0) {
-            textFrame = new Rectangle(new Point(textFrame.getTopLeft().getY(), textFrame.getTopLeft().getX()), textFrame.getHeight(), textFrame.getWidth(), page.getPageNumber());
-        } else if (page.getRotation() == 180) {
-            textFrame = new Rectangle(new Point(textFrame.getTopLeft().getX(), page.getPageHeight() - textFrame.getTopLeft().getY() - textFrame.getHeight()),
-                    textFrame.getWidth(),
-                    textFrame.getHeight(),
-                    0);
-        }
-        page.setBodyTextFrame(textFrame);
-    }
-
-
-    /**
-     * Calculates the frame that contains the main text, text outside the frame will be e.g. headers or footers.
-     * Note: This needs to use Pdf Coordinate System where {0,0} rotated with the page rotation.
-     * 0 -> LowerLeft
-     * 90 -> UpperLeft
-     * 180 -> UpperRight
-     * 270 -> LowerRight
-     * The aspect ratio of the page is also regarded.
-     *
-     * @param pages                   List of all pages
-     * @param documentFontSizeCounter Statistics of the document
-     * @param landscape               Calculate for landscape or portrait
-     * @return Rectangle of the text frame
-     */
-    public Rectangle calculateBodyTextFrame(List<ClassificationPage> pages, FloatFrequencyCounter documentFontSizeCounter, boolean landscape) {
-
-        BodyTextFrameExpansionsRectangle expansionsRectangle = new BodyTextFrameExpansionsRectangle();
-
-        for (ClassificationPage page : pages) {
-
-            if (page.getTextBlocks().isEmpty() || landscape != page.isLandscape()) {
-                continue;
-            }
-
-            for (AbstractPageBlock container : page.getTextBlocks()) {
-
-                if (container instanceof TextPageBlock) {
-                    TextPageBlock textBlock = (TextPageBlock) container;
-                    if (textBlock.getMostPopularWordFont() == null || textBlock.getMostPopularWordStyle() == null) {
-                        continue;
-                    }
-
-                    float approxLineCount = PositionUtils.getApproxLineCount(textBlock);
-                    if (applicationType.equals("DocuMine") && approxLineCount < APPROXIMATE_HEADER_LINE_COUNT && textBlock.getMaxY() >= page.getPageHeight() - (page.getPageHeight() / 10)
-                        || applicationType.equals("RedactManager") && approxLineCount < APPROXIMATE_HEADER_LINE_COUNT){
-                        continue;
-                    }
-
-                    if (documentFontSizeCounter.getMostPopular() != null && textBlock.getMostPopularWordFontSize() >= documentFontSizeCounter.getMostPopular()) {
-
-                        expandRectangle(textBlock, page, expansionsRectangle);
-                    }
-                }
-
-                if (container instanceof TablePageBlock) {
-                    TablePageBlock table = (TablePageBlock) container;
-                    for (List<Cell> row : table.getRows()) {
-                        for (Cell cell : row) {
-
-                            if (cell == null || cell.getTextBlocks() == null) {
-                                continue;
-                            }
-                            for (TextPageBlock textBlock : cell.getTextBlocks()) {
-                                expandRectangle(textBlock, page, expansionsRectangle);
-                            }
-                        }
-                    }
-                }
-            }
-        }
-        return new Rectangle(new Point(expansionsRectangle.minX, expansionsRectangle.minY),
-                expansionsRectangle.maxX - expansionsRectangle.minX,
-                expansionsRectangle.maxY - expansionsRectangle.minY,
-                0);
-    }
-
-
-    private void expandRectangle(TextPageBlock textBlock, ClassificationPage page, BodyTextFrameExpansionsRectangle expansionsRectangle) {
-
-        if (page.getPageWidth() > page.getPageHeight() && page.getRotation() != 0) {
-            if (textBlock.getPdfMinY() < expansionsRectangle.minX) {
-                expansionsRectangle.minX = textBlock.getPdfMinY();
-            }
-            if (textBlock.getPdfMaxY() > expansionsRectangle.maxX) {
-                expansionsRectangle.maxX = textBlock.getPdfMaxY();
-            }
-            if (textBlock.getPdfMinX() < expansionsRectangle.minY) {
-                expansionsRectangle.minY = textBlock.getPdfMinX();
-            }
-            if (textBlock.getPdfMaxX() > expansionsRectangle.maxY) {
-                expansionsRectangle.maxY = textBlock.getPdfMaxX();
-            }
-        } else {
-            if (textBlock.getPdfMinX() < expansionsRectangle.minX) {
-                expansionsRectangle.minX = textBlock.getPdfMinX();
-            }
-            if (textBlock.getPdfMaxX() > expansionsRectangle.maxX) {
-                expansionsRectangle.maxX = textBlock.getPdfMaxX();
-            }
-            if (textBlock.getPdfMinY() < expansionsRectangle.minY) {
-                expansionsRectangle.minY = textBlock.getPdfMinY();
-            }
-            if (textBlock.getPdfMaxY() > expansionsRectangle.maxY) {
-                expansionsRectangle.maxY = textBlock.getPdfMaxY();
-            }
-        }
-    }
-
-
-    private class BodyTextFrameExpansionsRectangle {
-
-        float minX = 10000;
-        float maxX = -100;
-        float minY = 10000;
-        float maxY = -100;
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/ClassificationService.java

@@ -1,7 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
-
-public interface ClassificationService {
-    void classifyDocument(ClassificationDocument document);
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/ColumnDetectionService.java

@@ -1,149 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.awt.geom.Rectangle2D;
-import java.util.Collection;
-import java.util.Comparator;
-import java.util.HashMap;
-import java.util.LinkedList;
-import java.util.List;
-import java.util.Map;
-import java.util.stream.IntStream;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public class ColumnDetectionService {
-
-    private static final double SPLITTABLE_LINE_PERCENTAGE_THRESHOLD = 0.6;
-    private static final int MAX_NUMBER_OF_COLUMNS = 4;
-
-
-    public List<Rectangle2D> detectColumns(List<TextPositionSequence> textPositionSequences, Rectangle2D mainBodyTextFrame) {
-
-        if (textPositionSequences.size() < 2) {
-            return List.of(mainBodyTextFrame);
-        }
-
-        List<List<Rectangle2D>> linesWithGaps = LineDetectionService.findLinesWithGaps(textPositionSequences);
-
-        Map<Integer, List<Integer>> linesWithMatchingGapIndices = new HashMap<>();
-        for (int numberOfColumns = 2; numberOfColumns <= MAX_NUMBER_OF_COLUMNS; numberOfColumns++) {
-            linesWithMatchingGapIndices.put(numberOfColumns, findConsecutiveLinesWithMatchingGaps(linesWithGaps, mainBodyTextFrame.getWidth(), numberOfColumns));
-        }
-
-        int optimalNumberOfColumns = findOptimalNumberOfColumns(linesWithMatchingGapIndices, linesWithGaps.size());
-        if (optimalNumberOfColumns == 1) {
-            return List.of(mainBodyTextFrame);
-        }
-        return buildColumns(mainBodyTextFrame, getLinesWithMatchingGaps(linesWithMatchingGapIndices.get(optimalNumberOfColumns), linesWithGaps), optimalNumberOfColumns);
-    }
-
-
-    private static List<Integer> findConsecutiveLinesWithMatchingGaps(List<List<Rectangle2D>> linesWithGaps, double width, int numberOfColumns) {
-
-        List<Boolean> booleans = lineHasMatchingGap(linesWithGaps, width, numberOfColumns);
-        return findConsecutiveTrueIndicesWithMaxLengthRun(booleans);
-    }
-
-
-    private List<Boolean> lineHasMatchingGap(List<List<Rectangle2D>> linesWithGaps, double width, int numberOfColumns) {
-
-        return linesWithGaps.stream()
-                .map(blocksWithGaps -> IntStream.range(1, numberOfColumns)
-                        .allMatch(columnIndex -> noBlocksIntersectX(blocksWithGaps, calculateGapLocation(width, numberOfColumns, columnIndex))))
-                .toList();
-    }
-
-
-    private List<Integer> findConsecutiveTrueIndicesWithMaxLengthRun(List<Boolean> booleans) {
-
-        List<Integer> maxConsecutiveTrueIndices = new LinkedList<>();
-        List<Integer> currentConsecutiveTrueIndices = new LinkedList<>();
-        for (int i = 0; i < booleans.size(); i++) {
-            if (!booleans.get(i)) {
-                if (currentConsecutiveTrueIndices.isEmpty()) {
-                    continue;
-                }
-                if (currentConsecutiveTrueIndices.size() > maxConsecutiveTrueIndices.size()) {
-                    maxConsecutiveTrueIndices = currentConsecutiveTrueIndices;
-                }
-                currentConsecutiveTrueIndices = new LinkedList<>();
-                continue;
-            }
-            currentConsecutiveTrueIndices.add(i);
-        }
-        if (currentConsecutiveTrueIndices.size() > maxConsecutiveTrueIndices.size()) {
-            return currentConsecutiveTrueIndices;
-        }
-        return maxConsecutiveTrueIndices;
-    }
-
-
-    private static int findOptimalNumberOfColumns(Map<Integer, List<Integer>> linesWithMatchingGapIndices, Integer numberOfLines) {
-
-        return linesWithMatchingGapIndices.entrySet()
-                .stream()
-                .max(comparePercentages(numberOfLines))
-                .filter(entry -> percentageIsAboveThreshold(entry, numberOfLines))
-                .map(Map.Entry::getKey)
-                .orElse(1);
-    }
-
-
-    private List<Rectangle2D> buildColumns(Rectangle2D mainBodyTextFrame, List<Rectangle2D> rectanglesToMerge, int optimalColumnCount) {
-
-        if (optimalColumnCount == 1 || rectanglesToMerge.isEmpty()) {
-            return List.of(mainBodyTextFrame);
-        }
-
-        double maxY = rectanglesToMerge.get(0).getMaxY();
-        double minY = rectanglesToMerge.get(rectanglesToMerge.size() - 1).getMinY();
-
-        List<Rectangle2D> columns = new LinkedList<>();
-        double width = mainBodyTextFrame.getWidth() / optimalColumnCount;
-        double height = maxY - minY;
-        for (int i = 0; i < optimalColumnCount; i++) {
-            columns.add(new Rectangle2D.Double(mainBodyTextFrame.getMinY() + i * width, minY, width, height));
-        }
-        return columns;
-    }
-
-
-    private Comparator<Map.Entry<Integer, List<Integer>>> comparePercentages(Integer numberOfLines) {
-
-        return Comparator.comparingDouble(entry -> calculatePercentage(entry.getValue().size(), numberOfLines));
-    }
-
-
-    private List<Rectangle2D> getLinesWithMatchingGaps(List<Integer> linesWithMatchingGapIndices, List<List<Rectangle2D>> linesWithGaps) {
-
-        return linesWithMatchingGapIndices.stream().map(linesWithGaps::get).flatMap(Collection::stream).toList();
-    }
-
-
-    private boolean percentageIsAboveThreshold(Map.Entry<Integer, List<Integer>> entry, Integer numberOfLines) {
-
-        return calculatePercentage(entry.getValue().size(), numberOfLines) > SPLITTABLE_LINE_PERCENTAGE_THRESHOLD;
-    }
-
-
-    private double calculatePercentage(Integer numberOfMatchingLines, Integer numberOfLines) {
-
-        return ((double) numberOfMatchingLines) / ((double) numberOfLines);
-    }
-
-
-    private double calculateGapLocation(double pageWidth, int numberOfColumns, int columnIndex) {
-
-        return (pageWidth / numberOfColumns) * columnIndex;
-    }
-
-
-    private Boolean noBlocksIntersectX(List<Rectangle2D> blocksWithGaps, double x) {
-
-        return blocksWithGaps.stream().noneMatch(rect -> rect.getMaxX() > x && rect.getMinX() < x);
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/DocuMineBlockificationService.java

@@ -1,207 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import static java.util.stream.Collectors.toSet;
-
-import java.util.ArrayList;
-import java.util.Comparator;
-import java.util.Iterator;
-import java.util.List;
-import java.util.regex.Matcher;
-import java.util.regex.Pattern;
-import java.util.stream.Collectors;
-
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.FloatFrequencyCounter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.Orientation;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.RulingTextDirAdjustUtil;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.Patterns;
-
-@Service
-@ConditionalOnProperty(prefix = "application", name = "type", havingValue = "DocuMine")
-@SuppressWarnings("all")
-public class DocuMineBlockificationService implements BlockificationService{
-
-    static final float THRESHOLD = 1f;
-
-    Pattern pattern = Patterns.getCompiledPattern("^(\\d{1,2}\\.){1,3}\\d{1,2}\\.?\\s[0-9A-Za-z ()-]{2,50}", true);
-
-
-    /**
-     * This method is building blocks by expanding the minX/maxX and minY/maxY value on each word that is not split by the conditions.
-     * This method must use text direction adjusted postions (DirAdj). Where {0,0} is on the upper left. Never try to change this!
-     * Rulings (Table lines) must be adjusted to the text directions as well, when checking if a block is split by a ruling.
-     * @param textPositions The words of a page.
-     * @param horizontalRulingLines Horizontal table lines.
-     * @param verticalRulingLines Vertical table lines.
-     * @return Page object that contains the Textblock and text statistics.
-     */
-    public ClassificationPage blockify(List<TextPositionSequence> textPositions, List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines) {
-
-        List<TextPositionSequence> chunkWords = new ArrayList<>();
-        List<AbstractPageBlock> chunkBlockList1 = new ArrayList<>();
-
-        float minX = 1000, maxX = 0, minY = 1000, maxY = 0;
-        TextPositionSequence prev = null;
-
-        boolean wasSplitted = false;
-        Float splitX1 = null;
-        for (TextPositionSequence word : textPositions) {
-
-            boolean lineSeparation = prev != null && word.getMinYDirAdj() - maxY > Math.min(word.getHeight(), prev.getHeight()) * 1.25;
-            boolean startFromTop = prev != null && word.getMinYDirAdj() < prev.getMinYDirAdj() - prev.getTextHeight();
-            boolean splitByX = prev != null && maxX + 50 < word.getMinXDirAdj() && prev.getMinYDirAdj() == word.getMinYDirAdj();
-            boolean negativeXGap = prev != null && word.getMinXDirAdj() - minX < -5;
-            boolean newLineAfterSplit = prev != null && word.getMinYDirAdj() != prev.getMinYDirAdj() && wasSplitted && splitX1 != word.getMinXDirAdj();
-            boolean isSplitByRuling = isSplitByRuling(minX, minY, maxX, maxY, word, horizontalRulingLines, verticalRulingLines);
-            boolean splitByDir = prev != null && !prev.getDir().equals(word.getDir());
-            boolean splitByOtherFontAndOtherY = prev != null && prev.getMaxYDirAdj() != word.getMaxYDirAdj() && (word.getFontStyle().contains("bold") && !prev.getFontStyle().contains("bold") || prev.getFontStyle().contains("bold") && !word.getFontStyle().contains("bold"));
-
-            Matcher matcher = pattern.matcher(chunkWords.stream().collect(Collectors.joining(" ")).toString());
-            boolean startsOnSameX = Math.abs(minX - word.getMinXDirAdj()) < 5 && matcher.matches();
-
-            if (prev != null && (lineSeparation || startFromTop || splitByDir || isSplitByRuling || splitByOtherFontAndOtherY || negativeXGap || startsOnSameX)) {
-
-                Orientation prevOrientation = null;
-                if (!chunkBlockList1.isEmpty()) {
-                    prevOrientation = chunkBlockList1.get(chunkBlockList1.size() - 1).getOrientation();
-                }
-
-                TextPageBlock cb1 = buildTextBlock(chunkWords);
-                chunkBlockList1.add(cb1);
-                chunkWords = new ArrayList<>();
-
-                if (splitByX && !isSplitByRuling) {
-                    wasSplitted = true;
-                    cb1.setOrientation(Orientation.LEFT);
-                    splitX1 = word.getMinXDirAdj();
-                } else if (newLineAfterSplit && !isSplitByRuling) {
-                    wasSplitted = false;
-                    cb1.setOrientation(Orientation.RIGHT);
-                    splitX1 = null;
-                } else if (prevOrientation != null && prevOrientation.equals(Orientation.RIGHT) && (lineSeparation || !startFromTop || !splitByX || !newLineAfterSplit || !isSplitByRuling)) {
-                    cb1.setOrientation(Orientation.LEFT);
-                }
-
-                minX = 1000;
-                maxX = 0;
-                minY = 1000;
-                maxY = 0;
-                prev = null;
-            }
-
-            chunkWords.add(word);
-
-            prev = word;
-            if (word.getMinXDirAdj() < minX) {
-                minX = word.getMinXDirAdj();
-            }
-            if (word.getMaxXDirAdj() > maxX) {
-                maxX = word.getMaxXDirAdj();
-            }
-            if (word.getMinYDirAdj() < minY) {
-                minY = word.getMinYDirAdj();
-            }
-            if (word.getMaxYDirAdj() > maxY) {
-                maxY = word.getMaxYDirAdj();
-            }
-        }
-
-        TextPageBlock cb1 = buildTextBlock(chunkWords);
-        if (cb1 != null) {
-            chunkBlockList1.add(cb1);
-        }
-
-        return new ClassificationPage(chunkBlockList1);
-    }
-
-
-    private boolean equalsWithThreshold(float f1, float f2) {
-
-        return Math.abs(f1 - f2) < THRESHOLD;
-    }
-
-
-    private TextPageBlock buildTextBlock(List<TextPositionSequence> wordBlockList) {
-
-        TextPageBlock textBlock = null;
-
-        FloatFrequencyCounter lineHeightFrequencyCounter = new FloatFrequencyCounter();
-        FloatFrequencyCounter fontSizeFrequencyCounter = new FloatFrequencyCounter();
-        FloatFrequencyCounter spaceFrequencyCounter = new FloatFrequencyCounter();
-        StringFrequencyCounter fontFrequencyCounter = new StringFrequencyCounter();
-        StringFrequencyCounter styleFrequencyCounter = new StringFrequencyCounter();
-
-        for (TextPositionSequence wordBlock : wordBlockList) {
-
-            lineHeightFrequencyCounter.add(wordBlock.getTextHeight());
-            fontSizeFrequencyCounter.add(wordBlock.getFontSize());
-            spaceFrequencyCounter.add(wordBlock.getSpaceWidth());
-            fontFrequencyCounter.add(wordBlock.getFont());
-            styleFrequencyCounter.add(wordBlock.getFontStyle());
-
-            if (textBlock == null) {
-                textBlock = new TextPageBlock(wordBlock.getMinXDirAdj(), wordBlock.getMaxXDirAdj(), wordBlock.getMinYDirAdj(), wordBlock.getMaxYDirAdj(), wordBlockList, wordBlock.getRotation());
-            } else {
-                TextPageBlock spatialEntity = textBlock.union(wordBlock);
-                textBlock.resize(spatialEntity.getMinX(), spatialEntity.getMinY(), spatialEntity.getWidth(), spatialEntity.getHeight());
-            }
-        }
-
-        if (textBlock != null) {
-            textBlock.setMostPopularWordFont(fontFrequencyCounter.getMostPopular());
-            textBlock.setMostPopularWordStyle(styleFrequencyCounter.getMostPopular());
-            textBlock.setMostPopularWordFontSize(fontSizeFrequencyCounter.getMostPopular());
-            textBlock.setMostPopularWordHeight(lineHeightFrequencyCounter.getMostPopular());
-            textBlock.setMostPopularWordSpaceWidth(spaceFrequencyCounter.getMostPopular());
-            textBlock.setHighestFontSize(fontSizeFrequencyCounter.getHighest());
-        }
-
-        if (textBlock != null && textBlock.getSequences() != null && textBlock.getSequences().stream().map(t -> round(t.getMinYDirAdj(), 3)).collect(toSet()).size() == 1) {
-            textBlock.getSequences().sort(Comparator.comparing(TextPositionSequence::getMinXDirAdj));
-        }
-        return textBlock;
-    }
-
-
-    private boolean isSplitByRuling(float minX,
-                                    float minY,
-                                    float maxX,
-                                    float maxY,
-                                    TextPositionSequence word,
-                                    List<Ruling> horizontalRulingLines,
-                                    List<Ruling> verticalRulingLines) {
-
-        return isSplitByRuling(maxX, minY, word.getMinXDirAdj(), word.getMinYDirAdj(), verticalRulingLines, word.getDir().getDegrees(), word.getPageWidth(), word.getPageHeight()) //
-                || isSplitByRuling(minX, minY, word.getMinXDirAdj(), word.getMaxYDirAdj(), horizontalRulingLines, word.getDir().getDegrees(), word.getPageWidth(), word.getPageHeight()) //
-                || isSplitByRuling(maxX, minY, word.getMinXDirAdj(), word.getMinYDirAdj(), horizontalRulingLines, word.getDir().getDegrees(), word.getPageWidth(), word.getPageHeight()) //
-                || isSplitByRuling(minX, minY, word.getMinXDirAdj(), word.getMaxYDirAdj(), verticalRulingLines, word.getDir().getDegrees(), word.getPageWidth(), word.getPageHeight()); //
-    }
-
-
-    private boolean isSplitByRuling(float previousX2, float previousY1, float currentX1, float currentY1, List<Ruling> rulingLines, float dir, float pageWidth, float pageHeight) {
-
-        for (Ruling ruling : rulingLines) {
-            var line = RulingTextDirAdjustUtil.convertToDirAdj(ruling, dir, pageWidth, pageHeight);
-            if (line.intersectsLine(previousX2, previousY1, currentX1, currentY1)) {
-                return true;
-            }
-        }
-        return false;
-    }
-
-
-    private double round(float value, int decimalPoints) {
-
-        var d = Math.pow(10, decimalPoints);
-        return Math.round(value * d) / d;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/DocuMineClassificationService.java

@@ -1,117 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.util.List;
-import java.util.Locale;
-import java.util.regex.Matcher;
-import java.util.regex.Pattern;
-
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.PositionUtils;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.Patterns;
-
-import lombok.RequiredArgsConstructor;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@Service
-@RequiredArgsConstructor
-@ConditionalOnProperty(prefix = "application", name = "type", havingValue = "DocuMine")
-public class DocuMineClassificationService implements ClassificationService {
-
-    private final BodyTextFrameService bodyTextFrameService;
-
-
-    public void classifyDocument(ClassificationDocument document) {
-
-        Rectangle bodyTextFrame = bodyTextFrameService.calculateBodyTextFrame(document.getPages(), document.getFontSizeCounter(), false);
-        Rectangle landscapeBodyTextFrame = bodyTextFrameService.calculateBodyTextFrame(document.getPages(), document.getFontSizeCounter(), true);
-
-        log.debug("Document FontSize counters are: {}", document.getFontSizeCounter().getCountPerValue());
-
-        for (ClassificationPage page : document.getPages()) {
-            bodyTextFrameService.setBodyTextFrameAdjustedToPage(page, bodyTextFrame, landscapeBodyTextFrame);
-            classifyPage(page, document);
-        }
-    }
-
-
-    private void classifyPage(ClassificationPage page, ClassificationDocument document) {
-
-        for (AbstractPageBlock textBlock : page.getTextBlocks()) {
-            if (textBlock instanceof TextPageBlock) {
-                classifyBlock((TextPageBlock) textBlock, page, document);
-            }
-        }
-    }
-
-
-    private void classifyBlock(TextPageBlock textBlock, ClassificationPage page, ClassificationDocument document) {
-
-        var bodyTextFrame = page.getBodyTextFrame();
-
-        var pattern = Patterns.getCompiledPattern("^(\\d{1,2}\\.){1,3}\\d{1,2}\\.?\\s[0-9A-Za-z \\[\\]]{2,50}", true);
-        var pattern2 = Patterns.getCompiledPattern("\\p{L}{3,}", true);
-        var pattern3 = Patterns.getCompiledPattern("^(\\d{1,1}\\.){1,3}\\d{1,2}\\.?\\s[a-z]{1,2}\\/[a-z]{1,2}.*", false);
-
-        Matcher matcher = pattern.matcher(textBlock.toString());
-        Matcher matcher2 = pattern2.matcher(textBlock.toString());
-        Matcher matcher3 = pattern3.matcher(textBlock.toString());
-
-        if (document.getFontSizeCounter().getMostPopular() == null) {
-            textBlock.setClassification(PageBlockType.OTHER);
-            return;
-        }
-        if (PositionUtils.isOverBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
-                .getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
-            textBlock.setClassification(PageBlockType.HEADER);
-
-        } else if (PositionUtils.isUnderBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
-                .getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
-            textBlock.setClassification(PageBlockType.FOOTER);
-        } else if (page.getPageNumber() == 1 && (PositionUtils.getHeightDifferenceBetweenChunkWordAndDocumentWord(textBlock,
-                document.getTextHeightCounter().getMostPopular()) > 2.5 && textBlock.getHighestFontSize() > document.getFontSizeCounter().getMostPopular() || page.getTextBlocks()
-                .size() == 1)) {
-            if (!Pattern.matches("[0-9]+", textBlock.toString())) {
-                textBlock.setClassification(PageBlockType.TITLE);
-            }
-        } else if (textBlock.getText().length() > 5 && (textBlock.getMostPopularWordHeight() > document.getTextHeightCounter()
-                .getMostPopular() || textBlock.getMostPopularWordFontSize() > document.getFontSizeCounter().getMostPopular()) && PositionUtils.getApproxLineCount(textBlock) < 5.9
-
-                && (textBlock.getMostPopularWordStyle().contains("bold") && Character.isDigit(textBlock.toString().charAt(0)) && !matcher2.matches() && !textBlock.toString()
-                .contains(":") || textBlock.toString().equals(textBlock.toString().toUpperCase(Locale.ROOT)) && !matcher2.matches() && !textBlock.toString()
-                .contains(":") || textBlock.toString().startsWith("APPENDIX") || textBlock.toString().startsWith("FIGURE") || textBlock.toString()
-                .startsWith("TABLE")) && !textBlock.toString().endsWith(":") && matcher2.find()) {
-            textBlock.setClassification(PageBlockType.getHeadlineType(1));
-            document.setHeadlines(true);
-
-        } else if (matcher.find() && PositionUtils.getApproxLineCount(textBlock) < 2.9 && matcher2.find() && !matcher3.matches()) {
-            textBlock.setClassification(PageBlockType.getHeadlineType(2));
-            document.setHeadlines(true);
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter()
-                .getMostPopular() && textBlock.getMostPopularWordStyle().equals("bold") && !document.getFontStyleCounter().getMostPopular().equals("bold")) {
-            textBlock.setClassification(PageBlockType.PARAGRAPH_BOLD);
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFont()
-                .equals(document.getFontCounter().getMostPopular()) && textBlock.getMostPopularWordStyle()
-                .equals(document.getFontStyleCounter().getMostPopular()) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter().getMostPopular()) {
-            textBlock.setClassification(PageBlockType.PARAGRAPH);
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter()
-                .getMostPopular() && textBlock.getMostPopularWordStyle().equals("italic") && !document.getFontStyleCounter()
-                .getMostPopular()
-                .equals("italic") && PositionUtils.getApproxLineCount(textBlock) < 2.9) {
-            textBlock.setClassification(PageBlockType.PARAGRAPH_ITALIC);
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock)) {
-            textBlock.setClassification(PageBlockType.PARAGRAPH_UNKNOWN);
-        } else {
-            textBlock.setClassification(PageBlockType.OTHER);
-        }
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/LineDetectionService.java

@@ -1,115 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.awt.geom.Rectangle2D;
-import java.util.Collections;
-import java.util.LinkedList;
-import java.util.List;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.document.utils.RectangleTransformations;
-import com.iqser.red.service.redaction.v1.server.document.utils.TextPositionSequenceComparator;
-
-import lombok.AllArgsConstructor;
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public class LineDetectionService {
-
-    private static final double X_GAP_FACTOR = 1; // multiplied with average text height, determines the minimum distance of gaps in lines
-
-
-    public static List<List<Rectangle2D>> findLinesWithGaps(List<TextPositionSequence> textPositionSequences) {
-
-        if (textPositionSequences.isEmpty()) {
-            return Collections.emptyList();
-        }
-
-        final double avgTextPositionHeight = textPositionSequences.stream().mapToDouble(TextPositionSequence::getHeight).average().orElseThrow();
-
-        Context context = Context.init();
-
-        List<TextPositionSequence> sortedTextPositionSequence = textPositionSequences.stream().sorted(new TextPositionSequenceComparator()).toList();
-
-        var previousTextPosition = sortedTextPositionSequence.get(0);
-        context.textPositionsToMerge.add(previousTextPosition);
-        for (TextPositionSequence currentTextPosition : sortedTextPositionSequence.subList(1, sortedTextPositionSequence.size())) {
-            if (isNewLine(currentTextPosition, previousTextPosition, avgTextPositionHeight) || isSplitByOrientation(currentTextPosition, previousTextPosition)) {
-                addBlockToLine(context);
-                startNewLine(currentTextPosition, context);
-            } else if (isXGap(currentTextPosition, previousTextPosition, avgTextPositionHeight)) {
-                addBlockToLine(context);
-                startNewBlock(currentTextPosition, context);
-            } else {
-                context.textPositionsToMerge.add(currentTextPosition);
-            }
-            previousTextPosition = currentTextPosition;
-        }
-        addBlockToLine(context);
-        return context.linesWithGaps;
-    }
-
-
-    private static boolean isXGap(TextPositionSequence currentTextPosition, TextPositionSequence previousTextPosition, double avgTextPositionHeight) {
-
-        return Math.abs(previousTextPosition.getMaxXDirAdj() - currentTextPosition.getMinXDirAdj()) > (avgTextPositionHeight * X_GAP_FACTOR);
-    }
-
-
-    private static boolean isSplitByOrientation(TextPositionSequence currentTextPosition, TextPositionSequence previousTextPosition) {
-
-        return !previousTextPosition.getDir().equals(currentTextPosition.getDir());
-    }
-
-
-    private static boolean isNewLine(TextPositionSequence currentTextPosition, TextPositionSequence previousTextPosition, double avgTextPositionHeight) {
-
-        return Math.abs(previousTextPosition.getMinYDirAdj() - currentTextPosition.getMinYDirAdj()) > avgTextPositionHeight;
-    }
-
-
-    private static void startNewBlock(TextPositionSequence currentTextPosition, Context context) {
-
-        context.textPositionsToMerge = new LinkedList<>();
-        context.textPositionsToMerge.add(currentTextPosition);
-    }
-
-
-    private static void addBlockToLine(Context context) {
-
-        context.blocksInLine.add(textPositionBBox(context.textPositionsToMerge));
-    }
-
-
-    private static void startNewLine(TextPositionSequence current, Context context) {
-
-        context.blocksInLine = new LinkedList<>();
-        startNewBlock(current, context);
-        context.linesWithGaps.add(context.blocksInLine);
-    }
-
-
-    private Rectangle2D textPositionBBox(List<TextPositionSequence> textPositionSequences) {
-
-        return RectangleTransformations.rectangleBBox(textPositionSequences.stream().map(TextPositionSequence::getRectangle).toList());
-    }
-
-
-    @AllArgsConstructor
-    private class Context {
-
-        List<List<Rectangle2D>> linesWithGaps;
-        List<Rectangle2D> blocksInLine;
-        List<TextPositionSequence> textPositionsToMerge;
-
-
-        public static Context init() {
-
-            List<List<Rectangle2D>> initialLinesWithGaps = new LinkedList<>();
-            List<Rectangle2D> initialBlocksInLine = new LinkedList<>();
-            initialLinesWithGaps.add(initialBlocksInLine);
-            return new Context(initialLinesWithGaps, initialBlocksInLine, new LinkedList<>());
-        }
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/PdfSegmentationService.java

@@ -1,185 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.io.File;
-import java.io.FileOutputStream;
-import java.io.IOException;
-import java.io.InputStream;
-import java.util.ArrayList;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-
-import org.apache.commons.io.IOUtils;
-import org.apache.pdfbox.io.MemoryUsageSetting;
-import org.apache.pdfbox.pdmodel.PDDocument;
-import org.apache.pdfbox.pdmodel.PDPage;
-import org.apache.pdfbox.pdmodel.common.PDRectangle;
-import org.springframework.beans.factory.annotation.Value;
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.ImageServiceResponseAdapter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.TableServiceResponseAdapter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.PdfTableCell;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.CleanRulings;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.parsing.PDFLinesTextStripper;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.FileUtils;
-import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
-
-import lombok.RequiredArgsConstructor;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@Service
-@RequiredArgsConstructor
-public class PdfSegmentationService {
-
-    private final RedactionServiceSettings redactionServiceSettings;
-    private final RulingCleaningService rulingCleaningService;
-    private final TableExtractionService tableExtractionService;
-    private final BlockificationService blockificationService;
-    private final ClassificationService classificationService;
-    private final SectionsBuilderService sectionsBuilderService;
-    private final ImageServiceResponseAdapter imageServiceResponseAdapter;
-    private final TableServiceResponseAdapter tableServiceResponseAdapter;
-
-    @Value("${application.type:RedactManager}")
-    private String applicationType;
-
-
-    public ClassificationDocument parseDocument(String dossierId,
-                                                String fileId,
-                                                InputStream documentInputStream,
-                                                Map<Integer, List<ClassifiedImage>> pdfImages) throws IOException {
-
-        PDDocument pdDocument = null;
-        File tempFile = null;
-        try {
-            Map<Integer, List<PdfTableCell>> pdfTableCells = new HashMap<>();
-            if (redactionServiceSettings.isCvTableParsingEnabled()) {
-                pdfTableCells = tableServiceResponseAdapter.convertTables(dossierId, fileId);
-            }
-
-            tempFile = FileUtils.createTempFile("document", ".pdf");
-            try (var fos = new FileOutputStream(tempFile)) {
-                IOUtils.copy(documentInputStream, fos);
-
-                // initialize required variables
-                ClassificationDocument document = new ClassificationDocument();
-                List<ClassificationPage> pages = new ArrayList<>();
-
-                pdDocument = PDDocument.load(tempFile, MemoryUsageSetting.setupMixed(67108864L));
-                pdDocument.setAllSecurityToBeRemoved(true);
-                long pageCount = pdDocument.getNumberOfPages();
-
-                for (int pageNumber = 1; pageNumber <= pageCount; pageNumber++) {
-                    processPage(pdfImages, pdDocument, pdfTableCells, document, pages, pageNumber);
-                }
-
-                document.setPages(pages);
-
-                classificationService.classifyDocument(document);
-                sectionsBuilderService.buildSections(document);
-                sectionsBuilderService.addImagesToSections(document);
-
-                return document;
-            }
-        } finally {
-            if (pdDocument != null) {
-                pdDocument.close();
-            }
-
-            FileUtils.deleteFile(tempFile);
-        }
-    }
-
-
-    private void processPage(Map<Integer, List<ClassifiedImage>> pdfImages,
-                             PDDocument pdDocument,
-                             Map<Integer, List<PdfTableCell>> pdfTableCells,
-                             ClassificationDocument document,
-                             List<ClassificationPage> pages,
-                             int pageNumber) throws IOException {
-
-        PDFLinesTextStripper stripper = new PDFLinesTextStripper();
-        PDPage pdPage = pdDocument.getPage(pageNumber - 1);
-        stripper.setPageNumber(pageNumber);
-        stripper.setStartPage(pageNumber);
-        stripper.setEndPage(pageNumber);
-        stripper.setPdpage(pdPage);
-
-        if(applicationType.equals("DocuMine")){
-            stripper.setSortByPosition(true);
-        }
-        
-        stripper.getText(pdDocument);
-
-        PDRectangle pdr = pdPage.getMediaBox();
-
-        int rotation = pdPage.getRotation();
-        boolean isLandscape = pdr.getWidth() > pdr.getHeight() && (rotation == 0 || rotation == 180) || pdr.getHeight() > pdr.getWidth() && (rotation == 90 || rotation == 270);
-
-        PDRectangle cropbox = pdPage.getCropBox();
-        CleanRulings cleanRulings = rulingCleaningService.getCleanRulings(pdfTableCells.get(pageNumber),
-                stripper.getRulings(),
-                stripper.getMinCharWidth(),
-                stripper.getMaxCharHeight());
-        // var columns = ColumnDetectionService.detectColumns(stripper.getTextPositionSequences(), RectangleTransformations.toRectangle2D(pdPage.getCropBox()));
-        ClassificationPage page = blockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
-
-        page.setRotation(rotation);
-        page.setLandscape(isLandscape);
-        page.setPageNumber(pageNumber);
-        page.setPageWidth(cropbox.getWidth());
-        page.setPageHeight(cropbox.getHeight());
-
-        // If images is ocr needs to be calculated before textBlocks are moved into tables, otherwise findOcr algorithm needs to be adopted.
-        if (pdfImages != null && pdfImages.containsKey(pageNumber)) {
-            page.setImages(pdfImages.get(pageNumber));
-            imageServiceResponseAdapter.findOcr(page);
-        }
-
-        tableExtractionService.extractTables(cleanRulings, page);
-        buildPageStatistics(page);
-        increaseDocumentStatistics(page, document);
-
-        pages.add(page);
-    }
-
-
-    private void increaseDocumentStatistics(ClassificationPage page, ClassificationDocument document) {
-
-        if (!page.isLandscape()) {
-            document.getFontSizeCounter().addAll(page.getFontSizeCounter().getCountPerValue());
-        }
-        document.getFontCounter().addAll(page.getFontCounter().getCountPerValue());
-        document.getTextHeightCounter().addAll(page.getTextHeightCounter().getCountPerValue());
-        document.getFontStyleCounter().addAll(page.getFontStyleCounter().getCountPerValue());
-    }
-
-
-    private void buildPageStatistics(ClassificationPage page) {
-
-        // Collect all statistics for the page, except from blocks inside tables, as tables will always be added to BodyTextFrame.
-        for (AbstractPageBlock textBlock : page.getTextBlocks()) {
-            if (textBlock instanceof TextPageBlock) {
-                if (((TextPageBlock) textBlock).getSequences() == null) {
-                    continue;
-                }
-                for (TextPositionSequence word : ((TextPageBlock) textBlock).getSequences()) {
-                    page.getTextHeightCounter().add(word.getTextHeight());
-                    page.getFontCounter().add(word.getFont());
-                    page.getFontSizeCounter().add(word.getFontSize());
-                    page.getFontStyleCounter().add(word.getFontStyle());
-                }
-            }
-        }
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/RedactManagerBlockificationService.java

@@ -1,280 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import static java.util.stream.Collectors.toSet;
-
-import java.util.ArrayList;
-import java.util.Comparator;
-import java.util.Iterator;
-import java.util.List;
-
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.FloatFrequencyCounter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.Orientation;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.StringFrequencyCounter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.RulingTextDirAdjustUtil;
-
-@Service
-@SuppressWarnings("all")
-@ConditionalOnProperty(prefix = "application", name = "type", havingValue = "RedactManager")
-public class RedactManagerBlockificationService implements BlockificationService {
-
-    static final float THRESHOLD = 1f;
-
-
-    /**
-     * This method is building blocks by expanding the minX/maxX and minY/maxY value on each word that is not split by the conditions.
-     * This method must use text direction adjusted postions (DirAdj). Where {0,0} is on the upper left. Never try to change this!
-     * Rulings (TablePageBlock lines) must be adjusted to the text directions as well, when checking if a block is split by a ruling.
-     *
-     * @param textPositions         The words of a page.
-     * @param horizontalRulingLines Horizontal table lines.
-     * @param verticalRulingLines   Vertical table lines.
-     * @return Page object that contains the Textblock and text statistics.
-     */
-    public ClassificationPage blockify(List<TextPositionSequence> textPositions, List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines) {
-
-        int indexOnPage = 0;
-        List<TextPositionSequence> chunkWords = new ArrayList<>();
-        List<AbstractPageBlock> chunkBlockList = new ArrayList<>();
-
-        float minX = 1000, maxX = 0, minY = 1000, maxY = 0;
-        TextPositionSequence prev = null;
-
-        boolean wasSplitted = false;
-        Float splitX1 = null;
-        for (TextPositionSequence word : textPositions) {
-
-            boolean lineSeparation = word.getMinYDirAdj() - maxY > word.getHeight() * 1.25;
-            boolean startFromTop = prev != null && word.getMinYDirAdj() < prev.getMinYDirAdj() - prev.getTextHeight();
-            boolean splitByX = prev != null && maxX + 50 < word.getMinXDirAdj() && prev.getMinYDirAdj() == word.getMinYDirAdj();
-            boolean xIsBeforeFirstX = prev != null && word.getMinXDirAdj() < minX;
-            boolean newLineAfterSplit = prev != null && word.getMinYDirAdj() != prev.getMinYDirAdj() && wasSplitted && splitX1 != word.getMinXDirAdj();
-            boolean isSplitByRuling = isSplitByRuling(minX, minY, maxX, maxY, word, horizontalRulingLines, verticalRulingLines);
-            boolean splitByDir = prev != null && !prev.getDir().equals(word.getDir());
-
-            if (prev != null && (lineSeparation || startFromTop || splitByX || splitByDir || isSplitByRuling)) {
-
-                Orientation prevOrientation = null;
-                if (!chunkBlockList.isEmpty()) {
-                    prevOrientation = chunkBlockList.get(chunkBlockList.size() - 1).getOrientation();
-                }
-
-                TextPageBlock cb1 = buildTextBlock(chunkWords, indexOnPage);
-                indexOnPage++;
-
-                chunkBlockList.add(cb1);
-                chunkWords = new ArrayList<>();
-
-                if (splitByX && !isSplitByRuling) {
-                    wasSplitted = true;
-                    cb1.setOrientation(Orientation.LEFT);
-                    splitX1 = word.getMinXDirAdj();
-                } else if (newLineAfterSplit && !isSplitByRuling) {
-                    wasSplitted = false;
-                    cb1.setOrientation(Orientation.RIGHT);
-                    splitX1 = null;
-                } else if (prevOrientation != null && prevOrientation.equals(Orientation.RIGHT) && (lineSeparation || !startFromTop || !splitByX || !newLineAfterSplit || !isSplitByRuling)) {
-                    cb1.setOrientation(Orientation.LEFT);
-                }
-
-                minX = 1000;
-                maxX = 0;
-                minY = 1000;
-                maxY = 0;
-                prev = null;
-            }
-
-            chunkWords.add(word);
-
-            prev = word;
-            if (word.getMinXDirAdj() < minX) {
-                minX = word.getMinXDirAdj();
-            }
-            if (word.getMaxXDirAdj() > maxX) {
-                maxX = word.getMaxXDirAdj();
-            }
-            if (word.getMinYDirAdj() < minY) {
-                minY = word.getMinYDirAdj();
-            }
-            if (word.getMaxYDirAdj() > maxY) {
-                maxY = word.getMaxYDirAdj();
-            }
-        }
-
-        TextPageBlock cb1 = buildTextBlock(chunkWords, indexOnPage);
-        if (cb1 != null) {
-            chunkBlockList.add(cb1);
-        }
-
-        Iterator<AbstractPageBlock> itty = chunkBlockList.iterator();
-
-        TextPageBlock previousLeft = null;
-        TextPageBlock previousRight = null;
-        while (itty.hasNext()) {
-            TextPageBlock block = (TextPageBlock) itty.next();
-
-            if (previousLeft != null && block.getOrientation().equals(Orientation.LEFT)) {
-                if (previousLeft.getMinY() > block.getMinY() && block.getMaxY() + block.getMostPopularWordHeight() > previousLeft.getMinY()) {
-                    previousLeft.add(block);
-                    itty.remove();
-                    continue;
-                }
-            }
-
-            if (previousRight != null && block.getOrientation().equals(Orientation.RIGHT)) {
-                if (previousRight.getMinY() > block.getMinY() && block.getMaxY() + block.getMostPopularWordHeight() > previousRight.getMinY()) {
-                    previousRight.add(block);
-                    itty.remove();
-                    continue;
-                }
-            }
-
-            if (block.getOrientation().equals(Orientation.LEFT)) {
-                previousLeft = block;
-            } else if (block.getOrientation().equals(Orientation.RIGHT)) {
-                previousRight = block;
-            }
-        }
-
-        itty = chunkBlockList.iterator();
-        TextPageBlock previous = null;
-        while (itty.hasNext()) {
-            TextPageBlock block = (TextPageBlock) itty.next();
-
-            if (previous != null && previous.getOrientation().equals(Orientation.LEFT) && block.getOrientation().equals(Orientation.LEFT) && equalsWithThreshold(block.getMaxY(),
-                    previous.getMaxY()) || previous != null && previous.getOrientation().equals(Orientation.LEFT) && block.getOrientation()
-                    .equals(Orientation.RIGHT) && equalsWithThreshold(block.getMaxY(), previous.getMaxY())) {
-                previous.add(block);
-                itty.remove();
-                continue;
-            }
-
-            previous = block;
-        }
-
-        return new ClassificationPage(chunkBlockList);
-    }
-
-
-    private boolean equalsWithThreshold(float f1, float f2) {
-
-        return Math.abs(f1 - f2) < THRESHOLD;
-    }
-
-
-    private TextPageBlock buildTextBlock(List<TextPositionSequence> wordBlockList, int indexOnPage) {
-
-        TextPageBlock textBlock = null;
-
-        FloatFrequencyCounter lineHeightFrequencyCounter = new FloatFrequencyCounter();
-        FloatFrequencyCounter fontSizeFrequencyCounter = new FloatFrequencyCounter();
-        FloatFrequencyCounter spaceFrequencyCounter = new FloatFrequencyCounter();
-        StringFrequencyCounter fontFrequencyCounter = new StringFrequencyCounter();
-        StringFrequencyCounter styleFrequencyCounter = new StringFrequencyCounter();
-
-        for (TextPositionSequence wordBlock : wordBlockList) {
-
-            lineHeightFrequencyCounter.add(wordBlock.getTextHeight());
-            fontSizeFrequencyCounter.add(wordBlock.getFontSize());
-            spaceFrequencyCounter.add(wordBlock.getSpaceWidth());
-            fontFrequencyCounter.add(wordBlock.getFont());
-            styleFrequencyCounter.add(wordBlock.getFontStyle());
-
-            if (textBlock == null) {
-                textBlock = new TextPageBlock(wordBlock.getMinXDirAdj(),
-                        wordBlock.getMaxXDirAdj(),
-                        wordBlock.getMinYDirAdj(),
-                        wordBlock.getMaxYDirAdj(),
-                        wordBlockList,
-                        wordBlock.getRotation());
-            } else {
-                TextPageBlock spatialEntity = textBlock.union(wordBlock);
-                textBlock.resize(spatialEntity.getMinX(), spatialEntity.getMinY(), spatialEntity.getWidth(), spatialEntity.getHeight());
-            }
-        }
-
-        if (textBlock != null) {
-            textBlock.setMostPopularWordFont(fontFrequencyCounter.getMostPopular());
-            textBlock.setMostPopularWordStyle(styleFrequencyCounter.getMostPopular());
-            textBlock.setMostPopularWordFontSize(fontSizeFrequencyCounter.getMostPopular());
-            textBlock.setMostPopularWordHeight(lineHeightFrequencyCounter.getMostPopular());
-            textBlock.setMostPopularWordSpaceWidth(spaceFrequencyCounter.getMostPopular());
-            textBlock.setHighestFontSize(fontSizeFrequencyCounter.getHighest());
-        }
-
-        if (textBlock != null && textBlock.getSequences() != null && textBlock.getSequences().stream().map(t -> round(t.getMinYDirAdj(), 3)).collect(toSet()).size() == 1) {
-            textBlock.getSequences().sort(Comparator.comparing(TextPositionSequence::getMinXDirAdj));
-        }
-        return textBlock;
-    }
-
-
-    private boolean isSplitByRuling(float minX,
-                                    float minY,
-                                    float maxX,
-                                    float maxY,
-                                    TextPositionSequence word,
-                                    List<Ruling> horizontalRulingLines,
-                                    List<Ruling> verticalRulingLines) {
-
-        return isSplitByRuling(maxX,
-                minY,
-                word.getMinXDirAdj(),
-                word.getMinYDirAdj(),
-                verticalRulingLines,
-                word.getDir().getDegrees(),
-                word.getPageWidth(),
-                word.getPageHeight()) //
-                || isSplitByRuling(minX,
-                minY,
-                word.getMinXDirAdj(),
-                word.getMaxYDirAdj(),
-                horizontalRulingLines,
-                word.getDir().getDegrees(),
-                word.getPageWidth(),
-                word.getPageHeight()) //
-                || isSplitByRuling(maxX,
-                minY,
-                word.getMinXDirAdj(),
-                word.getMinYDirAdj(),
-                horizontalRulingLines,
-                word.getDir().getDegrees(),
-                word.getPageWidth(),
-                word.getPageHeight()) //
-                || isSplitByRuling(minX,
-                minY,
-                word.getMinXDirAdj(),
-                word.getMaxYDirAdj(),
-                verticalRulingLines,
-                word.getDir().getDegrees(),
-                word.getPageWidth(),
-                word.getPageHeight());
-    }
-
-
-    private boolean isSplitByRuling(float previousX2, float previousY1, float currentX1, float currentY1, List<Ruling> rulingLines, float dir, float pageWidth, float pageHeight) {
-
-        for (Ruling ruling : rulingLines) {
-            var line = RulingTextDirAdjustUtil.convertToDirAdj(ruling, dir, pageWidth, pageHeight);
-            if (line.intersectsLine(previousX2, previousY1, currentX1, currentY1)) {
-                return true;
-            }
-        }
-        return false;
-    }
-
-
-    private double round(float value, int decimalPoints) {
-
-        var d = Math.pow(10, decimalPoints);
-        return Math.round(value * d) / d;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/RedactManagerClassificationService.java

@@ -1,117 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.util.List;
-import java.util.regex.Pattern;
-
-import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty;
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.PositionUtils;
-
-import lombok.RequiredArgsConstructor;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@Service
-@RequiredArgsConstructor
-@ConditionalOnProperty(prefix = "application", name = "type", havingValue = "RedactManager")
-public class RedactManagerClassificationService implements ClassificationService {
-
-    private final BodyTextFrameService bodyTextFrameService;
-
-
-    public void classifyDocument(ClassificationDocument document) {
-
-        Rectangle bodyTextFrame = bodyTextFrameService.calculateBodyTextFrame(document.getPages(), document.getFontSizeCounter(), false);
-        Rectangle landscapeBodyTextFrame = bodyTextFrameService.calculateBodyTextFrame(document.getPages(), document.getFontSizeCounter(), true);
-        List<Float> headlineFontSizes = document.getFontSizeCounter().getHighterThanMostPopular();
-
-        log.debug("Document FontSize counters are: {}", document.getFontSizeCounter().getCountPerValue());
-
-        for (ClassificationPage page : document.getPages()) {
-            bodyTextFrameService.setBodyTextFrameAdjustedToPage(page, bodyTextFrame, landscapeBodyTextFrame);
-            classifyPage(page, document, headlineFontSizes);
-        }
-    }
-
-
-    private void classifyPage(ClassificationPage page, ClassificationDocument document, List<Float> headlineFontSizes) {
-
-        for (AbstractPageBlock textBlock : page.getTextBlocks()) {
-            if (textBlock instanceof TextPageBlock) {
-                classifyBlock((TextPageBlock) textBlock, page, document, headlineFontSizes);
-            }
-        }
-    }
-
-
-    private void classifyBlock(TextPageBlock textBlock, ClassificationPage page, ClassificationDocument document, List<Float> headlineFontSizes) {
-
-        var bodyTextFrame = page.getBodyTextFrame();
-
-        if (document.getFontSizeCounter().getMostPopular() == null) {
-            textBlock.setClassification(PageBlockType.OTHER);
-            return;
-        }
-        if (PositionUtils.isOverBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
-                .getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
-            textBlock.setClassification(PageBlockType.HEADER);
-
-        } else if (PositionUtils.isUnderBodyTextFrame(bodyTextFrame, textBlock, page.getRotation()) && (document.getFontSizeCounter()
-                .getMostPopular() == null || textBlock.getHighestFontSize() <= document.getFontSizeCounter().getMostPopular())) {
-            textBlock.setClassification(PageBlockType.FOOTER);
-        } else if (page.getPageNumber() == 1 && (PositionUtils.getHeightDifferenceBetweenChunkWordAndDocumentWord(textBlock,
-                document.getTextHeightCounter().getMostPopular()) > 2.5 && textBlock.getHighestFontSize() > document.getFontSizeCounter().getMostPopular() || page.getTextBlocks()
-                .size() == 1)) {
-            if (!Pattern.matches("[0-9]+", textBlock.toString())) {
-                textBlock.setClassification(PageBlockType.TITLE);
-            }
-        } else if (textBlock.getMostPopularWordFontSize() > document.getFontSizeCounter()
-                .getMostPopular() && PositionUtils.getApproxLineCount(textBlock) < 4.9 && (textBlock.getMostPopularWordStyle().equals("bold") || !document.getFontStyleCounter()
-                .getCountPerValue()
-                .containsKey("bold") && textBlock.getMostPopularWordFontSize() > document.getFontSizeCounter().getMostPopular() + 1) && textBlock.getSequences()
-                .get(0)
-                .getTextPositions()
-                .get(0)
-                .getFontSizeInPt() >= textBlock.getMostPopularWordFontSize()) {
-
-            for (int i = 1; i <= headlineFontSizes.size(); i++) {
-                if (textBlock.getMostPopularWordFontSize() == headlineFontSizes.get(i - 1)) {
-                    textBlock.setClassification(PageBlockType.getHeadlineType(i));
-                    document.setHeadlines(true);
-                }
-            }
-        } else if (!textBlock.getText().startsWith("Figure ") && PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordStyle()
-                .equals("bold") && !document.getFontStyleCounter().getMostPopular().equals("bold") && PositionUtils.getApproxLineCount(textBlock) < 2.9 && textBlock.getSequences()
-                .get(0)
-                .getTextPositions()
-                .get(0)
-                .getFontSizeInPt() >= textBlock.getMostPopularWordFontSize()) {
-            textBlock.setClassification(PageBlockType.getHeadlineType(headlineFontSizes.size() + 1));
-            document.setHeadlines(true);
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter()
-                .getMostPopular() && textBlock.getMostPopularWordStyle().equals("bold") && !document.getFontStyleCounter().getMostPopular().equals("bold")) {
-            textBlock.setClassification(PageBlockType.PARAGRAPH_BOLD);
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFont()
-                .equals(document.getFontCounter().getMostPopular()) && textBlock.getMostPopularWordStyle()
-                .equals(document.getFontStyleCounter().getMostPopular()) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter().getMostPopular()) {
-            textBlock.setClassification(PageBlockType.PARAGRAPH);
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter()
-                .getMostPopular() && textBlock.getMostPopularWordStyle().equals("italic") && !document.getFontStyleCounter()
-                .getMostPopular()
-                .equals("italic") && PositionUtils.getApproxLineCount(textBlock) < 2.9) {
-            textBlock.setClassification(PageBlockType.PARAGRAPH_ITALIC);
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock)) {
-            textBlock.setClassification(PageBlockType.PARAGRAPH_UNKNOWN);
-        } else {
-            textBlock.setClassification(PageBlockType.OTHER);
-        }
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/RulingCleaningService.java

@@ -1,235 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.awt.geom.Line2D;
-import java.awt.geom.Point2D;
-import java.util.ArrayList;
-import java.util.Collection;
-import java.util.Collections;
-import java.util.Comparator;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.table.PdfTableCell;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.CleanRulings;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.Utils;
-import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
-
-import lombok.RequiredArgsConstructor;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@Service
-@RequiredArgsConstructor
-public class RulingCleaningService {
-
-    private final RedactionServiceSettings redactionServiceSettings;
-
-
-    public CleanRulings getCleanRulings(List<PdfTableCell> pdfTableCells, List<Ruling> rulings, float minCharWidth, float maxCharHeight) {
-
-        if (!rulings.isEmpty()) {
-            snapPoints(rulings, minCharWidth, maxCharHeight);
-        }
-
-        List<Ruling> vrs = new ArrayList<>();
-        for (Ruling vr : rulings) {
-            if (vr.vertical()) {
-                vrs.add(vr);
-            }
-        }
-        if (vrs.isEmpty() && redactionServiceSettings.isCvTableParsingEnabled()) {
-            vrs.addAll(extractVerticalRulings(pdfTableCells));
-        }
-        List<Ruling> verticalRulingLines = collapseOrientedRulings(vrs);
-
-        List<Ruling> hrs = new ArrayList<>();
-        for (Ruling hr : rulings) {
-            if (hr.horizontal()) {
-                hrs.add(hr);
-            }
-        }
-        if (hrs.isEmpty() && redactionServiceSettings.isCvTableParsingEnabled()) {
-            hrs.addAll(extractHorizontalRulings(pdfTableCells));
-        }
-        List<Ruling> horizontalRulingLines = collapseOrientedRulings(hrs);
-
-        return CleanRulings.builder().vertical(verticalRulingLines).horizontal(horizontalRulingLines).build();
-    }
-
-
-    public void snapPoints(List<? extends Line2D.Float> rulings, float xThreshold, float yThreshold) {
-
-        // collect points and keep a Line -> p1,p2 map
-        Map<Line2D.Float, Point2D[]> linesToPoints = new HashMap<>();
-        List<Point2D> points = new ArrayList<>();
-        for (Line2D.Float r : rulings) {
-            Point2D p1 = r.getP1();
-            Point2D p2 = r.getP2();
-            linesToPoints.put(r, new Point2D[]{p1, p2});
-            points.add(p1);
-            points.add(p2);
-        }
-
-        // snap by X
-        points.sort(Comparator.comparingDouble(Point2D::getX));
-
-        List<List<Point2D>> groupedPoints = new ArrayList<>();
-        groupedPoints.add(new ArrayList<>(Collections.singletonList(points.get(0))));
-
-        for (Point2D p : points.subList(1, points.size() - 1)) {
-            List<Point2D> last = groupedPoints.get(groupedPoints.size() - 1);
-            if (Math.abs(p.getX() - last.get(0).getX()) < xThreshold) {
-                groupedPoints.get(groupedPoints.size() - 1).add(p);
-            } else {
-                groupedPoints.add(new ArrayList<>(Collections.singletonList(p)));
-            }
-        }
-
-        for (List<Point2D> group : groupedPoints) {
-            float avgLoc = 0;
-            for (Point2D p : group) {
-                avgLoc += p.getX();
-            }
-            avgLoc /= group.size();
-            for (Point2D p : group) {
-                p.setLocation(avgLoc, p.getY());
-            }
-        }
-        // ---
-
-        // snap by Y
-        points.sort(Comparator.comparingDouble(Point2D::getY));
-
-        groupedPoints = new ArrayList<>();
-        groupedPoints.add(new ArrayList<>(Collections.singletonList(points.get(0))));
-
-        for (Point2D p : points.subList(1, points.size() - 1)) {
-            List<Point2D> last = groupedPoints.get(groupedPoints.size() - 1);
-            if (Math.abs(p.getY() - last.get(0).getY()) < yThreshold) {
-                groupedPoints.get(groupedPoints.size() - 1).add(p);
-            } else {
-                groupedPoints.add(new ArrayList<>(Collections.singletonList(p)));
-            }
-        }
-
-        for (List<Point2D> group : groupedPoints) {
-            float avgLoc = 0;
-            for (Point2D p : group) {
-                avgLoc += p.getY();
-            }
-            avgLoc /= group.size();
-            for (Point2D p : group) {
-                p.setLocation(p.getX(), avgLoc);
-            }
-        }
-        // ---
-
-        // finally, modify lines
-        for (Map.Entry<Line2D.Float, Point2D[]> ltp : linesToPoints.entrySet()) {
-            Point2D[] p = ltp.getValue();
-            ltp.getKey().setLine(p[0], p[1]);
-        }
-    }
-
-
-    private Collection<? extends Ruling> extractVerticalRulings(List<PdfTableCell> pdfTableCells) {
-
-        List<Ruling> vrs = new ArrayList<>();
-
-        if (pdfTableCells != null) {
-            for (PdfTableCell pdfTableCell : pdfTableCells) {
-                Ruling leftLine = createRuling(pdfTableCell.getX0(), pdfTableCell.getX0(), pdfTableCell.getY0(), pdfTableCell.getY1());
-                Ruling rightLine = createRuling(pdfTableCell.getX1(), pdfTableCell.getX1(), pdfTableCell.getY0(), pdfTableCell.getY1());
-                vrs.add(leftLine);
-                vrs.add(rightLine);
-            }
-        }
-        return vrs;
-    }
-
-
-    private Collection<? extends Ruling> extractHorizontalRulings(List<PdfTableCell> pdfTableCells) {
-
-        List<Ruling> hrs = new ArrayList<>();
-
-        if (pdfTableCells != null) {
-            for (PdfTableCell pdfTableCell : pdfTableCells) {
-                Ruling topLine = createRuling(pdfTableCell.getX0(), pdfTableCell.getX1(), pdfTableCell.getY1(), pdfTableCell.getY1());
-                Ruling baseLine = createRuling(pdfTableCell.getX0(), pdfTableCell.getX1(), pdfTableCell.getY0(), pdfTableCell.getY0());
-                hrs.add(topLine);
-                hrs.add(baseLine);
-            }
-        }
-
-        return hrs;
-    }
-
-
-    private Ruling createRuling(float tableCellX0, float tableCellX1, float tableCellY0, float tableCellY1) {
-
-        float x0 = tableCellX0;
-        float x1 = tableCellX1;
-        float y0 = tableCellY0;
-        float y1 = tableCellY1;
-
-        if (x1 < x0) {
-            x0 = tableCellX1;
-            x1 = tableCellX0;
-        }
-
-        if (y1 < y0) {
-            y0 = tableCellY1;
-            y1 = tableCellY0;
-        }
-
-        return new Ruling(new Point2D.Float(x0, y0), new Point2D.Float(x1, y1));
-    }
-
-
-    private List<Ruling> collapseOrientedRulings(List<Ruling> lines) {
-
-        int COLINEAR_OR_PARALLEL_PIXEL_EXPAND_AMOUNT = 1;
-        return collapseOrientedRulings(lines, COLINEAR_OR_PARALLEL_PIXEL_EXPAND_AMOUNT);
-    }
-
-
-    private List<Ruling> collapseOrientedRulings(List<Ruling> lines, int expandAmount) {
-
-        ArrayList<Ruling> rv = new ArrayList<>();
-        lines.sort((a, b) -> {
-            final float diff = a.getPosition() - b.getPosition();
-            return Float.compare(diff == 0 ? a.getStart() - b.getStart() : diff, 0f);
-        });
-
-        for (Ruling next_line : lines) {
-            Ruling last = rv.isEmpty() ? null : rv.get(rv.size() - 1);
-            // if current line colinear with next, and are "close enough": expand current line
-            if (last != null && Utils.feq(next_line.getPosition(), last.getPosition()) && last.nearlyIntersects(next_line, expandAmount)) {
-                final float lastStart = last.getStart();
-                final float lastEnd = last.getEnd();
-
-                final boolean lastFlipped = lastStart > lastEnd;
-                final boolean nextFlipped = next_line.getStart() > next_line.getEnd();
-
-                boolean differentDirections = nextFlipped != lastFlipped;
-                float nextS = differentDirections ? next_line.getEnd() : next_line.getStart();
-                float nextE = differentDirections ? next_line.getStart() : next_line.getEnd();
-
-                final float newStart = lastFlipped ? Math.max(nextS, lastStart) : Math.min(nextS, lastStart);
-                final float newEnd = lastFlipped ? Math.min(nextE, lastEnd) : Math.max(nextE, lastEnd);
-                last.setStartEnd(newStart, newEnd);
-                assert !last.oblique();
-            } else if (next_line.length() == 0) {
-                continue;
-            } else {
-                rv.add(next_line);
-            }
-        }
-        return rv;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/SectionsBuilderService.java

@@ -1,304 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.util.ArrayList;
-import java.util.Collections;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-import java.util.stream.Collectors;
-
-import org.apache.commons.collections4.CollectionUtils;
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationFooter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationHeader;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationSection;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.PageBlockType;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Cell;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.UnclassifiedText;
-
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@Service
-public class SectionsBuilderService {
-
-    public void buildSections(ClassificationDocument document) {
-
-        List<AbstractPageBlock> chunkWords = new ArrayList<>();
-        List<ClassificationSection> chunkBlockList = new ArrayList<>();
-        List<ClassificationHeader> headers = new ArrayList<>();
-        List<ClassificationFooter> footers = new ArrayList<>();
-        List<UnclassifiedText> unclassifiedTexts = new ArrayList<>();
-
-        AbstractPageBlock prev = null;
-
-        String lastHeadline = "";
-        TablePageBlock previousTable = null;
-        for (ClassificationPage page : document.getPages()) {
-            List<TextPageBlock> header = new ArrayList<>();
-            List<TextPageBlock> footer = new ArrayList<>();
-            List<TextPageBlock> unclassifiedText = new ArrayList<>();
-            for (AbstractPageBlock current : page.getTextBlocks()) {
-
-                if (current.getClassification() == null) {
-                    continue;
-                }
-
-                current.setPage(page.getPageNumber());
-
-                if (current.getClassification().equals(PageBlockType.HEADER)) {
-                    header.add((TextPageBlock) current);
-                    continue;
-                }
-
-                if (current.getClassification().equals(PageBlockType.FOOTER)) {
-                    footer.add((TextPageBlock) current);
-                    continue;
-                }
-
-                if (current.getClassification().equals(PageBlockType.OTHER)) {
-                    unclassifiedText.add((TextPageBlock) current);
-                    continue;
-                }
-
-                if (prev != null && current.getClassification().isHeadline() && !prev.getClassification().isHeadline() || !document.isHeadlines()) {
-                    ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline);
-                    chunkBlock.setHeadline(lastHeadline);
-                    if (document.isHeadlines()) {
-                        lastHeadline = current.getText();
-                    }
-                    chunkBlockList.add(chunkBlock);
-                    chunkWords = new ArrayList<>();
-                    if (CollectionUtils.isNotEmpty(chunkBlock.getTables())) {
-                        previousTable = chunkBlock.getTables().get(chunkBlock.getTables().size() - 1);
-                    }
-                }
-                if (current instanceof TablePageBlock table) {
-                    // Distribute header information for subsequent tables
-                    mergeTableMetadata(table, previousTable);
-                    previousTable = table;
-                }
-                chunkWords.add(current);
-                prev = current;
-            }
-
-            if (!header.isEmpty()) {
-                headers.add(new ClassificationHeader(header));
-            }
-            if (!footer.isEmpty()) {
-                footers.add(new ClassificationFooter(footer));
-            }
-            if (!unclassifiedText.isEmpty()) {
-                unclassifiedTexts.add(new UnclassifiedText(unclassifiedText));
-            }
-        }
-
-        ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline);
-        chunkBlock.setHeadline(lastHeadline);
-        chunkBlockList.add(chunkBlock);
-
-        document.setSections(chunkBlockList);
-        document.setHeaders(headers);
-        document.setFooters(footers);
-        document.setUnclassifiedTexts(unclassifiedTexts);
-    }
-
-
-    public void addImagesToSections(ClassificationDocument document) {
-
-        Map<Integer, List<ClassificationSection>> sectionMap = new HashMap<>();
-        for (ClassificationSection section : document.getSections()) {
-            for (AbstractPageBlock container : section.getPageBlocks()) {
-
-                List<ClassificationSection> sectionsOnPage = sectionMap.computeIfAbsent(container.getPage(), c -> new ArrayList<>());
-                if (sectionsOnPage.contains(section)) {
-                    continue;
-                }
-                sectionsOnPage.add(section);
-            }
-        }
-
-        if (sectionMap.isEmpty()) {
-            ClassificationSection section = new ClassificationSection();
-            document.getSections().add(section);
-            sectionMap.computeIfAbsent(1, x -> new ArrayList<>()).add(section);
-        }
-
-        // first page is always a paragraph, else we can't process pages 1..N,
-        // where N is the first found page with a paragraph
-        if (sectionMap.get(1) == null) {
-            ClassificationSection section = new ClassificationSection();
-            document.getSections().add(section);
-            sectionMap.computeIfAbsent(1, x -> new ArrayList<>()).add(section);
-        }
-
-        for (ClassificationPage page : document.getPages()) {
-            for (ClassifiedImage image : page.getImages()) {
-                List<ClassificationSection> sectionsOnPage = sectionMap.get(page.getPageNumber());
-                if (sectionsOnPage == null) {
-                    int i = page.getPageNumber();
-                    while (sectionsOnPage == null) {
-                        sectionsOnPage = sectionMap.get(i);
-                        i--;
-                    }
-                }
-                for (ClassificationSection section : sectionsOnPage) {
-                    Float xMin = null;
-                    Float yMin = null;
-                    Float xMax = null;
-                    Float yMax = null;
-
-                    for (AbstractPageBlock abs : section.getPageBlocks()) {
-                        if (abs.getPage() != page.getPageNumber()) {
-                            continue;
-                        }
-
-                        if (abs.getMinX() < abs.getMaxX()) {
-                            if (xMin == null || abs.getMinX() < xMin) {
-                                xMin = abs.getMinX();
-                            }
-                            if (xMax == null || abs.getMaxX() > xMax) {
-                                xMax = abs.getMaxX();
-                            }
-                        } else {
-                            if (xMin == null || abs.getMaxX() < xMin) {
-                                xMin = abs.getMaxX();
-                            }
-                            if (xMax == null || abs.getMinX() > xMax) {
-                                xMax = abs.getMinX();
-                            }
-                        }
-
-                        if (abs.getMinY() < abs.getMaxY()) {
-                            if (yMin == null || abs.getMinY() < yMin) {
-                                yMin = abs.getMinY();
-                            }
-                            if (yMax == null || abs.getMaxY() > yMax) {
-                                yMax = abs.getMaxY();
-                            }
-                        } else {
-                            if (yMin == null || abs.getMaxY() < yMin) {
-                                yMin = abs.getMaxY();
-                            }
-                            if (yMax == null || abs.getMinY() > yMax) {
-                                yMax = abs.getMinY();
-                            }
-                        }
-
-                    }
-
-                    log.debug("Image position x: {}, y: {}", image.getPosition().getX(), image.getPosition().getY());
-                    log.debug("Paragraph position xMin: {}, xMax: {}, yMin: {}, yMax: {}", xMin, xMax, yMin, yMax);
-
-                    if (xMin != null && xMax != null && yMin != null && yMax != null && image.getPosition().getX() >= xMin && image.getPosition()
-                            .getX() <= xMax && image.getPosition().getY() >= yMin && image.getPosition().getY() <= yMax) {
-                        section.getImages().add(image);
-                        image.setAppendedToSection(true);
-                    }
-                }
-                if (!image.isAppendedToSection()) {
-                    log.debug("Image uses first paragraph");
-                    sectionsOnPage.get(0).getImages().add(image);
-                    image.setAppendedToSection(true);
-                }
-            }
-        }
-    }
-
-
-    private void mergeTableMetadata(TablePageBlock currentTable, TablePageBlock previousTable) {
-
-        // Distribute header information for subsequent tables
-        if (previousTable != null && hasInvalidHeaderInformation(currentTable) && hasValidHeaderInformation(previousTable)) {
-            List<Cell> previousTableNonHeaderRow = getRowWithNonHeaderCells(previousTable);
-            List<Cell> tableNonHeaderRow = getRowWithNonHeaderCells(currentTable);
-            // Allow merging of tables if header row is separated from first logical non-header row
-            if (previousTableNonHeaderRow.isEmpty() && previousTable.getRowCount() == 1 && previousTable.getRows().get(0).size() == tableNonHeaderRow.size()) {
-                previousTableNonHeaderRow = previousTable.getRows().get(0).stream().map(cell -> {
-                    Cell fakeCell = new Cell(cell.getPoints()[0], cell.getPoints()[2]);
-                    fakeCell.setHeaderCells(Collections.singletonList(cell));
-                    return fakeCell;
-                }).collect(Collectors.toList());
-            }
-            if (previousTableNonHeaderRow.size() == tableNonHeaderRow.size()) {
-                for (int i = currentTable.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
-                    List<Cell> row = currentTable.getRows().get(i);
-                    if (row.size() == tableNonHeaderRow.size() && row.stream().allMatch(cell -> cell.getHeaderCells().isEmpty())) {
-                        for (int j = 0; j < row.size(); j++) {
-                            row.get(j).setHeaderCells(previousTableNonHeaderRow.get(j).getHeaderCells());
-                        }
-                    }
-                }
-            }
-        }
-    }
-
-
-    private ClassificationSection buildTextBlock(List<AbstractPageBlock> wordBlockList, String lastHeadline) {
-
-        ClassificationSection section = new ClassificationSection();
-
-        for (AbstractPageBlock container : wordBlockList) {
-            if (container instanceof TablePageBlock table) {
-
-                if (lastHeadline == null || lastHeadline.isEmpty()) {
-                    table.setHeadline("Text in table");
-                } else {
-                    table.setHeadline("TablePageBlock in: " + lastHeadline);
-                }
-
-                section.getPageBlocks().add(table);
-                continue;
-            }
-
-            TextPageBlock wordBlock = (TextPageBlock) container;
-            section.getPageBlocks().add(wordBlock);
-        }
-        return section;
-    }
-
-
-    private boolean hasValidHeaderInformation(TablePageBlock table) {
-
-        return !hasInvalidHeaderInformation(table);
-    }
-
-
-    private boolean hasInvalidHeaderInformation(TablePageBlock table) {
-
-        return table.getRows().stream().flatMap(row -> row.stream().filter(cell -> CollectionUtils.isNotEmpty(cell.getHeaderCells()))).findAny().isEmpty();
-
-    }
-
-
-    private List<Cell> getRowWithNonHeaderCells(TablePageBlock table) {
-
-        for (int i = table.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
-            List<Cell> row = table.getRows().get(i);
-            if (row.size() == 1) {
-                continue;
-            }
-            boolean allNonHeader = true;
-            for (Cell cell : row) {
-                if (cell.isHeaderCell()) {
-                    allNonHeader = false;
-                    break;
-                }
-            }
-            if (allNonHeader) {
-                return row;
-            }
-        }
-
-        return Collections.emptyList();
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/service/TableExtractionService.java

@@ -1,337 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service;
-
-import java.awt.geom.Point2D;
-import java.util.ArrayList;
-import java.util.Comparator;
-import java.util.HashMap;
-import java.util.HashSet;
-import java.util.Iterator;
-import java.util.List;
-import java.util.Map;
-import java.util.Set;
-import java.util.stream.Collectors;
-
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.AbstractPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationPage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Cell;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.CleanRulings;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Rectangle;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.TablePageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils.Utils;
-
-@Service
-public class TableExtractionService {
-
-    private static final Comparator<Point2D> X_FIRST_POINT_COMPARATOR = (arg0, arg1) -> {
-
-        int rv = 0;
-        float arg0X = Utils.round(arg0.getX(), 2);
-        float arg0Y = Utils.round(arg0.getY(), 2);
-        float arg1X = Utils.round(arg1.getX(), 2);
-        float arg1Y = Utils.round(arg1.getY(), 2);
-
-        if (arg0X > arg1X) {
-            rv = 1;
-        } else if (arg0X < arg1X) {
-            rv = -1;
-        } else if (arg0Y > arg1Y) {
-            rv = 1;
-        } else if (arg0Y < arg1Y) {
-            rv = -1;
-        }
-        return rv;
-    };
-    private static final Comparator<Point2D> POINT_COMPARATOR = (arg0, arg1) -> {
-
-        int rv = 0;
-        float arg0X = Utils.round(arg0.getX(), 2);
-        float arg0Y = Utils.round(arg0.getY(), 2);
-        float arg1X = Utils.round(arg1.getX(), 2);
-        float arg1Y = Utils.round(arg1.getY(), 2);
-
-        if (arg0Y > arg1Y) {
-            rv = 1;
-        } else if (arg0Y < arg1Y) {
-            rv = -1;
-        } else if (arg0X > arg1X) {
-            rv = 1;
-        } else if (arg0X < arg1X) {
-            rv = -1;
-        }
-        return rv;
-    };
-
-
-    /**
-     * Finds tables on a page and moves textblocks into cells of the found tables.
-     * Note: This algorithm uses Pdf Coordinate System where {0,0} rotated with the page rotation.
-     * 0 -> LowerLeft
-     * 90 -> UpperLeft
-     * 180 -> UpperRight
-     * 270 -> LowerRight
-     * <p>
-     * DirAdj (Text direction adjusted) values can not be used here.
-     *
-     * @param cleanRulings The lines used to build the table.
-     * @param page         Page object that contains textblocks and statistics.
-     */
-    public void extractTables(CleanRulings cleanRulings, ClassificationPage page) {
-
-        List<Cell> cells = findCells(cleanRulings.getHorizontal(), cleanRulings.getVertical());
-
-        List<TextPageBlock> toBeRemoved = new ArrayList<>();
-
-        for (AbstractPageBlock abstractPageBlock : page.getTextBlocks()) {
-            TextPageBlock textBlock = (TextPageBlock) abstractPageBlock;
-            for (Cell cell : cells) {
-                if (cell.hasMinimumSize() && cell.intersects(textBlock.getPdfMinX(),
-                        textBlock.getPdfMinY(),
-                        textBlock.getPdfMaxX() - textBlock.getPdfMinX(),
-                        textBlock.getPdfMaxY() - textBlock.getPdfMinY())) {
-                    cell.addTextBlock(textBlock);
-                    toBeRemoved.add(textBlock);
-                    break;
-                }
-            }
-        }
-
-        cells = new ArrayList<>(new HashSet<>(cells));
-        Utils.sort(cells, Rectangle.ILL_DEFINED_ORDER);
-
-        List<Rectangle> spreadsheetAreas = findSpreadsheetsFromCells(cells).stream().filter(r -> r.getWidth() > 0f && r.getHeight() > 0f).collect(Collectors.toList());
-
-        List<TablePageBlock> tables = new ArrayList<>();
-        for (Rectangle area : spreadsheetAreas) {
-
-            List<Cell> overlappingCells = new ArrayList<>();
-            for (Cell c : cells) {
-                if (c.hasMinimumSize() && c.intersects(area)) {
-                    overlappingCells.add(c);
-                }
-            }
-            tables.add(new TablePageBlock(overlappingCells, area, page.getRotation()));
-        }
-
-        for (TablePageBlock table : tables) {
-            int position = -1;
-
-            Iterator<AbstractPageBlock> itty = page.getTextBlocks().iterator();
-            while (itty.hasNext()) {
-                AbstractPageBlock textBlock = itty.next();
-                if (textBlock instanceof TextPageBlock ? table.containsBlock((TextPageBlock) textBlock) : table.contains(textBlock) && position == -1) {
-                    position = page.getTextBlocks().indexOf(textBlock);
-                }
-            }
-            if (position != -1) {
-                page.getTextBlocks().add(position, table);
-            }
-        }
-
-        page.getTextBlocks().removeAll(toBeRemoved);
-    }
-
-
-    public List<Cell> findCells(List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines) {
-
-        List<Cell> cellsFound = new ArrayList<>();
-        Map<Point2D, Ruling[]> intersectionPoints = Ruling.findIntersections(horizontalRulingLines, verticalRulingLines);
-        List<Point2D> intersectionPointsList = new ArrayList<>(intersectionPoints.keySet());
-        intersectionPointsList.sort(POINT_COMPARATOR);
-
-        for (int i = 0; i < intersectionPointsList.size(); i++) {
-            Point2D topLeft = intersectionPointsList.get(i);
-            Ruling[] hv = intersectionPoints.get(topLeft);
-
-            // CrossingPointsDirectlyBelow( topLeft );
-            List<Point2D> xPoints = new ArrayList<>();
-            // CrossingPointsDirectlyToTheRight( topLeft );
-            List<Point2D> yPoints = new ArrayList<>();
-
-            for (Point2D p : intersectionPointsList.subList(i, intersectionPointsList.size())) {
-                if (p.getX() == topLeft.getX() && p.getY() > topLeft.getY()) {
-                    xPoints.add(p);
-                }
-                if (p.getY() == topLeft.getY() && p.getX() > topLeft.getX()) {
-                    yPoints.add(p);
-                }
-            }
-            outer:
-            for (Point2D xPoint : xPoints) {
-                // is there a vertical edge b/w topLeft and xPoint?
-                if (!hv[1].equals(intersectionPoints.get(xPoint)[1])) {
-                    continue;
-                }
-                for (Point2D yPoint : yPoints) {
-                    // is there an horizontal edge b/w topLeft and yPoint ?
-                    if (!hv[0].equals(intersectionPoints.get(yPoint)[0])) {
-                        continue;
-                    }
-                    Point2D btmRight = new Point2D.Float((float) yPoint.getX(), (float) xPoint.getY());
-                    if (intersectionPoints.containsKey(btmRight) && intersectionPoints.get(btmRight)[0].equals(intersectionPoints.get(xPoint)[0]) && intersectionPoints.get(btmRight)[1].equals(
-                            intersectionPoints.get(yPoint)[1])) {
-                        cellsFound.add(new Cell(topLeft, btmRight));
-                        break outer;
-                    }
-                }
-            }
-        }
-
-        // TODO create cells for vertical ruling lines with aligned endpoints at the top/bottom of a grid
-        // that aren't connected with an horizontal ruler?
-        // see: https://github.com/jazzido/tabula-extractor/issues/78#issuecomment-41481207
-
-        return cellsFound;
-    }
-
-
-    private List<Rectangle> findSpreadsheetsFromCells(List<? extends Rectangle> cells) {
-        // via: http://stackoverflow.com/questions/13746284/merging-multiple-adjacent-rectangles-into-one-polygon
-        List<Rectangle> rectangles = new ArrayList<>();
-        Set<Point2D> pointSet = new HashSet<>();
-        Map<Point2D, Point2D> edgesH = new HashMap<>();
-        Map<Point2D, Point2D> edgesV = new HashMap<>();
-        int i = 0;
-
-        for (Rectangle cell : cells) {
-            for (Point2D pt : cell.getPoints()) {
-                if (pointSet.contains(pt)) { // shared vertex, remove it
-                    pointSet.remove(pt);
-                } else {
-                    pointSet.add(pt);
-                }
-            }
-        }
-
-        // X first sort
-        List<Point2D> pointsSortX = new ArrayList<>(pointSet);
-        pointsSortX.sort(X_FIRST_POINT_COMPARATOR);
-        // Y first sort
-        List<Point2D> pointsSortY = new ArrayList<>(pointSet);
-        pointsSortY.sort(POINT_COMPARATOR);
-
-        while (i < pointSet.size()) {
-            float currY = (float) pointsSortY.get(i).getY();
-            while (i < pointSet.size() && Utils.feq(pointsSortY.get(i).getY(), currY)) {
-                edgesH.put(pointsSortY.get(i), pointsSortY.get(i + 1));
-                edgesH.put(pointsSortY.get(i + 1), pointsSortY.get(i));
-                i += 2;
-            }
-        }
-
-        i = 0;
-        while (i < pointSet.size()) {
-            float currX = (float) pointsSortX.get(i).getX();
-            while (i < pointSet.size() && Utils.feq(pointsSortX.get(i).getX(), currX)) {
-                edgesV.put(pointsSortX.get(i), pointsSortX.get(i + 1));
-                edgesV.put(pointsSortX.get(i + 1), pointsSortX.get(i));
-                i += 2;
-            }
-        }
-
-        // Get all the polygons
-        List<List<PolygonVertex>> polygons = new ArrayList<>();
-        Point2D nextVertex;
-        while (!edgesH.isEmpty()) {
-            ArrayList<PolygonVertex> polygon = new ArrayList<>();
-            Point2D first = edgesH.keySet().iterator().next();
-            polygon.add(new PolygonVertex(first, Direction.HORIZONTAL));
-            edgesH.remove(first);
-
-            while (true) {
-                PolygonVertex curr = polygon.get(polygon.size() - 1);
-                PolygonVertex lastAddedVertex;
-                if (curr.direction == Direction.HORIZONTAL) {
-                    nextVertex = edgesV.get(curr.point);
-                    edgesV.remove(curr.point);
-                    lastAddedVertex = new PolygonVertex(nextVertex, Direction.VERTICAL);
-                } else {
-                    nextVertex = edgesH.get(curr.point);
-                    edgesH.remove(curr.point);
-                    lastAddedVertex = new PolygonVertex(nextVertex, Direction.HORIZONTAL);
-                }
-                polygon.add(lastAddedVertex);
-
-                if (lastAddedVertex.equals(polygon.get(0))) {
-                    // closed polygon
-                    polygon.remove(polygon.size() - 1);
-                    break;
-                }
-            }
-
-            for (PolygonVertex vertex : polygon) {
-                edgesH.remove(vertex.point);
-                edgesV.remove(vertex.point);
-            }
-            polygons.add(polygon);
-        }
-
-        // calculate grid-aligned minimum area rectangles for each found polygon
-        for (List<PolygonVertex> poly : polygons) {
-            float top = Float.MAX_VALUE;
-            float left = Float.MAX_VALUE;
-            float bottom = Float.MIN_VALUE;
-            float right = Float.MIN_VALUE;
-            for (PolygonVertex pt : poly) {
-                top = (float) Math.min(top, pt.point.getY());
-                left = (float) Math.min(left, pt.point.getX());
-                bottom = (float) Math.max(bottom, pt.point.getY());
-                right = (float) Math.max(right, pt.point.getX());
-            }
-            rectangles.add(new Rectangle(top, left, right - left, bottom - top));
-        }
-
-        return rectangles;
-    }
-
-
-    private enum Direction {
-        HORIZONTAL,
-        VERTICAL
-    }
-
-    static class PolygonVertex {
-
-        Point2D point;
-        Direction direction;
-
-
-        PolygonVertex(Point2D point, Direction direction) {
-
-            this.direction = direction;
-            this.point = point;
-        }
-
-
-        @Override
-        public boolean equals(Object other) {
-
-            if (this == other) {
-                return true;
-            }
-            if (!(other instanceof PolygonVertex)) {
-                return false;
-            }
-            return this.point.equals(((PolygonVertex) other).point);
-        }
-
-
-        @Override
-        public int hashCode() {
-
-            return this.point.hashCode();
-        }
-
-
-        @Override
-        public String toString() {
-
-            return String.format("%s[point=%s,direction=%s]", this.getClass().getName(), this.point.toString(), this.direction.toString());
-        }
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/utils/CohenSutherlandClipping.java

@@ -1,142 +0,0 @@
-/*
- * CohenSutherland.java
- * --------------------
- * (c) 2007 by Intevation GmbH
- *
- * @author Sascha L. Teichmann (teichmann@intevation.de)
- * @author Ludwig Reiter       (ludwig@intevation.de)
- *
- * This program is free software under the LGPL (>=v2.1)
- * Read the file LICENSE.txt coming with the sources for details.
- */
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
-
-import java.awt.geom.Line2D;
-import java.awt.geom.Rectangle2D;
-
-/**
- * Implements the well known Cohen Sutherland line
- * clipping algorithm (line against clip rectangle).
- */
-@SuppressWarnings("all")
-public final class CohenSutherlandClipping {
-
-    private static final int INSIDE = 0;
-    private static final int LEFT = 1;
-    private static final int RIGHT = 2;
-    private static final int BOTTOM = 4;
-    private static final int TOP = 8;
-    private double xMin;
-    private double yMin;
-    private double xMax;
-    private double yMax;
-
-
-    /**
-     * Creates a Cohen Sutherland clipper with clip rect (0, 0, 0, 0).
-     */
-    public CohenSutherlandClipping() {
-
-    }
-
-
-    /**
-     * Creates a Cohen Sutherland clipper with the given clip rectangle.
-     *
-     * @param clip the clip rectangle to use
-     */
-    public CohenSutherlandClipping(Rectangle2D clip) {
-
-        setClip(clip);
-    }
-
-
-    /**
-     * Sets the clip rectangle.
-     *
-     * @param clip the clip rectangle
-     */
-    public void setClip(Rectangle2D clip) {
-
-        xMin = clip.getX();
-        xMax = xMin + clip.getWidth();
-        yMin = clip.getY();
-        yMax = yMin + clip.getHeight();
-    }
-
-
-    private final int regionCode(double x, double y) {
-
-        int code = x < xMin ? LEFT : x > xMax ? RIGHT : INSIDE;
-        if (y < yMin) {
-            code |= BOTTOM;
-        } else if (y > yMax) {
-            code |= TOP;
-        }
-        return code;
-    }
-
-
-    /**
-     * Clips a given line against the clip rectangle.
-     * The modification (if needed) is done in place.
-     *
-     * @param line the line to clip
-     * @return true if line is clipped, false if line is
-     * totally outside the clip rect.
-     */
-    public boolean clip(Line2D.Float line) {
-
-        double p1x = line.getX1();
-        double p1y = line.getY1();
-        double p2x = line.getX2();
-        double p2y = line.getY2();
-
-        double qx = 0d;
-        double qy = 0d;
-
-        boolean vertical = p1x == p2x;
-
-        double slope = vertical ? 0d : (p2y - p1y) / (p2x - p1x);
-
-        int c1 = regionCode(p1x, p1y);
-        int c2 = regionCode(p2x, p2y);
-
-        while (c1 != INSIDE || c2 != INSIDE) {
-
-            if ((c1 & c2) != INSIDE) {
-                return false;
-            }
-
-            int c = c1 == INSIDE ? c2 : c1;
-
-            if ((c & LEFT) != INSIDE) {
-                qx = xMin;
-                qy = (Utils.feq(qx, p1x) ? 0 : qx - p1x) * slope + p1y;
-            } else if ((c & RIGHT) != INSIDE) {
-                qx = xMax;
-                qy = (Utils.feq(qx, p1x) ? 0 : qx - p1x) * slope + p1y;
-            } else if ((c & BOTTOM) != INSIDE) {
-                qy = yMin;
-                qx = vertical ? p1x : (Utils.feq(qy, p1y) ? 0 : qy - p1y) / slope + p1x;
-            } else if ((c & TOP) != INSIDE) {
-                qy = yMax;
-                qx = vertical ? p1x : (Utils.feq(qy, p1y) ? 0 : qy - p1y) / slope + p1x;
-            }
-
-            if (c == c1) {
-                p1x = qx;
-                p1y = qy;
-                c1 = regionCode(p1x, p1y);
-            } else {
-                p2x = qx;
-                p2y = qy;
-                c2 = regionCode(p2x, p2y);
-            }
-        }
-        line.setLine(p1x, p1y, p2x, p2y);
-        return true;
-    }
-
-}
-// end of file

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/utils/FileUtils.java

@@ -1,56 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
-
-import java.io.File;
-import java.io.IOException;
-import java.nio.file.Files;
-
-import lombok.experimental.UtilityClass;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@UtilityClass
-public class FileUtils {
-
-    public File createTempFile(String filenamePrefix, String filenameSuffix) throws IOException {
-
-        File tempFile = Files.createTempFile(filenamePrefix, filenameSuffix).toFile();
-        setRWPermissionsOnlyForOwner(tempFile);
-
-        return tempFile;
-    }
-
-
-    /**
-     * Deletes a file; logs a message with the reason if the deletion fails.
-     * This method is null-safe.
-     *
-     * @param file The file to delete.  Can be null.
-     */
-    public void deleteFile(File file) {
-
-        if (file != null) {
-            try {
-                Files.deleteIfExists(file.toPath());
-            } catch (IOException ex) {
-                log.warn("Could not delete file!", ex);
-            }
-        }
-    }
-
-
-    // We don't need to check the results of the permission setters below,
-    // since we're manipulating a file we created ourselves.
-    @SuppressWarnings({"ResultOfMethodCallIgnored", "squid:S899"})
-    private void setRWPermissionsOnlyForOwner(File tempFile) {
-
-        try {
-            tempFile.setReadable(true, true);
-            tempFile.setWritable(true, true);
-            tempFile.setExecutable(false);
-        } catch (SecurityException ex) {
-            // This should never happen since we're creating a temp file ourselves.
-            log.warn("Caught an exception during temp file creation.  This should not happend.  Check the code.", ex);
-        }
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/utils/PositionUtils.java

@@ -1,121 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
-
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.TextPageBlock;
-
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-@SuppressWarnings("all")
-public final class PositionUtils {
-
-    // TODO This currently uses pdf coord system. In the futher this should use java coord system.
-    // Note: DirAdj (TextDirection Adjusted) can not be user for this.
-    public boolean isWithinBodyTextFrame(Rectangle btf, TextPageBlock textBlock) {
-
-        if (btf == null || textBlock == null) {
-            return false;
-        }
-
-        double threshold = textBlock.getMostPopularWordHeight() * 3;
-
-        if (textBlock.getPdfMinX() + threshold > btf.getTopLeft().getX() && textBlock.getPdfMaxX() - threshold < btf.getTopLeft()
-                .getX() + btf.getWidth() && textBlock.getPdfMinY() + threshold > btf.getTopLeft().getY() && textBlock.getPdfMaxY() - threshold < btf.getTopLeft()
-                .getY() + btf.getHeight()) {
-            return true;
-        } else {
-            return false;
-        }
-
-    }
-
-
-    // TODO This currently uses pdf coord system. In the futher this should use java coord system.
-    // Note: DirAdj (TextDirection Adjusted) can not be user for this.
-    public boolean isOverBodyTextFrame(Rectangle btf, TextPageBlock textBlock, int rotation) {
-
-        if (btf == null || textBlock == null) {
-            return false;
-        }
-
-        if (rotation == 90 && textBlock.getPdfMaxX() < btf.getTopLeft().getX()) {
-            return true;
-        }
-
-        if (rotation == 180 && textBlock.getPdfMaxY() < btf.getTopLeft().getY()) {
-            return true;
-        }
-
-        if (rotation == 270 && textBlock.getPdfMinX() > btf.getTopLeft().getX() + btf.getWidth()) {
-            return true;
-        }
-
-        if (rotation == 0 && textBlock.getPdfMinY() > btf.getTopLeft().getY() + btf.getHeight()) {
-            return true;
-        } else {
-            return false;
-        }
-
-    }
-
-
-    // TODO This currently uses pdf coord system. In the futher this should use java coord system.
-    // Note: DirAdj (TextDirection Adjusted) can not be user for this.
-    public boolean isUnderBodyTextFrame(Rectangle btf, TextPageBlock textBlock, int rotation) {
-
-        if (btf == null || textBlock == null) {
-            return false;
-        }
-
-        if (rotation == 90 && textBlock.getPdfMinX() > btf.getTopLeft().getX() + btf.getWidth()) {
-            return true;
-        }
-
-        if (rotation == 180 && textBlock.getPdfMinY() > btf.getTopLeft().getY() + btf.getHeight()) {
-            return true;
-        }
-
-        if (rotation == 270 && textBlock.getPdfMaxX() < btf.getTopLeft().getX()) {
-            return true;
-        }
-
-        if (rotation == 0 && textBlock.getPdfMaxY() < btf.getTopLeft().getY()) {
-            return true;
-        } else {
-            return false;
-        }
-
-    }
-
-
-    // TODO This currently uses pdf coord system. In the futher this should use java coord system.
-    // Note: DirAdj (TextDirection Adjusted) can not be user for this.
-    public boolean isTouchingUnderBodyTextFrame(Rectangle btf, TextPageBlock textBlock) {
-
-        //TODO Currently this is not working for rotated pages.
-
-        if (btf == null || textBlock == null) {
-            return false;
-        }
-
-        if (textBlock.getMinY() < btf.getTopLeft().getY()) {
-            return true;
-        } else {
-            return false;
-        }
-
-    }
-
-
-    public float getHeightDifferenceBetweenChunkWordAndDocumentWord(TextPageBlock textBlock, Float documentMostPopularWordHeight) {
-
-        return textBlock.getMostPopularWordHeight() - documentMostPopularWordHeight;
-    }
-
-
-    public Float getApproxLineCount(TextPageBlock textBlock) {
-
-        return textBlock.getHeight() / textBlock.getMostPopularWordHeight();
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/utils/QuickSort.java

@@ -1,109 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
-
-import java.util.ArrayDeque;
-import java.util.Comparator;
-import java.util.Deque;
-import java.util.List;
-
-import lombok.experimental.UtilityClass;
-
-/**
- * Copied and minimal modified from PDFBox.
- */
-@UtilityClass
-public final class QuickSort {
-
-    private static final Comparator<? extends Comparable> OBJCOMP = new Comparator<Comparable>() {
-        @Override
-        public int compare(Comparable object1, Comparable object2) {
-
-            return object1.compareTo(object2);
-        }
-    };
-
-
-    /**
-     * Sorts the given list using the given comparator.
-     *
-     * @param <T>  type of the objects to be sorted.
-     * @param list list to be sorted
-     * @param cmp  comparator used to compare the objects within the list
-     */
-    public static <T> void sort(List<T> list, Comparator<? super T> cmp) {
-
-        int size = list.size();
-        if (size < 2) {
-            return;
-        }
-        quicksort(list, cmp);
-    }
-
-
-    /**
-     * Sorts the given list using compareTo as comparator.
-     *
-     * @param <T>  type of the objects to be sorted.
-     * @param list list to be sorted
-     */
-    public static <T extends Comparable> void sort(List<T> list) {
-
-        sort(list, (Comparator<T>) OBJCOMP);
-    }
-
-
-    private static <T> void quicksort(List<T> list, Comparator<? super T> cmp) {
-
-        Deque<Integer> stack = new ArrayDeque<Integer>();
-        stack.push(0);
-        stack.push(list.size());
-        while (!stack.isEmpty()) {
-            int right = stack.pop();
-            int left = stack.pop();
-            if (right - left < 2) {
-                continue;
-            }
-            int p = left + ((right - left) / 2);
-            p = partition(list, cmp, p, left, right);
-
-            stack.push(p + 1);
-            stack.push(right);
-
-            stack.push(left);
-            stack.push(p);
-        }
-    }
-
-
-    private static <T> int partition(List<T> list, Comparator<? super T> cmp, int p, int start, int end) {
-
-        int l = start;
-        int h = end - 2;
-        T piv = list.get(p);
-        swap(list, p, end - 1);
-
-        while (l < h) {
-            if (cmp.compare(list.get(l), piv) <= 0) {
-                l++;
-            } else if (cmp.compare(piv, list.get(h)) <= 0) {
-                h--;
-            } else {
-                swap(list, l, h);
-            }
-        }
-        int idx = h;
-        if (cmp.compare(list.get(h), piv) < 0) {
-            idx++;
-        }
-        swap(list, end - 1, idx);
-        return idx;
-    }
-
-
-    private static <T> void swap(List<T> list, int i, int j) {
-
-        T tmp = list.get(i);
-        list.set(i, list.get(j));
-        list.set(j, tmp);
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/utils/RulingTextDirAdjustUtil.java

@@ -1,64 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
-
-import java.awt.geom.Line2D;
-import java.awt.geom.Point2D;
-
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.table.Ruling;
-
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public final class RulingTextDirAdjustUtil {
-
-    /**
-     * Converts a ruling (line of a table) the same way TextPositions are converted in PDFBox.
-     * This will get the y position of the text, adjusted so that 0,0 is upper left and it is adjusted based on the text direction.
-     * <p>
-     * See org.apache.pdfbox.text.TextPosition
-     */
-    public Line2D.Float convertToDirAdj(Ruling ruling, float dir, float pageWidth, float pageHeight) {
-
-        return new Line2D.Float(convertPoint(ruling.x1, ruling.y1, dir, pageWidth, pageHeight), convertPoint(ruling.x2, ruling.y2, dir, pageWidth, pageHeight));
-    }
-
-
-    private Point2D convertPoint(float x, float y, float dir, float pageWidth, float pageHeight) {
-
-        var xAdj = getXRot(x, y, dir, pageWidth, pageHeight);
-        var yLowerLeftRot = getYLowerLeftRot(x, y, dir, pageWidth, pageHeight);
-        var yAdj = dir == 0 || dir == 180 ? pageHeight - yLowerLeftRot : pageWidth - yLowerLeftRot;
-        return new Point2D.Float(xAdj, yAdj);
-    }
-
-
-    @SuppressWarnings("SuspiciousNameCombination")
-    private float getXRot(float x, float y, float dir, float pageWidth, float pageHeight) {
-
-        if (dir == 0) {
-            return x;
-        } else if (dir == 90) {
-            return y;
-        } else if (dir == 180) {
-            return pageWidth - x;
-        } else if (dir == 270) {
-            return pageHeight - y;
-        }
-        return 0;
-    }
-
-
-    private float getYLowerLeftRot(float x, float y, float dir, float pageWidth, float pageHeight) {
-
-        if (dir == 0) {
-            return y;
-        } else if (dir == 90) {
-            return pageWidth - x;
-        } else if (dir == 180) {
-            return pageHeight - y;
-        } else if (dir == 270) {
-            return x;
-        }
-        return 0;
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/layoutparsing/classification/utils/Utils.java

@@ -1,43 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.layoutparsing.classification.utils;
-
-import java.math.BigDecimal;
-import java.util.Comparator;
-import java.util.List;
-
-import lombok.experimental.UtilityClass;
-import lombok.extern.slf4j.Slf4j;
-
-@Slf4j
-@UtilityClass
-@SuppressWarnings("all")
-public final class Utils {
-
-    private final static float EPSILON = 0.1f;
-
-
-    public static boolean feq(double f1, double f2) {
-
-        return (Math.abs(f1 - f2) < EPSILON);
-    }
-
-
-    public static float round(double d, int decimalPlace) {
-
-        BigDecimal bd = BigDecimal.valueOf(d);
-        bd = bd.setScale(decimalPlace, BigDecimal.ROUND_HALF_UP);
-        return bd.floatValue();
-    }
-
-
-    public static <T> void sort(List<T> list, Comparator<? super T> comparator) {
-
-        try {
-            QuickSort.sort(list, comparator);
-        } catch (IllegalArgumentException e) {
-            // This should not happen since we use QuickSort from PDFBox
-            log.warn(e.getMessage());
-        }
-    }
-
-}
-

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/service/AnalyzeService.java

@@ -14,7 +14,6 @@ import org.springframework.stereotype.Service;
 import org.springframework.web.bind.annotation.RequestBody;
 
 import com.iqser.gin4.commons.metrics.meters.FunctionTimerValues;
-import com.iqser.red.service.dictionarymerge.commons.DictionaryEntryModel;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.AnalyzeRequest;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.AnalyzeResult;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.FileAttribute;
@@ -24,29 +23,15 @@ import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlo
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogChanges;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogEntry;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogLegalBasis;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.section.SectionGrid;
-import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
 import com.iqser.red.service.redaction.v1.server.client.LegalBasisClient;
 import com.iqser.red.service.redaction.v1.server.client.model.NerEntitiesModel;
-import com.iqser.red.service.redaction.v1.server.document.data.DocumentData;
 import com.iqser.red.service.redaction.v1.server.document.data.mapper.DocumentGraphMapper;
-import com.iqser.red.service.redaction.v1.server.document.factory.DocumentGraphFactory;
 import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Section;
 import com.iqser.red.service.redaction.v1.server.document.graph.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.adapter.ImageServiceResponseAdapter;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.ClassificationDocument;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.image.ClassifiedImage;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.SimplifiedSectionText;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.model.text.SimplifiedText;
-import com.iqser.red.service.redaction.v1.server.layoutparsing.classification.service.PdfSegmentationService;
-
 import com.iqser.red.service.redaction.v1.server.redaction.adapter.NerEntities;
 import com.iqser.red.service.redaction.v1.server.redaction.adapter.NerEntitiesAdapter;
 import com.iqser.red.service.redaction.v1.server.redaction.model.dictionary.Dictionary;
 import com.iqser.red.service.redaction.v1.server.redaction.model.dictionary.DictionaryIncrement;
-import com.iqser.red.service.redaction.v1.server.redaction.model.dictionary.DictionaryModel;
 import com.iqser.red.service.redaction.v1.server.redaction.model.dictionary.DictionaryVersion;
 import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
 import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
@@ -71,63 +56,15 @@ public class AnalyzeService {
     EntityRedactionService entityRedactionService;
     RedactionLogCreatorService redactionLogCreatorService;
     RedactionStorageService redactionStorageService;
-    PdfSegmentationService pdfSegmentationService;
     RedactionChangeLogService redactionChangeLogService;
     LegalBasisClient legalBasisClient;
     RedactionServiceSettings redactionServiceSettings;
-    SectionGridCreatorService sectionGridCreatorService;
-    ImageServiceResponseAdapter imageServiceResponseAdapter;
     ImportedRedactionService importedRedactionService;
     SectionFinderService sectionFinderService;
 
     FunctionTimerValues redactmanagerAnalyzePagewiseValues;
 
 
-    @Timed("redactmanager_analyzeDocumentStructure")
-    public AnalyzeResult analyzeDocumentStructure(StructureAnalyzeRequest analyzeRequest) {
-
-        long startTime = System.currentTimeMillis();
-
-        ClassificationDocument classifiedDoc;
-
-        try {
-            var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(analyzeRequest.getDossierId(),
-                    analyzeRequest.getFileId(),
-                    FileType.ORIGIN));
-            log.info("Loaded PDF for file {} in dossier {}", analyzeRequest.getFileId(), analyzeRequest.getDossierId());
-            Map<Integer, List<ClassifiedImage>> pdfImages = null;
-            if (redactionServiceSettings.isEnableImageClassification()) {
-                pdfImages = imageServiceResponseAdapter.convertImages(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
-                log.info("Loaded image service response for file {} in dossier {}", analyzeRequest.getFileId(), analyzeRequest.getDossierId());
-            }
-
-            classifiedDoc = pdfSegmentationService.parseDocument(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), storedObjectStream, pdfImages);
-            log.info("Parsed document for file {} in dossier {}", analyzeRequest.getFileId(), analyzeRequest.getDossierId());
-        } catch (Exception e) {
-            throw new RedactionException(e);
-        }
-
-        Document document = DocumentGraphFactory.buildDocumentGraph(classifiedDoc);
-        log.info("Built Document Graph for file {} in dossier {}", analyzeRequest.getFileId(), analyzeRequest.getDossierId());
-
-        SectionGrid sectionGrid = sectionGridCreatorService.createSectionGrid(document);
-        log.info("Built section grid for file {} in dossier {}", analyzeRequest.getFileId(), analyzeRequest.getDossierId());
-
-        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.TEXT, DocumentData.fromDocument(document));
-        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.SIMPLIFIED_TEXT, toSimplifiedText(document));
-        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.SECTION_GRID, sectionGrid);
-        log.info("Stored document graph, text, simplified text, and section grid for file {} in dossier {}", analyzeRequest.getFileId(), analyzeRequest.getDossierId());
-
-        return AnalyzeResult.builder()
-                .dossierId(analyzeRequest.getDossierId())
-                .fileId(analyzeRequest.getFileId())
-                .duration(System.currentTimeMillis() - startTime)
-                .numberOfPages(document.getNumberOfPages())
-                .analysisVersion(redactionServiceSettings.getAnalysisVersion())
-                .build();
-    }
-
-
     @Timed("redactmanager_analyze")
     public AnalyzeResult analyze(AnalyzeRequest analyzeRequest) {
 
@@ -367,18 +304,4 @@ public class AnalyzeService {
             }));
         }
     }
-
-
-    private SimplifiedText toSimplifiedText(Document document) {
-
-        List<SimplifiedSectionText> simplifiedSectionTexts = document.getMainSections().stream().map(this::toSimplifiedSectionText).toList();
-        return SimplifiedText.builder().numberOfPages(document.getNumberOfPages()).sectionTexts(simplifiedSectionTexts).build();
-    }
-
-
-    private SimplifiedSectionText toSimplifiedSectionText(Section section) {
-
-        return SimplifiedSectionText.builder().sectionNumber(section.getTreeId().get(0)).text(section.getTextBlock().getSearchText()).build();
-    }
-
 }

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/service/SectionGridCreatorService.java

@@ -1,145 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.redaction.service;
-
-import java.awt.geom.Rectangle2D;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-import java.util.Objects;
-import java.util.Set;
-import java.util.function.BiConsumer;
-import java.util.function.BinaryOperator;
-import java.util.function.Function;
-import java.util.function.Supplier;
-import java.util.stream.Collector;
-import java.util.stream.Stream;
-
-import org.springframework.stereotype.Service;
-
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Point;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.section.CellRectangle;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.section.SectionGrid;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.section.SectionRectangle;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.Table;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.TableCell;
-
-import lombok.RequiredArgsConstructor;
-
-@Service
-@RequiredArgsConstructor
-public class SectionGridCreatorService {
-
-    public SectionGrid createSectionGrid(Document document) {
-
-        Map<Integer, List<SectionRectangle>> sectionBBox = document.streamAllSubNodesOfType(NodeType.SECTION).map(SemanticNode::getBBox).collect(new SectionGridCollector());
-        Map<Integer, List<SectionRectangle>> paragraphBBox = document.streamAllSubNodesOfType(NodeType.PARAGRAPH).map(SemanticNode::getBBox).collect(new SectionGridCollector());
-        Map<Integer, List<SectionRectangle>> headlineBBox = document.streamAllSubNodesOfType(NodeType.HEADLINE).map(SemanticNode::getBBox).collect(new SectionGridCollector());
-        Map<Integer, List<SectionRectangle>> tableBBox = document.streamAllSubNodesOfType(NodeType.TABLE).map(node -> (Table) node).collect(new TableGridCollector());
-        var sectionGrid = new SectionGrid();
-
-        sectionGrid.setRectanglesPerPage(mergeMapsByConcatenatingLists(//
-                mergeMapsByConcatenatingLists(paragraphBBox, headlineBBox), //
-                mergeMapsByConcatenatingLists(sectionBBox, tableBBox)));
-        return sectionGrid;
-    }
-
-
-    private static abstract class GridCollector<T> implements Collector<T, Map<Integer, List<SectionRectangle>>, Map<Integer, List<SectionRectangle>>> {
-
-        @Override
-        public Supplier<Map<Integer, List<SectionRectangle>>> supplier() {
-
-            return HashMap::new;
-        }
-
-
-        @Override
-        public Function<Map<Integer, List<SectionRectangle>>, Map<Integer, List<SectionRectangle>>> finisher() {
-
-            return Function.identity();
-        }
-
-
-        @Override
-        public BinaryOperator<Map<Integer, List<SectionRectangle>>> combiner() {
-
-            return SectionGridCreatorService::mergeMapsByConcatenatingLists;
-        }
-
-
-        @Override
-        public Set<Characteristics> characteristics() {
-
-            return Set.of(Characteristics.IDENTITY_FINISH, Characteristics.CONCURRENT, Characteristics.UNORDERED);
-        }
-
-    }
-
-    private static class TableGridCollector extends GridCollector<Table> {
-
-        @Override
-        public BiConsumer<Map<Integer, List<SectionRectangle>>, Table> accumulator() {
-
-            return (map, table) -> table.getPages()
-                    .forEach(page -> map.merge(page.getNumber(), List.of(toSectionRectangle(table, page, table.getPages().size())), SectionGridCreatorService::concatLists));
-        }
-
-
-        private static SectionRectangle toSectionRectangle(Table table, Page page, int numberOfParts) {
-
-            Rectangle2D rect = table.getBBox().get(page);
-            List<CellRectangle> tableCellRectangles = table.streamTableCells()
-                    .map(TableCell::getBBox)
-                    .map(map -> map.get(page))
-                    .filter(Objects::nonNull)
-                    .map(rectangle2D -> new CellRectangle(new Point((float) rectangle2D.getX(), (float) rectangle2D.getY()),
-                            (float) rectangle2D.getWidth(),
-                            (float) rectangle2D.getHeight()))
-                    .toList();
-            return new SectionRectangle(new Point((float) rect.getX(), (float) rect.getY()),
-                    (float) rect.getWidth(),
-                    (float) rect.getHeight(),
-                    1,
-                    numberOfParts,
-                    tableCellRectangles);
-        }
-
-    }
-
-    private static class SectionGridCollector extends GridCollector<Map<Page, Rectangle2D>> {
-
-        @Override
-        public BiConsumer<Map<Integer, List<SectionRectangle>>, Map<Page, Rectangle2D>> accumulator() {
-
-            return (mapToKeep, mapToMerge) -> mapToMerge.forEach((page, rectangle) -> mapToKeep.merge(page.getNumber(),
-                    List.of(toSectionRectangle(rectangle, mapToMerge.values().size())),
-                    SectionGridCreatorService::concatLists));
-
-        }
-
-
-        private static SectionRectangle toSectionRectangle(Rectangle2D rect, int numberOfParts) {
-
-            return new SectionRectangle(new Point((float) rect.getX(), (float) rect.getY()), (float) rect.getWidth(), (float) rect.getHeight(), 1, numberOfParts, null);
-        }
-
-    }
-
-
-    private static Map<Integer, List<SectionRectangle>> mergeMapsByConcatenatingLists(Map<Integer, List<SectionRectangle>> mapToKeep,
-                                                                                      Map<Integer, List<SectionRectangle>> mapToMerge) {
-
-        mapToMerge.forEach((page, rectangle) -> mapToKeep.merge(page, rectangle, SectionGridCreatorService::concatLists));
-        return mapToKeep;
-    }
-
-
-    private static List<SectionRectangle> concatLists(List<SectionRectangle> l1, List<SectionRectangle> l2) {
-
-        return Stream.concat(l1.stream(), l2.stream()).toList();
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/storage/RedactionStorageService.java

@@ -13,6 +13,10 @@ import com.iqser.red.service.redaction.v1.server.exception.NotFoundException;
 import com.iqser.red.service.redaction.v1.server.document.data.DocumentData;
 import com.iqser.red.storage.commons.exception.StorageObjectDoesNotExist;
 import com.iqser.red.storage.commons.service.StorageService;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentPage;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentPositionData;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentStructure;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentTextData;
 import com.knecon.fforesight.tenantcommons.TenantContext;
 
 import io.micrometer.core.annotation.Timed;
@@ -82,7 +86,12 @@ public class RedactionStorageService {
     public DocumentData getDocumentData(String dossierId, String fileId) {
 
         try {
-            return storageService.readJSONObject(TenantContext.getTenantId(), StorageIdUtils.getStorageId(dossierId, fileId, FileType.TEXT), DocumentData.class);
+            return DocumentData.builder()
+                    .documentStructure(storageService.readJSONObject(TenantContext.getTenantId(), StorageIdUtils.getStorageId(dossierId, fileId, FileType.DOCUMENT_STRUCTURE), DocumentStructure.class))
+                    .documentTextData(storageService.readJSONObject(TenantContext.getTenantId(), StorageIdUtils.getStorageId(dossierId, fileId, FileType.DOCUMENT_TEXT), DocumentTextData[].class))
+                    .documentPositionData(storageService.readJSONObject(TenantContext.getTenantId(), StorageIdUtils.getStorageId(dossierId, fileId, FileType.DOCUMENT_POSITION), DocumentPositionData[].class))
+                    .documentPages(storageService.readJSONObject(TenantContext.getTenantId(), StorageIdUtils.getStorageId(dossierId, fileId, FileType.DOCUMENT_PAGES), DocumentPage[].class))
+                    .build();
         } catch (StorageObjectDoesNotExist e) {
             log.debug("DocumentData not available.");
             return null;

redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/AbstractRedactionIntegrationTest.java

@@ -13,6 +13,7 @@ import java.util.HashMap;
 import java.util.HashSet;
 import java.util.List;
 import java.util.Map;
+import java.util.Optional;
 import java.util.Set;
 import java.util.stream.Collectors;
 
@@ -28,6 +29,7 @@ import com.fasterxml.jackson.databind.ObjectMapper;
 import com.iqser.red.service.dictionarymerge.commons.DictionaryEntry;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.AnalyzeRequest;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.configuration.Colors;
+
 import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.dossier.file.FileType;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.type.Type;
 import com.iqser.red.service.redaction.v1.server.annotate.AnnotationService;
@@ -41,6 +43,10 @@ import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
 import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
 import com.iqser.red.storage.commons.service.StorageService;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingFinishedEvent;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingRequest;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
+import com.knecon.fforesight.service.layoutparser.processor.LayoutParsingPipeline;
 import com.knecon.fforesight.tenantcommons.TenantContext;
 import com.knecon.fforesight.tenantcommons.TenantsClient;
 
@@ -112,6 +118,9 @@ public abstract class AbstractRedactionIntegrationTest {
     @Autowired
     protected ManualRedactionSurroundingTextService manualRedactionSurroundingTextService;
 
+    @Autowired
+    private LayoutParsingPipeline layoutParsingPipeline;
+
     @MockBean
     protected AmazonS3 amazonS3;
 
@@ -437,6 +446,32 @@ public abstract class AbstractRedactionIntegrationTest {
     }
 
 
+    @SneakyThrows
+    protected LayoutParsingFinishedEvent analyzeDocumentStructure(LayoutParsingType layoutParsingType, AnalyzeRequest request) {
+
+        var originId = RedactionStorageService.StorageIdUtils.getStorageId(request.getDossierId(), request.getFileId(), FileType.ORIGIN);
+        var tableId = RedactionStorageService.StorageIdUtils.getStorageId(request.getDossierId(), request.getFileId(), FileType.TABLES);
+        var imageId = RedactionStorageService.StorageIdUtils.getStorageId(request.getDossierId(), request.getFileId(), FileType.IMAGE_INFO);
+        var sectionGridId = RedactionStorageService.StorageIdUtils.getStorageId(request.getDossierId(), request.getFileId(), FileType.SECTION_GRID);
+        var structureId = RedactionStorageService.StorageIdUtils.getStorageId(request.getDossierId(), request.getFileId(), FileType.DOCUMENT_STRUCTURE);
+        var textId = RedactionStorageService.StorageIdUtils.getStorageId(request.getDossierId(), request.getFileId(), FileType.DOCUMENT_TEXT);
+        var positionID = RedactionStorageService.StorageIdUtils.getStorageId(request.getDossierId(), request.getFileId(), FileType.DOCUMENT_POSITION);
+        var pagesID = RedactionStorageService.StorageIdUtils.getStorageId(request.getDossierId(), request.getFileId(), FileType.DOCUMENT_PAGES);
+        var layoutParsingRequest = LayoutParsingRequest.builder()
+                .layoutParsingType(layoutParsingType)
+                .originFileStorageId(originId)
+                .tablesFileStorageId(Optional.of(tableId))
+                .imagesFileStorageId(Optional.of(imageId))
+                .structureFileStorageId(structureId)
+                .textBlockFileStorageId(textId)
+                .positionBlockFileStorageId(positionID)
+                .pageFileStorageId(pagesID)
+                .sectionGridStorageId(sectionGridId)
+                .build();
+        return layoutParsingPipeline.parseLayoutAndSaveFilesToStorage(layoutParsingRequest);
+    }
+
+
     @SneakyThrows
     protected AnalyzeRequest prepareStorage(String file, String cvServiceResponseFile) {
 

redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/DocumineFloraTest.java

@@ -25,12 +25,13 @@ import com.iqser.red.service.persistence.service.v1.api.shared.model.AnalyzeRequ
 import com.iqser.red.service.persistence.service.v1.api.shared.model.AnalyzeResult;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.common.JSONPrimitive;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.type.Type;
-import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
 import com.iqser.red.service.redaction.v1.server.annotate.AnnotateRequest;
 import com.iqser.red.service.redaction.v1.server.annotate.AnnotateResponse;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.OsUtils;
 import com.iqser.red.storage.commons.StorageAutoConfiguration;
 import com.iqser.red.storage.commons.service.StorageService;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
+import com.knecon.fforesight.service.layoutparser.processor.LayoutParsingServiceProcessorConfiguration;
 import com.knecon.fforesight.tenantcommons.TenantContext;
 
 @ExtendWith(SpringExtension.class)
@@ -51,7 +52,7 @@ public class DocumineFloraTest extends AbstractRedactionIntegrationTest {
 //                "files/Documine/Flora/ProblemDocs/SOLICITA_VICTRATO-GOLD-II_Item 21_Mutacao_Genica (1).TABLES.json");
 
         System.out.println("Start Full integration test");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.DOCUMINE, request);
         System.out.println("Finished structure analysis");
         AnalyzeResult result = analyzeService.analyze(request);
         System.out.println("Finished analysis");
@@ -77,7 +78,7 @@ public class DocumineFloraTest extends AbstractRedactionIntegrationTest {
                 "files/Documine/Flora/A8591B/2009773e2e05919bb9e46aeedcc8b924.ORIGIN.TABLES.json");
 
         System.out.println("Start Full integration test");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.DOCUMINE, request);
         System.out.println("Finished structure analysis");
         AnalyzeResult result = analyzeService.analyze(request);
         System.out.println("Finished analysis");
@@ -104,7 +105,7 @@ public class DocumineFloraTest extends AbstractRedactionIntegrationTest {
         AnalyzeRequest request = uploadFileToStorage("files/Documine/Flora/ProblemDocs/SOLICITA_VICTRATO-GOLD-II_Item 15_Toxicidade Oral Aguda.pdf");
 
         System.out.println("Start Full integration test");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.DOCUMINE, request);
         System.out.println("Finished structure analysis");
         AnalyzeResult result = analyzeService.analyze(request);
         System.out.println("Finished analysis");
@@ -123,6 +124,7 @@ public class DocumineFloraTest extends AbstractRedactionIntegrationTest {
 
     @Configuration
     @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class})
+    @Import(LayoutParsingServiceProcessorConfiguration.class)
     @ComponentScan(excludeFilters = {@ComponentScan.Filter(type = FilterType.ASSIGNABLE_TYPE, value = StorageAutoConfiguration.class)})
     static class RedactionIntegrationTestConfiguration {
 

redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/HeadlinesGoldStandardIntegrationTest.java

@@ -1,217 +0,0 @@
-package com.iqser.red.service.redaction.v1.server;
-
-import java.io.InputStream;
-import java.time.OffsetDateTime;
-import java.util.ArrayList;
-import java.util.HashSet;
-import java.util.List;
-import java.util.Set;
-import java.util.stream.Collectors;
-
-import org.assertj.core.api.Assertions;
-import org.junit.jupiter.api.AfterEach;
-import org.junit.jupiter.api.Test;
-import org.junit.jupiter.api.extension.ExtendWith;
-import org.springframework.beans.factory.annotation.Autowired;
-import org.springframework.boot.autoconfigure.EnableAutoConfiguration;
-import org.springframework.boot.autoconfigure.amqp.RabbitAutoConfiguration;
-import org.springframework.boot.test.context.SpringBootTest;
-import org.springframework.boot.test.mock.mockito.MockBean;
-import org.springframework.context.annotation.Bean;
-import org.springframework.context.annotation.ComponentScan;
-import org.springframework.context.annotation.Configuration;
-import org.springframework.context.annotation.FilterType;
-import org.springframework.context.annotation.Import;
-import org.springframework.context.annotation.Primary;
-import org.springframework.core.io.ClassPathResource;
-import org.springframework.test.context.junit.jupiter.SpringExtension;
-
-import com.fasterxml.jackson.databind.ObjectMapper;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.AnalyzeRequest;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.dossier.file.FileType;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.ChangeType;
-import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLog;
-import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
-import com.iqser.red.service.redaction.v1.server.document.data.mapper.DocumentGraphMapper;
-import com.iqser.red.service.redaction.v1.server.document.graph.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.redaction.service.AnalyzeService;
-import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
-import com.iqser.red.storage.commons.StorageAutoConfiguration;
-import com.iqser.red.storage.commons.service.StorageService;
-import com.knecon.fforesight.tenantcommons.TenantContext;
-import com.knecon.fforesight.tenantcommons.TenantsClient;
-
-import lombok.AllArgsConstructor;
-import lombok.Data;
-import lombok.EqualsAndHashCode;
-import lombok.SneakyThrows;
-import lombok.ToString;
-
-@ExtendWith(SpringExtension.class)
-@SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT)
-@Import(HeadlinesGoldStandardIntegrationTest.RedactionIntegrationTestConfiguration.class)
-public class HeadlinesGoldStandardIntegrationTest {
-
-    @MockBean
-    private TenantsClient tenantsClient;
-
-    @Autowired
-    private AnalyzeService analyzeService;
-
-    @Autowired
-    private ObjectMapper objectMapper;
-
-    @Autowired
-    private RedactionStorageService redactionStorageService;
-
-    @Autowired
-    private StorageService storageService;
-
-    private final static String TEST_DOSSIER_TEMPLATE_ID = "123";
-    private final static String TEST_DOSSIER_ID = "123";
-    private final static String TEST_FILE_ID = "123";
-
-
-    @Test
-    public void testHeadlineDetection() {
-
-        List<Metrics> metrics = new ArrayList<>();
-        metrics.add(getMetrics("files/RSS/01 - CGA100251 - Acute Oral Toxicity (Up and Down Procedure) - Rat (1).pdf",
-                "files/Headlines/01 - CGA100251 - Acute Oral Toxicity (Up and Down Procedure) - Rat (1)_REDACTION_LOG.json"));
-        metrics.add(getMetrics("files/Trinexapac/91 Trinexapac-ethyl_RAR_01_Volume_1_2018-02-23.pdf",
-                "files/Headlines/91 Trinexapac-ethyl_RAR_01_Volume_1_2018-02-23_REDACTION_LOG.json"));
-        metrics.add(getMetrics("files/Metolachlor/S-Metolachlor_RAR_01_Volume_1_2018-09-06.pdf", "files/Headlines/S-Metolachlor_RAR_01_Volume_1_2018-09-06_REDACTION_LOG.json"));
-
-        double precision = metrics.stream().mapToDouble(Metrics::getPrecision).average().orElse(1.0);
-        double recall = metrics.stream().mapToDouble(Metrics::getRecall).average().orElse(1.0);
-
-        System.out.println("Precision is: " + precision + " recall is: " + recall);
-
-        Assertions.assertThat(precision).isGreaterThanOrEqualTo(0.44f);
-        Assertions.assertThat(recall).isGreaterThanOrEqualTo(0.69f);
-    }
-
-
-    @SneakyThrows
-    private Metrics getMetrics(String fileUrl, String redactionLogUrl) {
-
-        ClassPathResource redactionLogResource = new ClassPathResource(redactionLogUrl);
-
-        Set<Headline> goldStandardHeadlines = new HashSet<>();
-        var goldStandardLog = objectMapper.readValue(redactionLogResource.getInputStream(), RedactionLog.class);
-        goldStandardLog.getRedactionLogEntry().removeIf(r -> !r.isRedacted() || r.getChanges().get(r.getChanges().size() - 1).getType().equals(ChangeType.REMOVED));
-        goldStandardLog.getRedactionLogEntry().forEach(e -> goldStandardHeadlines.add(new Headline(e.getPositions().get(0).getPage(), e.getValue())));
-
-        AnalyzeRequest request = prepareStorage(fileUrl);
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
-
-        var documentGraph = DocumentGraphMapper.toDocumentGraph(redactionStorageService.getDocumentData(TEST_DOSSIER_ID, TEST_FILE_ID));
-        var foundHeadlines = documentGraph.streamAllSubNodes()
-                .map(SemanticNode::getHeadline)
-                .distinct()
-                .map(headlineNode -> new Headline(headlineNode.getPages().stream().findFirst().get().getNumber(), headlineNode.getTextBlock().getSearchText().stripTrailing()))
-                .toList();
-
-        Set<Headline> correct = new HashSet<>();
-        Set<Headline> missing;
-        Set<Headline> falsePositive = new HashSet<>();
-        for (Headline headline : foundHeadlines) {
-            if (goldStandardHeadlines.contains(headline)) {
-                correct.add(headline);
-            } else {
-                falsePositive.add(headline);
-            }
-        }
-
-        missing = goldStandardHeadlines.stream().filter(h -> !correct.contains(h)).collect(Collectors.toSet());
-
-        float precision = (float) correct.size() / (float) foundHeadlines.size();
-        float recall = (float) correct.size() / ((float) correct.size() + (float) missing.size());
-
-        return new Metrics(precision, recall);
-    }
-
-
-    @Configuration
-    @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class/*, StorageAutoConfiguration.class*/})
-    @ComponentScan(excludeFilters = {@ComponentScan.Filter(type = FilterType.ASSIGNABLE_TYPE, value = StorageAutoConfiguration.class)})
-    public static class RedactionIntegrationTestConfiguration {
-
-        @Bean
-        @Primary
-        public StorageService inmemoryStorage() {
-
-            return new FileSystemBackedStorageService();
-        }
-
-    }
-
-
-    @AfterEach
-    public void cleanupStorage() {
-
-        if (this.storageService instanceof FileSystemBackedStorageService) {
-            ((FileSystemBackedStorageService) this.storageService).clearStorage();
-        }
-    }
-
-
-    @SneakyThrows
-    private AnalyzeRequest prepareStorage(String file) {
-
-        return prepareStorage(file, "files/cv_service_empty_response.json");
-    }
-
-
-    @SneakyThrows
-    private AnalyzeRequest prepareStorage(String file, String cvServiceResponseFile) {
-
-        ClassPathResource pdfFileResource = new ClassPathResource(file);
-        ClassPathResource cvServiceResponseFileResource = new ClassPathResource(cvServiceResponseFile);
-
-        return prepareStorage(pdfFileResource.getInputStream(), cvServiceResponseFileResource.getInputStream());
-    }
-
-
-    @SneakyThrows
-    private AnalyzeRequest prepareStorage(InputStream fileStream, InputStream cvServiceResponseFileStream) {
-
-        AnalyzeRequest request = AnalyzeRequest.builder()
-                .dossierTemplateId(TEST_DOSSIER_TEMPLATE_ID)
-                .dossierId(TEST_DOSSIER_ID)
-                .fileId(TEST_FILE_ID)
-                .lastProcessed(OffsetDateTime.now())
-                .build();
-
-        storageService.storeObject(TenantContext.getTenantId(),
-                RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.TABLES),
-                cvServiceResponseFileStream);
-        storageService.storeObject(TenantContext.getTenantId(), RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.ORIGIN), fileStream);
-
-        return request;
-    }
-
-
-    @Data
-    @EqualsAndHashCode
-    @AllArgsConstructor
-    @ToString
-    private class Metrics {
-
-        private float precision;
-        private float recall;
-
-    }
-
-    @Data
-    @EqualsAndHashCode
-    @AllArgsConstructor
-    @ToString
-    private class Headline {
-
-        private int page;
-        private String headline;
-
-    }
-
-}

redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/RedactionAcceptanceTest.java

@@ -1,7 +1,6 @@
 package com.iqser.red.service.redaction.v1.server;
 
 import static org.mockito.Mockito.when;
-import static org.wildfly.common.Assert.assertFalse;
 import static org.wildfly.common.Assert.assertTrue;
 
 import java.io.FileOutputStream;
@@ -34,12 +33,13 @@ import com.iqser.red.service.persistence.service.v1.api.shared.model.common.JSON
 import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.type.Type;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLog;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogEntry;
-import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
 import com.iqser.red.service.redaction.v1.server.annotate.AnnotateRequest;
 import com.iqser.red.service.redaction.v1.server.annotate.AnnotateResponse;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.OsUtils;
 import com.iqser.red.storage.commons.StorageAutoConfiguration;
 import com.iqser.red.storage.commons.service.StorageService;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
+import com.knecon.fforesight.service.layoutparser.processor.LayoutParsingServiceProcessorConfiguration;
 import com.knecon.fforesight.tenantcommons.TenantContext;
 
 @ExtendWith(SpringExtension.class)
@@ -51,6 +51,7 @@ public class RedactionAcceptanceTest extends AbstractRedactionIntegrationTest {
 
     @Configuration
     @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class})
+    @Import(LayoutParsingServiceProcessorConfiguration.class)
     @ComponentScan(excludeFilters = {@ComponentScan.Filter(type = FilterType.ASSIGNABLE_TYPE, value = StorageAutoConfiguration.class)})
     public static class RedactionIntegrationTestConfiguration {
 
@@ -101,7 +102,7 @@ public class RedactionAcceptanceTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage("files/new/SYNGENTA_EFSA_sanitisation_GFL_v1_moreSections.pdf");
         System.out.println("Start Full integration test");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         System.out.println("Finished structure analysis");
         AnalyzeResult result = analyzeService.analyze(request);
         System.out.println("Finished analysis");
@@ -158,7 +159,7 @@ public class RedactionAcceptanceTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage("files/new/SYNGENTA_EFSA_sanitisation_GFL_v1_moreSections.pdf");
         System.out.println("Start Full integration test");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         System.out.println("Finished structure analysis");
         AnalyzeResult result = analyzeService.analyze(request);
         System.out.println("Finished analysis");

redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/RedactionIntegrationTest.java

@@ -66,7 +66,6 @@ import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlo
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Point;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLog;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogEntry;
-import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
 import com.iqser.red.service.redaction.v1.server.annotate.AnnotateRequest;
 import com.iqser.red.service.redaction.v1.server.annotate.AnnotateResponse;
 import com.iqser.red.service.redaction.v1.server.document.data.mapper.DocumentGraphMapper;
@@ -81,6 +80,8 @@ import com.iqser.red.service.redaction.v1.server.redaction.utils.OsUtils;
 import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
 import com.iqser.red.storage.commons.StorageAutoConfiguration;
 import com.iqser.red.storage.commons.service.StorageService;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
+import com.knecon.fforesight.service.layoutparser.processor.LayoutParsingServiceProcessorConfiguration;
 import com.knecon.fforesight.tenantcommons.TenantContext;
 
 import lombok.SneakyThrows;
@@ -94,8 +95,10 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
     @Autowired
     private EntityEnrichmentService entityEnrichmentService;
 
+
     @Configuration
     @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class})
+    @Import({LayoutParsingServiceProcessorConfiguration.class})
     @ComponentScan(excludeFilters = {@ComponentScan.Filter(type = FilterType.ASSIGNABLE_TYPE, value = StorageAutoConfiguration.class)})
     public static class RedactionIntegrationTestConfiguration {
 
@@ -145,7 +148,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
     public void test270Rotated() {
 
         AnalyzeRequest request = uploadFileToStorage("files/Minimal Examples/270Rotated.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
         assertThat(result).isNotNull();
     }
@@ -156,7 +159,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
     public void testLargeScannedFileOOM() {
 
         AnalyzeRequest request = uploadFileToStorage("scanned/VV-377031.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
         assertThat(result).isNotNull();
     }
@@ -168,7 +171,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = uploadFileToStorage("files/Minimal Examples/merge_images.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         Map<String, List<RedactionLogEntry>> duplicates = new HashMap<>();
@@ -209,7 +212,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         // Lastname M., Doe J., Mustermann M.
         AnalyzeRequest request = uploadFileToStorage("files/Minimal Examples/ExpansionTest.pdf");
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -225,7 +228,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage("files/new/SYNGENTA_EFSA_sanitisation_GFL_v1_moreSections.pdf");
         System.out.println("Start Full integration test");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         System.out.println("Finished structure analysis");
         AnalyzeResult result = analyzeService.analyze(request);
         System.out.println("Finished analysis");
@@ -247,7 +250,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage("files/RSS/01 - CGA100251 - Acute Oral Toxicity (Up and Down Procedure) - Rat (1).pdf");
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         assertThat(result.getAddedFileAttributes().size()).isEqualTo(1);
@@ -264,7 +267,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage("files/new/test-ignore-hint.pdf");
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -298,8 +301,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         long start = System.currentTimeMillis();
         System.out.println("noExceptionShouldBeThrownForAnyFiles");
-        ClassLoader loader = getClass().getClassLoader();
+        URL url = Thread.currentThread().getContextClassLoader().getResource("files");
-        URL url = loader.getResource("files");
         Path path = Paths.get(URI.create(url.toString()));
         when(dictionaryClient.getDictionaryForType(anyString(), anyLong())).thenReturn(new Type());
         Files.walk(path)//
@@ -308,7 +310,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
                 .forEach(currentPath -> {
                     AnalyzeRequest request = uploadFileToStorage(currentPath.toString());
                     System.out.println("Redacting file : " + currentPath.getFileName());
-                    analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+                    analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
 
                     long fstart = System.currentTimeMillis();
                     AnalyzeResult result = analyzeService.analyze(request);
@@ -360,7 +362,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
                 .value("true")
                 .build()));
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -414,13 +416,12 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         ManualRedactions manualRedactions = new ManualRedactions();
 
         manualRedactions.setEntriesToAdd(Set.of(ManualRedactionEntry.builder()
-                        .value("Redact")
+                .value("Redact")
-                        .addToDictionary(true)
+                .addToDictionary(true)
-                                .addToDossierDictionary(true)
+                .addToDossierDictionary(true)
-                        .positions(List.of(new Rectangle(new Point(95.96979999999999f, 515.7984f), 19.866899999999987f, 46.953f, 2)
+                .positions(List.of(new Rectangle(new Point(95.96979999999999f, 515.7984f), 19.866899999999987f, 46.953f, 2)))
-                                )).type("dossier_redaction").build()));
+                .type("dossier_redaction")
-
+                .build()));
-
 
         request.setManualRedactions(manualRedactions);
 
@@ -460,7 +461,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage(fileName);
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         analyzeService.analyze(request);
 
         dictionary.get(DICTIONARY_AUTHOR).add("report");
@@ -516,7 +517,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
                 .value("true")
                 .build()));
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -614,7 +615,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         List<Type> types = objectMapper.readValue(typeResource.getInputStream(), typeRefForTypes);
 
         AnalyzeRequest request = uploadFileToStorage("files/new/PublishedInformationTest.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         ManualRedactions manualRedactions = new ManualRedactions();
         manualRedactions.getIdsToRemove()
                 .add(IdRemoval.builder()
@@ -673,7 +674,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = uploadFileToStorage("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -694,7 +695,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         String filePath = "files/new/crafted document.pdf";
         AnalyzeRequest request = uploadFileToStorage(filePath);
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         String testEntityValue1 = "Desiree";
@@ -766,7 +767,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         String tableServiceResponseFile = "files/cv_table_response_VV-511309.json";
 
         AnalyzeRequest request = prepareStorage(fileName, tableServiceResponseFile);
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -790,7 +791,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         String tableServiceResponseFile = "files/cv_table_response_VV-511309.json";
 
         AnalyzeRequest request = prepareStorage(fileName, tableServiceResponseFile);
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -811,7 +812,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = uploadFileToStorage("files/new/unicodeProblem.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -833,7 +834,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = uploadFileToStorage("files/new/RotateTestFile.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -859,7 +860,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = uploadFileToStorage("files/new/RotateTestFileSimple.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -881,7 +882,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = uploadFileToStorage("files/Minimal Examples/NoHeaderTable.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -934,7 +935,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         request.setManualRedactions(manualRedactions);
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -993,7 +994,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage(pdfFile);
         request.setManualRedactions(manualRedactions);
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         manualRedactions.getEntriesToAdd().add(manualRedactionEntry);
@@ -1026,7 +1027,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage("files/Minimal Examples/Phantom Cells.pdf");
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -1046,7 +1047,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage("files/Minimal Examples/sponsor_companies.pdf");
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -1108,7 +1109,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage(pdfFile);
         request.setManualRedactions(manualRedactions);
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         manualRedactions.getEntriesToAdd().add(manualRedactionEntry);
@@ -1163,7 +1164,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = uploadFileToStorage("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());
@@ -1219,7 +1220,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage(pdfFile);
         request.setManualRedactions(manualRedactions);
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder()
@@ -1252,7 +1253,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
                 RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.IMPORTED_REDACTIONS),
                 importedRedactions.getInputStream());
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         AnalyzeResult result = analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -1292,7 +1293,7 @@ public class RedactionIntegrationTest extends AbstractRedactionIntegrationTest {
 
         AnalyzeRequest request = uploadFileToStorage(fileName);
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = annotationService.annotate(AnnotateRequest.builder().dossierId(TEST_DOSSIER_ID).fileId(TEST_FILE_ID).build());

redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/RedactionIntegrationV2Test.java

@@ -24,9 +24,10 @@ import com.iqser.red.service.persistence.service.v1.api.shared.model.common.JSON
 import com.iqser.red.service.persistence.service.v1.api.shared.model.dossiertemplate.type.Type;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Engine;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogEntry;
-import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
 import com.iqser.red.storage.commons.StorageAutoConfiguration;
 import com.iqser.red.storage.commons.service.StorageService;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
+import com.knecon.fforesight.service.layoutparser.processor.LayoutParsingServiceProcessorConfiguration;
 import com.knecon.fforesight.tenantcommons.TenantContext;
 
 import lombok.SneakyThrows;
@@ -40,6 +41,7 @@ public class RedactionIntegrationV2Test extends AbstractRedactionIntegrationTest
 
     @Configuration
     @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class})
+    @Import(LayoutParsingServiceProcessorConfiguration.class)
     @ComponentScan(excludeFilters = {@ComponentScan.Filter(type = FilterType.ASSIGNABLE_TYPE, value = StorageAutoConfiguration.class)})
     static class RedactionIntegrationTestConfiguration {
 
@@ -112,7 +114,7 @@ public class RedactionIntegrationV2Test extends AbstractRedactionIntegrationTest
         when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(3L);
         mockDictionaryCalls(0L);
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -150,7 +152,7 @@ public class RedactionIntegrationV2Test extends AbstractRedactionIntegrationTest
         dictionary.put(DICTIONARY_AUTHOR, List.of(entryAuthorDictionary));
         falsePositive.put(DICTIONARY_PII, List.of("Dr. Alan Miller COMPLETION DATE:"));
 
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER, request);
         analyzeService.analyze(request);
 
         var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);

redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/RulesTest.java

@@ -24,6 +24,7 @@ import java.util.HashMap;
 import java.util.HashSet;
 import java.util.List;
 import java.util.Map;
+import java.util.Optional;
 import java.util.Set;
 import java.util.stream.Collectors;
 
@@ -72,18 +73,22 @@ import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlo
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogComment;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogEntry;
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.RedactionLogLegalBasis;
-import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
 import com.iqser.red.service.redaction.v1.server.client.DictionaryClient;
 import com.iqser.red.service.redaction.v1.server.client.LegalBasisClient;
 import com.iqser.red.service.redaction.v1.server.client.RulesClient;
 import com.iqser.red.service.redaction.v1.server.controller.RedactionController;
-import com.iqser.red.service.redaction.v1.server.redaction.service.ManualRedactionSurroundingTextService;
 import com.iqser.red.service.redaction.v1.server.redaction.service.AnalyzeService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.ManualRedactionSurroundingTextService;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
 import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
 import com.iqser.red.storage.commons.StorageAutoConfiguration;
 import com.iqser.red.storage.commons.service.StorageService;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingFinishedEvent;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingRequest;
+import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
+import com.knecon.fforesight.service.layoutparser.processor.LayoutParsingPipeline;
+import com.knecon.fforesight.service.layoutparser.processor.LayoutParsingServiceProcessorConfiguration;
 import com.knecon.fforesight.tenantcommons.TenantContext;
 import com.knecon.fforesight.tenantcommons.TenantsClient;
 
@@ -237,6 +242,8 @@ public class RulesTest {
     @Autowired
     private RedactionStorageService redactionStorageService;
     @Autowired
+    private LayoutParsingPipeline layoutParsingPipeline;
+    @Autowired
     private StorageService storageService;
     @Autowired
     private ManualRedactionSurroundingTextService manualRedactionSurroundingTextService;
@@ -399,7 +406,7 @@ public class RulesTest {
             loadNerForTest();
 
             AnalyzeRequest request = prepareStorage(fileName);
-            analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+
             analyzeService.analyze(request);
 
             RedactionLog redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -422,7 +429,7 @@ public class RulesTest {
         loadNerForTest();
 
         AnalyzeRequest request = prepareStorage(fileName);
-        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        analyzeDocumentStructure(LayoutParsingType.REDACT_MANAGER);
         analyzeService.analyze(request);
 
         RedactionLog redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
@@ -814,9 +821,35 @@ public class RulesTest {
         }
     }
 
+    @SneakyThrows
+    protected LayoutParsingFinishedEvent analyzeDocumentStructure(LayoutParsingType layoutParsingType) {
+
+        var originId = RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.ORIGIN);
+        var tableId = RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.TABLES);
+        var imageId = RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.IMAGE_INFO);
+        var sectionGridId = RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.SECTION_GRID);
+        var structureId = RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.DOCUMENT_STRUCTURE);
+        var textId = RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.DOCUMENT_TEXT);
+        var positionID = RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.DOCUMENT_POSITION);
+        var pagesID = RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.DOCUMENT_PAGES);
+        var layoutParsingRequest = LayoutParsingRequest.builder()
+                .layoutParsingType(layoutParsingType)
+                .originFileStorageId(originId)
+                .tablesFileStorageId(Optional.of(tableId))
+                .imagesFileStorageId(Optional.of(imageId))
+                .structureFileStorageId(structureId)
+                .textBlockFileStorageId(textId)
+                .positionBlockFileStorageId(positionID)
+                .pageFileStorageId(pagesID)
+                .sectionGridStorageId(sectionGridId)
+                .build();
+        return layoutParsingPipeline.parseLayoutAndSaveFilesToStorage(layoutParsingRequest);
+    }
+
 
     @Configuration
     @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class})
+    @Import(LayoutParsingServiceProcessorConfiguration.class)
     @ComponentScan(excludeFilters = {@ComponentScan.Filter(type = FilterType.ASSIGNABLE_TYPE, value = StorageAutoConfiguration.class)})
     public static class RulesTestConfiguration {