Merge branch 'RM-231-bp-bp' into 'release/0.159.x'

RM-231: missing whitespace in name

See merge request fforesight/layout-parser!265

.gitlab-ci.yml

@@ -21,6 +21,5 @@ deploy:
       dotenv: version.env
   rules:
     - if: $CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH 
-    - if: $CI_COMMIT_BRANCH =~ /^feature/ && $CI_COMMIT_TAG == ""
     - if: $CI_COMMIT_BRANCH =~ /^release/
     - if: $CI_COMMIT_TAG

buildSrc/src/main/kotlin/com.knecon.fforesight.java-conventions.gradle.kts

@@ -8,8 +8,6 @@ plugins {
 
 group = "com.knecon.fforesight"
 
-val documentVersion by rootProject.extra { "4.433.0" }
-
 java.sourceCompatibility = JavaVersion.VERSION_17
 java.targetCompatibility = JavaVersion.VERSION_17
 

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/data/redaction/DocumentData.java

@@ -0,0 +1,28 @@
+package com.knecon.fforesight.service.layoutparser.internal.api.data.redaction;
+
+import java.io.Serializable;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+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)
+@Schema(description = "Object containing the complete document layout parsing information. It is split into 4 categories, structure, text, positions and pages: " + "The document tree structure of SemanticNodes such as Section, Paragraph, Headline, etc. " + "The text, which is stored as separate blocks of data. " + "The text positions, which are also stored as separate blocks. The Blocks are equal to the text blocks in length and order. " + "The page information.")
+public class DocumentData implements Serializable {
+
+    @Schema(description = "Contains information about the document's pages.")
+    DocumentPage[] documentPages;
+    @Schema(description = "Contains information about the document's text.")
+    DocumentTextData[] documentTextData;
+    @Schema(description = "Contains information about the document's text positions.")
+    DocumentPositionData[] documentPositions;
+    @Schema(description = "Contains information about the document's semantic structure.")
+    DocumentStructure documentStructure;
+
+}

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/data/redaction/DocumentPage.java

@@ -0,0 +1,30 @@
+package com.knecon.fforesight.service.layoutparser.internal.api.data.redaction;
+
+import java.io.Serializable;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+@Schema(description = "Object containing information about the document's pages.")
+public class DocumentPage implements Serializable {
+
+    @Schema(description = "The page number, starting with 1.")
+    int number;
+    @Schema(description = "The page height in PDF user units.", example = "792")
+    int height;
+    @Schema(description = "The page width in PDF user units.", example = "694")
+    int width;
+    @Schema(description = "The page rotation as specified by the PDF.", example = "90", allowableValues = {"0", "90", "180", "270"})
+    int rotation;
+
+}

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/data/redaction/DocumentPositionData.java

@@ -0,0 +1,28 @@
+package com.knecon.fforesight.service.layoutparser.internal.api.data.redaction;
+
+import java.io.Serializable;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+@Schema(description = "Object containing text positional information of a specific text block. A document is split into multiple text blocks, which are supposed to be read in order. Every text block can only occur on a single page.")
+public class DocumentPositionData implements Serializable {
+
+    @Schema(description = "Identifier of the text block.")
+    Long id;
+    @Schema(description = "For each string coordinate in the search text of the text block, the array contains an entry relating the string coordinate to the position coordinate. This is required due to the text and position coordinates not being equal.")
+    int[] stringIdxToPositionIdx;
+    @Schema(description = "The bounding box for each glyph as a rectangle. This matrix is of size (n,4), where n is the number of glyphs in the text block. The second dimension specifies the rectangle with the value x, y, width, height, with x, y specifying the lower left corner. In order to access this information, the stringIdxToPositionIdx array must be used to transform the coordinates.")
+    float[][] positions;
+
+}

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/data/redaction/DocumentStructure.java

@@ -0,0 +1,172 @@
+package com.knecon.fforesight.service.layoutparser.internal.api.data.redaction;
+
+import java.awt.geom.Rectangle2D;
+import java.io.Serializable;
+import java.util.Arrays;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Stream;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+@Schema(description = "Object containing information about the parsed tree structure of the SemanticNodes, such as Section, Paragraph, Headline etc inside of the document.")
+public class DocumentStructure implements Serializable {
+
+    @Schema(description = "The root EntryData represents the Document.")
+    EntryData root;
+
+    @Schema(description = "Object containing the extra field names, a table has in its properties field.")
+    public static class TableProperties implements Serializable {
+
+        public static final String NUMBER_OF_ROWS = "numberOfRows";
+        public static final String NUMBER_OF_COLS = "numberOfCols";
+
+    }
+
+    @Schema(description = "Object containing the extra field names, an Image has in its properties field.")
+    public static class ImageProperties implements Serializable {
+
+        public static final String TRANSPARENT = "transparent";
+        public static final String IMAGE_TYPE = "imageType";
+        public static final String POSITION = "position";
+        public static final String ID = "id";
+
+        public static final String REPRESENTATION_HASH = "representationHash";
+
+    }
+
+    @Schema(description = "Object containing the extra field names, a table cell has in its properties field.")
+    public static class TableCellProperties implements Serializable {
+
+        public static final String B_BOX = "bBox";
+        public static final String ROW = "row";
+        public static final String COL = "col";
+        public static final String HEADER = "header";
+
+    }
+
+    @Schema(description = "Object containing the extra field names, a duplicate paragraph has in its properties field.")
+    public static class DuplicateParagraphProperties implements Serializable {
+
+        public static final String UNSORTED_TEXTBLOCK_ID = "utbid";
+
+    }
+
+    public static final String RECTANGLE_DELIMITER = ";";
+
+
+    public static Rectangle2D parseRectangle2D(String bBox) {
+
+        List<Float> floats = Arrays.stream(bBox.split(RECTANGLE_DELIMITER))
+                .map(Float::parseFloat)
+                .toList();
+        return new Rectangle2D.Float(floats.get(0), floats.get(1), floats.get(2), floats.get(3));
+    }
+
+
+    public static double[] parseRepresentationVector(String representationHash) {
+
+        String[] stringArray = representationHash.split("[,\\s]+");
+        double[] doubleArray = new double[stringArray.length];
+        for (int i = 0; i < stringArray.length; i++) {
+            doubleArray[i] = Double.parseDouble(stringArray[i]);
+        }
+
+        return doubleArray;
+    }
+
+
+    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(DocumentStructure::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(DocumentStructure::flatten));
+    }
+
+
+    @Data
+    @Builder
+    @NoArgsConstructor
+    @AllArgsConstructor
+    @FieldDefaults(level = AccessLevel.PRIVATE)
+    @Schema(description = "Object containing information of a SemanticNode and also structuring the layout with children.")
+    public static class EntryData implements Serializable {
+
+        @Schema(description = "Type of the semantic node.", allowableValues = {"DOCUMENT", "SECTION", "PARAGRAPH", "HEADLINE", "TABLE", "TABLE_CELL", "HEADER", "FOOTER", "IMAGE"})
+        NodeType type;
+        @Schema(description = "Specifies the position in the parsed tree structure.", example = "[1, 0, 2]")
+        int[] treeId;
+        @Schema(description = "Specifies the text block IDs associated with this semantic node. The value should be joined with the DocumentTextData/DocumentPositionData. Is empty, if no text block is directly associated with this semantic node. Only Paragraph, Headline, Header or Footer is directly associated with a text block.", example = "[1]")
+        Long[] atomicBlockIds;
+        @Schema(description = "Specifies the pages this semantic node appears on. The value should be joined with the PageData.", example = "[1, 2, 3]")
+        Long[] pageNumbers;
+        @Schema(description = "Some semantic nodes have additional information, this information is stored in this Map. The extra fields are specified by the Properties subclasses.", example = "For a Table: {\"numberOfRows\": 3, \"numberOfCols\": 4}")
+        Map<String, String> properties;
+        @Schema(description = "All child Entries of this Entry.", example = "[1, 2, 3]")
+        List<EntryData> children;
+        @Schema(description = "Describes the origin of the semantic node", example = "[ALGORITHM]")
+        Set<LayoutEngine> engines;
+
+
+        @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();
+        }
+
+    }
+
+}

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/data/redaction/DocumentTextData.java

@@ -0,0 +1,36 @@
+package com.knecon.fforesight.service.layoutparser.internal.api.data.redaction;
+
+import java.io.Serializable;
+
+import io.swagger.v3.oas.annotations.media.Schema;
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+@Schema(description = "Object containing text information of a specific text block. A document is split into multiple text blocks, which are supposed to be read in order. Every text block can only occur on a single page.")
+public class DocumentTextData implements Serializable {
+
+    @Schema(description = "Identifier of the text block.")
+    Long id;
+    @Schema(description = "The page the text block occurs on.")
+    Long page;
+    @Schema(description = "The text the text block.")
+    String searchText;
+    @Schema(description = "Each text block is assigned a number on a page, starting from 0.")
+    int numberOnPage;
+    @Schema(description = "The text blocks are ordered, this number represents the start of the text block as a string offset.")
+    int start;
+    @Schema(description = "The text blocks are ordered, this number represents the end of the text block as a string offset.")
+    int end;
+    @Schema(description = "The line breaks in the text of this semantic node in string offsets. They are exclusive end. At the end of each semantic node there is an implicit linebreak.", example = "[5, 10]")
+    int[] lineBreaks;
+
+}

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/data/redaction/LayoutEngine.java

@@ -0,0 +1,7 @@
+package com.knecon.fforesight.service.layoutparser.internal.api.data.redaction;
+
+public enum LayoutEngine {
+    ALGORITHM,
+    AI,
+    OUTLINE
+}

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/data/redaction/NodeType.java

@@ -0,0 +1,23 @@
+package com.knecon.fforesight.service.layoutparser.internal.api.data.redaction;
+
+import java.io.Serializable;
+import java.util.Locale;
+
+public enum NodeType implements Serializable {
+    DOCUMENT,
+    SECTION,
+    SUPER_SECTION,
+    HEADLINE,
+    PARAGRAPH,
+    TABLE,
+    TABLE_CELL,
+    IMAGE,
+    HEADER,
+    FOOTER;
+
+
+    public String toString() {
+
+        return this.name().charAt(0) + this.name().substring(1).toLowerCase(Locale.ROOT);
+    }
+}

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/data/redaction/SimplifiedText.java

@@ -21,14 +21,5 @@ public class SimplifiedText {
     @Schema(description = "A List of simplified Sections, which contains almost exclusively text.")
     @Builder.Default
     private List<SimplifiedSectionText> sectionTexts = new ArrayList<>();
-    @Schema(description = "A list of the main section numbers ")
-    @Builder.Default
-    private List<String> mainSectionNumbers = new ArrayList<>();
-    @Schema(description = "A list of the header section numbers ")
-    @Builder.Default
-    private List<String> headerSectionNumbers = new ArrayList<>();
-    @Schema(description = "A list of the footer section numbers ")
-    @Builder.Default
-    private List<String> footerSectionNumbers = new ArrayList<>();
 
 }

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/queue/LayoutParsingFinishedEvent.java

@@ -8,20 +8,13 @@ import lombok.Builder;
 @Builder
 @Schema(description = "Object containing information about the layout parsing.")
 public record LayoutParsingFinishedEvent(
-        @Schema(description = "General purpose identifier. It is returned exactly the same way it is inserted with the LayoutParsingRequest.") //
-        Map<String, String> identifier,
-
-        @Schema(description = "The duration of a single layout parsing in ms.") //
-        long duration,
-
-        @Schema(description = "The number of pages of the parsed document.") //
-        int numberOfPages,
-
-        @Schema(description = "A general message. It contains some information useful for a developer, like the paths where the files are stored. Not meant to be machine readable.") //
-        String message,
-
-        @Schema(description = "The app version of the layout parser.") //
-        String layoutParserVersion
-) {
+        @Schema(description = "General purpose identifier. It is returned exactly the same way it is inserted with the LayoutParsingRequest.")
+        Map<String, String> identifier,//
+        @Schema(description = "The duration of a single layout parsing in ms.")
+        long duration,//
+        @Schema(description = "The number of pages of the parsed document.")
+        int numberOfPages,//
+        @Schema(description = "A general message. It contains some information useful for a developer, like the paths where the files are stored. Not meant to be machine readable.")
+        String message) {
 
 }

layoutparser-service/layoutparser-service-internal-api/src/main/java/com/knecon/fforesight/service/layoutparser/internal/api/queue/LayoutParsingQueueNames.java

@@ -2,9 +2,6 @@ package com.knecon.fforesight.service.layoutparser.internal.api.queue;
 
 public class LayoutParsingQueueNames {
 
-    public static final String LAYOUT_PARSING_REQUEST_QUEUE_PREFIX = "layout_parsing_request";
-    public static final String LAYOUT_PARSING_REQUEST_EXCHANGE = "layout_parsing_request_exchange";
-    public static final String LAYOUT_PARSING_RESPONSE_QUEUE_PREFIX = "layout_parsing_response";
-    public static final String LAYOUT_PARSING_RESPONSE_EXCHANGE = "layout_parsing_response_exchange";
-    public static final String LAYOUT_PARSING_DLQ = "layout_parsing_error";
+    public static final String LAYOUT_PARSING_REQUEST_QUEUE = "layout_parsing_request_queue";
+    public static final String LAYOUT_PARSING_FINISHED_EVENT_QUEUE = "layout_parsing_response_queue";
 }

layoutparser-service/layoutparser-service-processor/build.gradle.kts

@@ -8,20 +8,16 @@ description = "layoutparser-service-processor"
 val jacksonVersion = "2.15.2"
 val pdfBoxVersion = "3.0.0"
 
-
 dependencies {
     implementation(project(":layoutparser-service-internal-api"))
     implementation(project(":viewer-doc-processor"))
 
-    implementation("com.knecon.fforesight:document:${rootProject.extra.get("documentVersion")}")
-    implementation("com.iqser.red.service:persistence-service-shared-api-v1:2.564.0-RED9010.0") {
+    implementation("com.iqser.red.service:persistence-service-shared-api-v1:2.144.0") {
         exclude("org.springframework.boot", "spring-boot-starter-security")
         exclude("org.springframework.boot", "spring-boot-starter-validation")
     }
-    implementation("com.knecon.fforesight:tenant-commons:0.30.0") {
-        exclude("com.iqser.red.commons", "storage-commons")
-    }
-    implementation("com.iqser.red.commons:storage-commons:2.50.0")
+    implementation("com.knecon.fforesight:tenant-commons:0.21.0")
+    implementation("com.iqser.red.commons:storage-commons:2.45.0")
 
     implementation("org.apache.pdfbox:pdfbox:${pdfBoxVersion}")
     implementation("org.apache.pdfbox:pdfbox-tools:${pdfBoxVersion}")
@@ -37,4 +33,5 @@ dependencies {
     implementation("org.commonmark:commonmark-ext-gfm-tables:0.22.0")
     implementation("com.pdftron:PDFNet:10.11.0")
     implementation("org.apache.commons:commons-text:1.12.0")
+
 }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/LayoutParsingPipeline.java

@@ -20,18 +20,12 @@ import org.apache.pdfbox.pdmodel.PDDocument;
 import org.apache.pdfbox.pdmodel.PDPage;
 import org.apache.pdfbox.pdmodel.common.PDRectangle;
 import org.apache.pdfbox.pdmodel.documentinterchange.markedcontent.PDMarkedContent;
-import org.springframework.beans.factory.annotation.Value;
 import org.springframework.stereotype.Service;
 
-import com.iqser.red.service.redaction.v1.server.mapper.DocumentDataMapper;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.ImageType;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.NodeType;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
 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.docstrum.model.Character;
-import com.knecon.fforesight.service.layoutparser.processor.model.DocumentWithVisualization;
-import com.knecon.fforesight.service.layoutparser.processor.model.table.Ruling;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.RedTextPosition;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.TextDirection;
 import com.knecon.fforesight.service.layoutparser.processor.services.classification.ClassificationService;
@@ -39,12 +33,14 @@ import com.knecon.fforesight.service.layoutparser.processor.services.mapper.Mark
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationDocument;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationPage;
-
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.ImageType;
 import com.knecon.fforesight.service.layoutparser.processor.model.image.ClassifiedImage;
 import com.knecon.fforesight.service.layoutparser.processor.model.outline.OutlineExtractorService;
-import com.knecon.fforesight.service.layoutparser.processor.model.outline.SectionTreeBuilderService;
-import com.knecon.fforesight.service.layoutparser.processor.model.outline.SectionTreeEnhancementService;
-import com.knecon.fforesight.service.layoutparser.processor.model.outline.SectionTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.outline.OutlineObject;
+import com.knecon.fforesight.service.layoutparser.processor.model.outline.OutlineValidationService;
+import com.knecon.fforesight.service.layoutparser.processor.model.outline.TOCEnrichmentService;
+import com.knecon.fforesight.service.layoutparser.processor.model.outline.TableOfContents;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.Cell;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.CleanRulings;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPageBlock;
@@ -68,6 +64,7 @@ import com.knecon.fforesight.service.layoutparser.processor.services.blockificat
 import com.knecon.fforesight.service.layoutparser.processor.services.factory.DocumentGraphFactory;
 import com.knecon.fforesight.service.layoutparser.processor.services.graphics.Box;
 import com.knecon.fforesight.service.layoutparser.processor.services.graphics.GraphicExtractorService;
+import com.knecon.fforesight.service.layoutparser.processor.services.mapper.DocumentDataMapper;
 import com.knecon.fforesight.service.layoutparser.processor.services.mapper.TaasDocumentDataMapper;
 import com.knecon.fforesight.service.layoutparser.processor.services.parsing.PDFLinesTextStripper;
 import com.knecon.fforesight.service.layoutparser.processor.services.visualization.LayoutGridService;
@@ -88,32 +85,29 @@ import lombok.extern.slf4j.Slf4j;
 @Slf4j
 @Service
 @RequiredArgsConstructor
-@FieldDefaults(level = AccessLevel.PRIVATE)
+@FieldDefaults(makeFinal = true, level = AccessLevel.PRIVATE)
 public class LayoutParsingPipeline {
 
-    final ImageServiceResponseAdapter imageServiceResponseAdapter;
-    final CvTableParsingAdapter cvTableParsingAdapter;
-    final LayoutParsingStorageService layoutParsingStorageService;
-    final SectionsBuilderService sectionsBuilderService;
-    final SimplifiedSectionTextService simplifiedSectionTextService;
-    final RulingCleaningService rulingCleaningService;
-    final TableExtractionService tableExtractionService;
-    final DocuMineBlockificationService docuMineBlockificationService;
-    final RedactManagerBlockificationService redactManagerBlockificationService;
-    final BlockificationPostprocessingService blockificationPostprocessingService;
-    final DocstrumBlockificationService docstrumBlockificationService;
-    final LayoutGridService layoutGridService;
-    final ObservationRegistry observationRegistry;
-    final VisualLayoutParsingAdapter visualLayoutParsingAdapter;
-    final GraphicExtractorService graphicExtractorService;
-    final OutlineExtractorService outlineExtractorService;
-    final SectionTreeBuilderService sectionTreeBuilderService;
-    final SectionTreeEnhancementService sectionTreeEnhancementService;
-    final LayoutParserSettings settings;
-    final ClassificationService classificationService;
-
-    @Value("${LAYOUT_PARSER_VERSION:}")
-    private String layoutParserVersion;
+    ImageServiceResponseAdapter imageServiceResponseAdapter;
+    CvTableParsingAdapter cvTableParsingAdapter;
+    LayoutParsingStorageService layoutParsingStorageService;
+    SectionsBuilderService sectionsBuilderService;
+    SimplifiedSectionTextService simplifiedSectionTextService;
+    RulingCleaningService rulingCleaningService;
+    TableExtractionService tableExtractionService;
+    DocuMineBlockificationService docuMineBlockificationService;
+    RedactManagerBlockificationService redactManagerBlockificationService;
+    BlockificationPostprocessingService blockificationPostprocessingService;
+    DocstrumBlockificationService docstrumBlockificationService;
+    LayoutGridService layoutGridService;
+    ObservationRegistry observationRegistry;
+    VisualLayoutParsingAdapter visualLayoutParsingAdapter;
+    GraphicExtractorService graphicExtractorService;
+    OutlineExtractorService outlineExtractorService;
+    OutlineValidationService outlineValidationService;
+    TOCEnrichmentService tocEnrichmentService;
+    LayoutparserSettings settings;
+    ClassificationService classificationService;
 
 
     public LayoutParsingFinishedEvent parseLayoutAndSaveFilesToStorage(LayoutParsingRequest layoutParsingRequest) throws IOException {
@@ -122,23 +116,17 @@ public class LayoutParsingPipeline {
         log.info("Starting layout parsing for {}", layoutParsingRequest.identifier());
 
         File originFile = layoutParsingStorageService.getOriginFile(layoutParsingRequest.originFileStorageId());
-        File viewerDocumentFile = layoutParsingStorageService.getViewerDocFile(layoutParsingRequest.viewerDocumentStorageId())
-                .orElse(originFile);
+        File viewerDocumentFile = layoutParsingStorageService.getViewerDocFile(layoutParsingRequest.viewerDocumentStorageId()).orElse(originFile);
 
         VisualLayoutParsingResponse visualLayoutParsingResponse = layoutParsingRequest.visualLayoutParsingFileId()
-                .map(layoutParsingStorageService::getVisualLayoutParsingFile)
-                .orElse(new VisualLayoutParsingResponse());
+                .map(layoutParsingStorageService::getVisualLayoutParsingFile).orElse(new VisualLayoutParsingResponse());
         ImageServiceResponse imageServiceResponse = layoutParsingRequest.imagesFileStorageId()
-                .map(layoutParsingStorageService::getImagesFile)
-                .orElse(new ImageServiceResponse());
+                .map(layoutParsingStorageService::getImagesFile).orElse(new ImageServiceResponse());
         TableServiceResponse tableServiceResponse = layoutParsingRequest.tablesFileStorageId()
-                .map(layoutParsingStorageService::getTablesFile)
-                .orElse(new TableServiceResponse());
+                .map(layoutParsingStorageService::getTablesFile).orElse(new TableServiceResponse());
 
-        LayoutParsingType layoutParsingType = settings.getLayoutParsingTypeOverride() == null //
-                ? layoutParsingRequest.layoutParsingType() : settings.getLayoutParsingTypeOverride();
-
-        ClassificationDocument classificationDocument = parseLayout(layoutParsingType,
+        ClassificationDocument classificationDocument = parseLayout(settings.getLayoutParsingTypeOverride() == null //
+                                                                            ? layoutParsingRequest.layoutParsingType() : settings.getLayoutParsingTypeOverride(),
                                                                     originFile,
                                                                     imageServiceResponse,
                                                                     tableServiceResponse,
@@ -147,37 +135,36 @@ public class LayoutParsingPipeline {
 
         log.info("Building document graph for {}", layoutParsingRequest.identifier());
 
-        DocumentWithVisualization documentWithVisualization = observeBuildDocumentGraph(layoutParsingType, classificationDocument);
+        Document documentGraph = observeBuildDocumentGraph(settings.getLayoutParsingTypeOverride() == null //
+                                                                   ? layoutParsingRequest.layoutParsingType() : settings.getLayoutParsingTypeOverride(), classificationDocument);
 
         log.info("Creating viewer document for {}", layoutParsingRequest.identifier());
 
-        layoutGridService.addLayoutGrid(viewerDocumentFile, documentWithVisualization, viewerDocumentFile, layoutParsingType, layoutParserVersion, false);
+        layoutGridService.addLayoutGrid(viewerDocumentFile, documentGraph, viewerDocumentFile, false);
 
         log.info("Storing resulting files for {}", layoutParsingRequest.identifier());
 
-        layoutParsingStorageService.storeDocumentData(layoutParsingRequest, DocumentDataMapper.toDocumentData(documentWithVisualization.document()));
-        if (layoutParsingRequest.documentMarkdownFileStorageId()
-                .isPresent()) {
-            layoutParsingStorageService.storeMarkdownFile(layoutParsingRequest.documentMarkdownFileStorageId()
-                                                                  .get(), new MarkdownMapper().toMarkdownContent(documentWithVisualization.document()));
+        layoutParsingStorageService.storeDocumentData(layoutParsingRequest, DocumentDataMapper.toDocumentData(documentGraph));
+        if (layoutParsingRequest.documentMarkdownFileStorageId().isPresent()) {
+            layoutParsingStorageService.storeMarkdownFile(layoutParsingRequest.documentMarkdownFileStorageId().get(), new MarkdownMapper().toMarkdownContent(documentGraph));
         }
-        layoutParsingStorageService.storeSimplifiedText(layoutParsingRequest, simplifiedSectionTextService.toSimplifiedText(documentWithVisualization.document()));
+        layoutParsingStorageService.storeSimplifiedText(layoutParsingRequest, simplifiedSectionTextService.toSimplifiedText(documentGraph));
         layoutParsingStorageService.storeViewerDocument(layoutParsingRequest, viewerDocumentFile);
 
         if (layoutParsingRequest.researchDocumentStorageId() != null) {
             log.info("Building research document data for {}", layoutParsingRequest.identifier());
-            var researchDocumentData = TaasDocumentDataMapper.fromDocument(documentWithVisualization.document());
+            var researchDocumentData = TaasDocumentDataMapper.fromDocument(documentGraph);
             layoutParsingStorageService.storeResearchDocumentData(layoutParsingRequest, researchDocumentData);
         }
 
         if (!viewerDocumentFile.equals(originFile)) {
-            assert !viewerDocumentFile.exists() || viewerDocumentFile.delete();
+            viewerDocumentFile.delete();
         }
-        assert !originFile.exists() || originFile.delete();
+        originFile.delete();
 
         return LayoutParsingFinishedEvent.builder()
                 .identifier(layoutParsingRequest.identifier())
-                .numberOfPages(documentWithVisualization.document().getNumberOfPages())
+                .numberOfPages(documentGraph.getNumberOfPages())
                 .duration(System.currentTimeMillis() - start)
                 .message(format("""
                                         Layout parsing has finished in %.02f s.
@@ -192,22 +179,21 @@ public class LayoutParsingPipeline {
                                         Viewer Doc: %s""",
                                 ((float) (System.currentTimeMillis() - start)) / 1000,
                                 layoutParsingRequest.identifier(),
-                                buildSemanticNodeCountMessage(documentWithVisualization.document().getNumberOfPages(), documentWithVisualization.buildSemanticNodeCounts()),
+                                buildSemanticNodeCountMessage(documentGraph.getNumberOfPages(), documentGraph.buildSemanticNodeCounts()),
                                 layoutParsingRequest.structureFileStorageId(),
                                 layoutParsingRequest.textBlockFileStorageId(),
                                 layoutParsingRequest.positionBlockFileStorageId(),
                                 layoutParsingRequest.pageFileStorageId(),
                                 layoutParsingRequest.simplifiedTextStorageId(),
                                 layoutParsingRequest.viewerDocumentStorageId()))
-                .layoutParserVersion(layoutParserVersion)
                 .build();
 
     }
 
 
-    private DocumentWithVisualization observeBuildDocumentGraph(LayoutParsingType layoutParsingType, ClassificationDocument classificationDocument) {
+    private Document observeBuildDocumentGraph(LayoutParsingType layoutParsingType, ClassificationDocument classificationDocument) {
 
-        AtomicReference<DocumentWithVisualization> documentReference = new AtomicReference<>();
+        AtomicReference<Document> documentReference = new AtomicReference<>();
 
         Observation.createNotStarted("LayoutParsingPipeline", observationRegistry)
                 .contextualName("build-document-graph")
@@ -254,8 +240,12 @@ public class LayoutParsingPipeline {
         }
 
         List<ClassificationPage> classificationPages = new ArrayList<>();
+        OutlineObject lastProcessedOutlineObject = null;
 
-        classificationDocument.setOutlineObjectTree(outlineExtractorService.getOutlineObjectTree(originDocument));
+        // parsing the structure elements could be useful as well
+        if (layoutParsingType != LayoutParsingType.REDACT_MANAGER_OLD) {
+            classificationDocument.setOutlineObjectTree(outlineExtractorService.getOutlineObjectTree(originDocument));
+        }
 
         long pageCount = originDocument.getNumberOfPages();
 
@@ -287,15 +277,18 @@ public class LayoutParsingPipeline {
             if (layoutParsingType.equals(LayoutParsingType.DOCUMINE_OLD)) {
                 var lines = TextPositionOperations.groupByLine(new HashSet<>(words));
                 classificationDocument.getLayoutDebugLayer().addLineVisualizationsFromNestedTextPosition(lines, pageNumber);
-                words = TextPositionOperations.sortWords(lines);
+                words = TextPositionOperations.sortLines(lines);
             }
             classificationDocument.getLayoutDebugLayer().addTextVisualizations(words, pageNumber);
 
             PDRectangle pdr = pdPage.getMediaBox();
 
-            List<Ruling> rulings = stripper.getRulings();
-            classificationDocument.getLayoutDebugLayer().addRulingVisualization(rulings, pageNumber);
-            CleanRulings cleanRulings = rulingCleaningService.deduplicateAndStraightenRulings(pdfTableCells.get(pageNumber), rulings);
+            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();
+            classificationDocument.getLayoutDebugLayer().addRulingVisualization(stripper.getRulings(), pageNumber);
+            CleanRulings cleanRulings = rulingCleaningService.deduplicateAndStraightenRulings(pdfTableCells.get(pageNumber), stripper.getRulings());
 
             PageInformation pageInformation = PageInformation.fromPDPage(pageNumber, pdPage);
             List<Cell> emptyTableCells = TableExtractionService.findCells(cleanRulings.getHorizontals(), cleanRulings.getVerticals(), pageInformation);
@@ -315,7 +308,8 @@ public class LayoutParsingPipeline {
                                     .toList());
 
             ClassificationPage classificationPage = switch (layoutParsingType) {
-                case REDACT_MANAGER_OLD -> redactManagerBlockificationService.blockify(stripper.getWords(), cleanRulings, classificationDocument.getLayoutDebugLayer());
+                case REDACT_MANAGER_OLD ->
+                        redactManagerBlockificationService.blockify(stripper.getWords(), cleanRulings, classificationDocument.getLayoutDebugLayer());
                 case DOCUMINE_OLD -> docuMineBlockificationService.blockify(words, cleanRulings);
                 case DOCUMINE, REDACT_MANAGER, REDACT_MANAGER_PARAGRAPH_DEBUG, REDACT_MANAGER_WITHOUT_DUPLICATE_PARAGRAPH ->
                         docstrumBlockificationService.blockify(words, cleanRulings, true, classificationDocument.getLayoutDebugLayer(), layoutParsingType);
@@ -323,9 +317,27 @@ public class LayoutParsingPipeline {
                         docstrumBlockificationService.blockify(words, cleanRulings, false, classificationDocument.getLayoutDebugLayer(), layoutParsingType);
             };
 
-            updateClassificationPage(pdPage, pdr, classificationPage, cleanRulings, pageNumber, pageInformation);
+            classificationPage.setCleanRulings(cleanRulings);
+            classificationPage.setRotation(rotation);
+            classificationPage.setLandscape(isLandscape);
+            classificationPage.setPageNumber(pageNumber);
+            classificationPage.setPageWidth(cropbox.getWidth());
+            classificationPage.setPageHeight(cropbox.getHeight());
 
-            blockificationPostprocessingService.findHeadlinesFromOutline(classificationDocument, pageNumber, classificationPage, pageInformation);
+            if (layoutParsingType != LayoutParsingType.REDACT_MANAGER_OLD) {
+                List<OutlineObject> outlineObjects = classificationDocument.getOutlineObjectTree().getOutlineObjectsPerPage().getOrDefault(pageNumber, new ArrayList<>());
+
+                OutlineObject notFoundOutlineObject = null;
+                if (lastProcessedOutlineObject != null && !lastProcessedOutlineObject.isFound()) {
+                    lastProcessedOutlineObject.resetPoint();
+                    notFoundOutlineObject = lastProcessedOutlineObject;
+                }
+                if (!outlineObjects.isEmpty()) {
+                    classificationPage.setOutlineObjects(outlineObjects);
+                    lastProcessedOutlineObject = blockificationPostprocessingService.sanitizeOutlineBlocks(classificationPage, notFoundOutlineObject);
+                }
+                classificationDocument.getLayoutDebugLayer().addOutlineObjects(outlineObjects, pageInformation);
+            }
 
             classificationDocument.getLayoutDebugLayer().addMarkedContentVisualizations(stripper.getMarkedContents(), pageNumber);
             // MarkedContent needs to be converted at this point, otherwise it leads to GC Problems in Pdfbox.
@@ -357,48 +369,29 @@ public class LayoutParsingPipeline {
 
         classificationService.classify(classificationDocument, layoutParsingType, identifier);
 
-        SectionTree sectionTree = sectionTreeBuilderService.createSectionTree(classificationDocument);
-        classificationDocument.setSectionTree(sectionTree);
+        TableOfContents tableOfContents = outlineValidationService.createToC(classificationDocument);
+        classificationDocument.setTableOfContents(tableOfContents);
 
         log.info("Building Sections for {}", identifier);
 
         switch (layoutParsingType) {
             case CLARIFYND_PARAGRAPH_DEBUG, REDACT_MANAGER_PARAGRAPH_DEBUG -> sectionsBuilderService.buildParagraphDebugSections(classificationDocument);
-            default -> sectionTreeEnhancementService.assignSectionBlocksAndImages(classificationDocument);
+            default -> tocEnrichmentService.assignSectionBlocksAndImages(classificationDocument);
         }
 
         return classificationDocument;
     }
 
 
-    private static void updateClassificationPage(PDPage pdPage,
-                                                 PDRectangle pdr,
-                                                 ClassificationPage classificationPage,
-                                                 CleanRulings cleanRulings,
-                                                 int pageNumber,
-                                                 PageInformation pageInformation) {
-
-        int rotation = pdPage.getRotation();
-        boolean isLandscape = pdr.getWidth() > pdr.getHeight() && (rotation == 0 || rotation == 180) || pdr.getHeight() > pdr.getWidth() && (rotation == 90 || rotation == 270);
-        classificationPage.setCleanRulings(cleanRulings);
-        classificationPage.setRotation(rotation);
-        classificationPage.setLandscape(isLandscape);
-        classificationPage.setPageNumber(pageNumber);
-        classificationPage.setPageWidth((float) pageInformation.width());
-        classificationPage.setPageHeight((float) pageInformation.height());
-    }
-
-
     private static void rotateDirAdjExactly(List<Word> words, PDPage pdPage) {
 
         for (TextDirection dir : TextDirection.values()) {
+
             double averageRotation = words.stream()
-                    .map(Word::getCharacters)
+                    .map(Word::getTextPositions)
                     .flatMap(Collection::stream)
-                    .map(Character::getTextPosition)
                     .filter(pos -> pos.getDir().equals(dir))
-                    .mapToDouble(RedTextPosition::getExactDir).average()
-                    .orElse(0);
+                    .mapToDouble(RedTextPosition::getExactDir).average().orElse(0);
 
             if (averageRotation == 0) {
                 continue;

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/LayoutParsingStorageService.java

@@ -11,14 +11,12 @@ import java.nio.file.Path;
 import java.nio.file.Paths;
 import java.nio.file.StandardOpenOption;
 import java.util.Optional;
-import java.util.concurrent.CompletableFuture;
 
-import org.springframework.core.task.TaskExecutor;
 import org.springframework.stereotype.Service;
 
 import com.fasterxml.jackson.databind.ObjectMapper;
-import com.iqser.red.service.redaction.v1.server.data.DocumentData;
 import com.iqser.red.storage.commons.service.StorageService;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentData;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.SimplifiedText;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.taas.ResearchDocumentData;
 import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingRequest;
@@ -41,8 +39,6 @@ public class LayoutParsingStorageService {
     private final StorageService storageService;
     private final ObjectMapper objectMapper;
 
-    private final TaskExecutor taskExecutor;
-
 
     @Observed(name = "LayoutParsingStorageService", contextualName = "get-origin-file")
     public File getOriginFile(String storageId) throws IOException {
@@ -104,35 +100,13 @@ public class LayoutParsingStorageService {
     }
 
 
-    @SneakyThrows
     @Observed(name = "LayoutParsingStorageService", contextualName = "store-document-data")
     public void storeDocumentData(LayoutParsingRequest layoutParsingRequest, DocumentData documentData) {
 
-        Runnable storeDocumentStructureRunnable = () -> storageService.storeProtoObject(TenantContext.getTenantId(),
-                                                                                       layoutParsingRequest.structureFileStorageId(),
-                                                                                       documentData.getDocumentStructure());
-
-        CompletableFuture<Void> storeDocumentStructureFuture = CompletableFuture.runAsync(storeDocumentStructureRunnable, taskExecutor);
-
-        Runnable storeDocumentTextDataRunnable = () -> storageService.storeProtoObject(TenantContext.getTenantId(),
-                                                                                      layoutParsingRequest.textBlockFileStorageId(),
-                                                                                      documentData.getDocumentTextData());
-
-        CompletableFuture<Void> storeDocumentTextDataFuture = CompletableFuture.runAsync(storeDocumentTextDataRunnable, taskExecutor);
-
-        Runnable storeDocumentPositionsRunnable = () -> storageService.storeProtoObject(TenantContext.getTenantId(),
-                                                                                       layoutParsingRequest.positionBlockFileStorageId(),
-                                                                                       documentData.getDocumentPositionData());
-
-        CompletableFuture<Void> storeDocumentPositionsFuture = CompletableFuture.runAsync(storeDocumentPositionsRunnable, taskExecutor);
-
-        Runnable storeDocumentPagesRunnable = () -> storageService.storeProtoObject(TenantContext.getTenantId(),
-                                                                                   layoutParsingRequest.pageFileStorageId(),
-                                                                                   documentData.getDocumentPages());
-
-        CompletableFuture<Void> storeDocumentPagesFuture = CompletableFuture.runAsync(storeDocumentPagesRunnable, taskExecutor);
-
-        CompletableFuture.allOf(storeDocumentStructureFuture, storeDocumentTextDataFuture, storeDocumentPositionsFuture, storeDocumentPagesFuture).join();
+        storageService.storeJSONObject(TenantContext.getTenantId(), layoutParsingRequest.structureFileStorageId(), documentData.getDocumentStructure());
+        storageService.storeJSONObject(TenantContext.getTenantId(), layoutParsingRequest.textBlockFileStorageId(), documentData.getDocumentTextData());
+        storageService.storeJSONObject(TenantContext.getTenantId(), layoutParsingRequest.positionBlockFileStorageId(), documentData.getDocumentPositions());
+        storageService.storeJSONObject(TenantContext.getTenantId(), layoutParsingRequest.pageFileStorageId(), documentData.getDocumentPages());
     }
 
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/LayoutparserSettings.java

@@ -13,8 +13,9 @@ import lombok.experimental.FieldDefaults;
 @Configuration
 @ConfigurationProperties("layoutparser")
 @FieldDefaults(level = AccessLevel.PRIVATE)
-public class LayoutParserSettings {
+public class LayoutparserSettings {
 
     boolean debug;
     LayoutParsingType layoutParsingTypeOverride;
+    String pdftronLicense;
 }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/docstrum/DocstrumSegmentationService.java

@@ -80,12 +80,16 @@ public class DocstrumSegmentationService {
 
     private List<Zone> computeZones(List<Word> textPositions, CleanRulings rulings, LayoutDebugLayer visualizations, TextDirection direction) {
 
-        List<Character> characters = textPositions.stream()
+        List<RedTextPosition> positions = textPositions.stream()
                 .filter(t -> t.getDir() == direction)
-                .map(Word::getCharacters)
+                .map(Word::getTextPositions)
                 .flatMap(List::stream)
                 .toList();
 
+        List<Character> characters = positions.stream()
+                .map(Character::new)
+                .collect(Collectors.toList());
+
         nearestNeighbourService.findNearestNeighbors(characters);
 
         double characterSpacing = spacingService.computeCharacterSpacing(characters);

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/docstrum/model/Line.java

@@ -36,13 +36,18 @@ public class Line extends TextBoundingBox {
     @EqualsAndHashCode.Include
     private final double y1;
 
+    private final double height;
+
     private FontStyle fontStyle;
 
-    private final List<Word> words;
+    private final List<Character> characters;
+    private final List<Word> words = new ArrayList<>();
 
 
     public Line(List<Character> characters, double wordSpacing) {
 
+        this.characters = characters;
+
         if (characters.size() >= 2) {
             // linear regression
             double sx = 0.0;
@@ -71,25 +76,13 @@ public class Line extends TextBoundingBox {
             this.y0 = character.getY() - dy;
             this.y1 = character.getY() + dy;
         }
-        this.words = new ArrayList<>();
-        computeWords(characters, wordSpacing * WORD_DISTANCE_MULTIPLIER);
+        height = computeHeight();
+        computeWords(wordSpacing * WORD_DISTANCE_MULTIPLIER);
         buildBBox();
         computeFontStyle();
     }
 
 
-    public Line(List<Word> words) {
-
-        this.words = words;
-        buildBBox();
-        x0 = getMinX();
-        y0 = getMinY();
-        x1 = getMaxX();
-        y1 = getMaxY();
-        computeFontStyle();
-    }
-
-
     private void computeFontStyle() {
 
         EnumMap<FontStyle, AtomicInteger> fontStyleCounter = new EnumMap<>(FontStyle.class);
@@ -107,7 +100,8 @@ public class Line extends TextBoundingBox {
         fontStyle = fontStyleCounter.entrySet()
                 .stream()
                 .max(Comparator.comparing(entry -> entry.getValue().get()))
-                .map(Map.Entry::getKey).orElse(FontStyle.REGULAR);
+                .map(Map.Entry::getKey)
+                .orElse(FontStyle.REGULAR);
     }
 
 
@@ -123,6 +117,14 @@ public class Line extends TextBoundingBox {
     }
 
 
+    private double computeHeight() {
+
+        return characters.stream()
+                       .map(Character::getHeight)
+                       .reduce(0d, Double::sum) / characters.size();
+    }
+
+
     public double angularDifference(Line j) {
 
         double diff = Math.abs(getAngle() - j.getAngle());
@@ -155,7 +157,7 @@ public class Line extends TextBoundingBox {
     }
 
 
-    private void computeWords(List<Character> characters, double wordSpacing) {
+    private void computeWords(double wordSpacing) {
 
         // Imo, the width of space should be scaled with the font size, but it only depends on the median distance between horizontal neighbours.
         // If there are large differences in fontsize on a page, this might lead to missing spaces for the smaller fonts and too many for larger fonts.
@@ -170,7 +172,7 @@ public class Line extends TextBoundingBox {
                     word = new Word();
                 }
             }
-            word.add(current);
+            word.getTextPositions().add(current.getTextPosition());
             previous = current;
         }
         words.add(word);
@@ -179,7 +181,9 @@ public class Line extends TextBoundingBox {
 
     private void buildBBox() {
 
-        this.setToBBoxOfComponents(words);
+        this.setToBBoxOfComponents(characters.stream()
+                                           .map(Character::getTextPosition)
+                                           .toList());
     }
 
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/docstrum/model/Zone.java

@@ -18,6 +18,7 @@ public class Zone extends TextBoundingBox {
     @SuppressWarnings("PMD.ConstructorCallsOverridableMethod")
     public Zone(List<Line> lines) {
 
+        lines.sort(Comparator.comparingDouble(Line::getY0));
         this.lines = lines;
         setToBBoxOfComponents(lines);
     }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/docstrum/service/LineBuilderService.java

@@ -17,7 +17,7 @@ public class LineBuilderService {
 
     private static final double CHARACTER_SPACING_DISTANCE_MULTIPLIER = 3.5;
     private static final double LINE_SPACING_THRESHOLD_MULTIPLIER = 0.67;
-    private static final double ANGLE_TOLERANCE = Math.toRadians(5);
+    private static final double ANGLE_TOLERANCE = Math.PI / 6;
 
 
     public List<Line> buildLines(List<Character> characters, double characterSpacing, double lineSpacing, CleanRulings rulings) {

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/docstrum/service/ZoneBuilderService.java

@@ -1,7 +1,9 @@
 package com.knecon.fforesight.service.layoutparser.processor.docstrum.service;
 
+import static com.knecon.fforesight.service.layoutparser.processor.model.SectionIdentifier.numericalIdentifierPattern;
+
 import java.util.ArrayList;
-import java.util.Collection;
+import java.util.Comparator;
 import java.util.HashSet;
 import java.util.List;
 import java.util.Set;
@@ -9,12 +11,11 @@ import java.util.stream.Collectors;
 
 import org.springframework.stereotype.Service;
 
+import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.Character;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.Line;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.UnionFind;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.Zone;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.CleanRulings;
-import com.knecon.fforesight.service.layoutparser.processor.model.text.Word;
-import com.knecon.fforesight.service.layoutparser.processor.utils.TextPositionOperations;
 
 @Service
 public class ZoneBuilderService {
@@ -30,7 +31,7 @@ public class ZoneBuilderService {
 
     private static final double MAX_LINE_SIZE_SCALE = 2.5;
 
-    private static final double ANGLE_TOLERANCE = Math.toRadians(5);
+    private static final double ANGLE_TOLERANCE = Math.PI / 6;
 
     private static final double MAX_VERTICAL_MERGE_DISTANCE = 0.5;
 
@@ -113,14 +114,64 @@ public class ZoneBuilderService {
 
     private Zone mergeLinesInZone(List<Line> lines, double characterSpacing, double lineSpacing) {
 
-        Set<Word> words = lines.stream()
-                .map(Line::getWords)
-                .flatMap(Collection::stream)
-                .collect(Collectors.toSet());
-        Collection<Set<Word>> groupedLines = TextPositionOperations.groupByLine(words);
+        double maxHorizontalDistance = 0;
+        double minVerticalDistance = 0;
+        double maxVerticalDistance = lineSpacing * MAX_VERTICAL_MERGE_DISTANCE;
 
-        List<Line> sortedLines = TextPositionOperations.sortLines(groupedLines);
-        return new Zone(sortedLines);
+        UnionFind<Line> unionFind = new UnionFind<>(new HashSet<>(lines));
+
+        lines.forEach(outer -> {
+            lines.forEach(inner -> {
+                if (inner == outer) {
+                    return;
+                }
+
+                double horizontalDistance = outer.horizontalDistance(inner);
+                double verticalDistance = outer.verticalDistance(inner);
+
+                if (horizontalDistance <= maxHorizontalDistance && minVerticalDistance <= verticalDistance && verticalDistance <= maxVerticalDistance) {
+
+                    unionFind.union(outer, inner);
+
+                } else if (minVerticalDistance <= verticalDistance
+                           && verticalDistance <= maxVerticalDistance
+                           && Math.abs(horizontalDistance - Math.min(outer.getLength(), inner.getLength())) < 0.1) {
+
+                    boolean characterOverlap = false;
+                    int overlappingCount = 0;
+                    for (Character outerCharacter : outer.getCharacters()) {
+                        for (Character innerCharacter : inner.getCharacters()) {
+                            double characterOverlapDistance = outerCharacter.overlappingDistance(innerCharacter);
+                            if (characterOverlapDistance > 2) {
+                                characterOverlap = true;
+                            }
+                            if (characterOverlapDistance > 0) {
+                                overlappingCount++;
+                            }
+                        }
+                    }
+                    if (!characterOverlap && overlappingCount <= 2) {
+                        unionFind.union(outer, inner);
+                    }
+                }
+
+            });
+        });
+
+        List<Line> outputZone = new ArrayList<>();
+        for (Set<Line> group : unionFind.getGroups()) {
+            List<Character> characters = new ArrayList<>();
+            for (Line line : group) {
+                characters.addAll(line.getCharacters());
+            }
+            characters.sort(Comparator.comparingDouble(Character::getX));
+
+            outputZone.add(new Line(characters, characterSpacing));
+        }
+
+        return new Zone(outputZone.stream()
+                                .sorted(Comparator.comparing(Line::getY0))
+                                .collect(Collectors.toList()));
     }
 
 }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/AbstractPageBlock.java

@@ -4,7 +4,7 @@ import java.util.HashSet;
 import java.util.Set;
 
 import com.fasterxml.jackson.annotation.JsonIgnore;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.LayoutEngine;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.LayoutEngine;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.BoundingBox;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPageBlock;
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/ClassificationDocument.java

@@ -4,7 +4,7 @@ import java.util.ArrayList;
 import java.util.List;
 
 import com.knecon.fforesight.service.layoutparser.processor.model.outline.OutlineObjectTree;
-import com.knecon.fforesight.service.layoutparser.processor.model.outline.SectionTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.outline.TableOfContents;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.StringFrequencyCounter;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.UnclassifiedText;
 import com.knecon.fforesight.service.layoutparser.processor.visualization.LayoutDebugLayer;
@@ -31,6 +31,6 @@ public class ClassificationDocument {
     private long rulesVersion;
 
     private OutlineObjectTree outlineObjectTree;
-    private SectionTree sectionTree;
+    private TableOfContents tableOfContents;
 
 }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/DocumentWithVisualization.java

@@ -1,19 +0,0 @@
-package com.knecon.fforesight.service.layoutparser.processor.model;
-
-import java.util.Map;
-import java.util.stream.Collectors;
-
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
-import com.knecon.fforesight.service.layoutparser.processor.visualization.LayoutDebugLayer;
-
-public record DocumentWithVisualization(Document document, LayoutDebugLayer layoutDebugLayer) {
-
-    public Map<NodeType, Long> buildSemanticNodeCounts() {
-
-        return document.streamAllSubNodes()
-                .collect(Collectors.groupingBy(SemanticNode::getType, Collectors.counting()));
-    }
-
-}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/PageBlockType.java

@@ -14,7 +14,6 @@ public enum PageBlockType {
     PARAGRAPH_ITALIC,
     PARAGRAPH_UNKNOWN,
     OTHER,
-    TABLE_OF_CONTENTS_HEADLINE,
     TABLE_OF_CONTENTS_ITEM,
     LIST_ITEM,
     TABLE;
@@ -36,7 +35,7 @@ public enum PageBlockType {
     public static int getHeadlineNumber(PageBlockType pageBlockType) {
 
         return switch (pageBlockType) {
-            case H1, TABLE_OF_CONTENTS_HEADLINE -> 1;
+            case H1 -> 1;
             case H2 -> 2;
             case H3 -> 3;
             case H4 -> 4;
@@ -48,6 +47,6 @@ public enum PageBlockType {
 
     public boolean isHeadline() {
 
-        return this.equals(H1) || this.equals(H2) || this.equals(H3) || this.equals(H4) || this.equals(H5) || this.equals(H6) || this.equals(TABLE_OF_CONTENTS_HEADLINE);
+        return this.equals(H1) || this.equals(H2) || this.equals(H3) || this.equals(H4) || this.equals(H5) || this.equals(H6);
     }
 }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/AbstractNodeVisitor.java

@@ -0,0 +1,94 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph;
+
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Footer;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Header;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Headline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Image;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Paragraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Section;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SuperSection;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
+
+public abstract class AbstractNodeVisitor implements NodeVisitor {
+
+    @Override
+    public void visit(Document document) {
+
+        visitChildren(document);
+    }
+
+
+    @Override
+    public void visit(SuperSection superSection) {
+
+        visitChildren(superSection);
+    }
+
+
+    @Override
+    public void visit(Section section) {
+
+        visitChildren(section);
+    }
+
+
+    @Override
+    public void visit(Headline headline) {
+
+        visitChildren(headline);
+    }
+
+
+    @Override
+    public void visit(Paragraph paragraph) {
+
+        visitChildren(paragraph);
+    }
+
+
+    @Override
+    public void visit(Footer footer) {
+
+        visitChildren(footer);
+    }
+
+
+    @Override
+    public void visit(Header header) {
+
+        visitChildren(header);
+    }
+
+
+    @Override
+    public void visit(Image image) {
+
+        visitChildren(image);
+    }
+
+
+    @Override
+    public void visit(Table table) {
+
+        visitChildren(table);
+    }
+
+
+    @Override
+    public void visit(TableCell tableCell) {
+
+        visitChildren(tableCell);
+    }
+
+
+    protected void visitChildren(SemanticNode semanticNode) {
+
+        semanticNode.streamChildren()
+                .forEach(node -> node.accept(this));
+
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/DocumentTree.java

@@ -0,0 +1,230 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph;
+
+import static java.lang.String.format;
+
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.stream.Stream;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.GenericSemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlockCollector;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.Getter;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@EqualsAndHashCode
+public class DocumentTree {
+
+    private final Entry root;
+
+
+    public DocumentTree(Document document) {
+
+        root = Entry.builder().treeId(Collections.emptyList()).children(new LinkedList<>()).node(document).build();
+    }
+
+
+    public TextBlock buildTextBlock() {
+
+        return allEntriesInOrder().map(Entry::getNode)
+                .filter(SemanticNode::isLeaf)
+                .map(SemanticNode::getLeafTextBlock)
+                .collect(new TextBlockCollector());
+    }
+
+
+    public List<Integer> createNewMainEntryAndReturnId(GenericSemanticNode node) {
+
+        return createNewChildEntryAndReturnIdImpl(Collections.emptyList(), node);
+    }
+
+
+    public List<Integer> createNewChildEntryAndReturnId(GenericSemanticNode parentNode, GenericSemanticNode node) {
+
+        return createNewChildEntryAndReturnIdImpl(parentNode.getTreeId(), node);
+    }
+
+
+    public List<Integer> createNewChildEntryAndReturnId(GenericSemanticNode parentNode, Table node) {
+
+        return createNewChildEntryAndReturnIdImpl(parentNode.getTreeId(), node);
+    }
+
+
+    public List<Integer> createNewTableChildEntryAndReturnId(Table parentTable, TableCell tableCell) {
+
+        return createNewChildEntryAndReturnIdImpl(parentTable.getTreeId(), tableCell);
+    }
+
+
+    @SuppressWarnings("PMD.UnusedPrivateMethod") // PMD actually flags this wrong
+    private List<Integer> createNewChildEntryAndReturnIdImpl(List<Integer> parentId, SemanticNode node) {
+
+        if (!entryExists(parentId)) {
+            throw new IllegalArgumentException(format("parentId %s does not exist!", parentId));
+        }
+
+        Entry parent = getEntryById(parentId);
+        List<Integer> newId = new LinkedList<>(parentId);
+        newId.add(parent.children.size());
+        parent.children.add(Entry.builder().treeId(newId).node(node).build());
+
+        return newId;
+    }
+
+
+    private boolean entryExists(List<Integer> treeId) {
+
+        if (treeId.isEmpty()) {
+            return root != null;
+        }
+        Entry entry = root.children.get(treeId.get(0));
+        for (int id : treeId.subList(1, treeId.size())) {
+            if (id >= entry.children.size() || 0 > id) {
+                return false;
+            }
+            entry = entry.children.get(id);
+        }
+        return true;
+    }
+
+
+    public Entry getParentEntryById(List<Integer> treeId) {
+
+        return getEntryById(getParentId(treeId));
+    }
+
+
+    public boolean hasParentById(List<Integer> treeId) {
+
+        return !treeId.isEmpty();
+    }
+
+
+    public Stream<SemanticNode> childNodes(List<Integer> treeId) {
+
+        return getEntryById(treeId).children.stream()
+                .map(Entry::getNode);
+    }
+
+
+    public Stream<SemanticNode> childNodesOfType(List<Integer> treeId, NodeType nodeType) {
+
+        return getEntryById(treeId).children.stream()
+                .filter(entry -> entry.node.getType().equals(nodeType))
+                .map(Entry::getNode);
+    }
+
+
+    private static List<Integer> getParentId(List<Integer> treeId) {
+
+        if (treeId.isEmpty()) {
+            throw new UnsupportedOperationException("Root has no parent!");
+        }
+        if (treeId.size() < 2) {
+            return Collections.emptyList();
+        }
+        return treeId.subList(0, treeId.size() - 1);
+    }
+
+
+    public Entry getEntryById(List<Integer> treeId) {
+
+        if (treeId.isEmpty()) {
+            return root;
+        }
+        Entry entry = root;
+        for (int id : treeId) {
+            entry = entry.children.get(id);
+        }
+        return entry;
+    }
+
+
+    public Stream<Entry> mainEntries() {
+
+        return root.children.stream();
+    }
+
+
+    public Stream<Entry> allEntriesInOrder() {
+
+        return Stream.of(root)
+                .flatMap(DocumentTree::flatten);
+    }
+
+
+    public Stream<Entry> allSubEntriesInOrder(List<Integer> parentId) {
+
+        return getEntryById(parentId).children.stream()
+                .flatMap(DocumentTree::flatten);
+    }
+
+
+    @Override
+    public String toString() {
+
+        return String.join("\n",
+                           allEntriesInOrder().map(Entry::toString)
+                                   .toList());
+    }
+
+
+    private static Stream<Entry> flatten(Entry entry) {
+
+        return Stream.concat(Stream.of(entry),
+                             entry.children.stream()
+                                     .flatMap(DocumentTree::flatten));
+    }
+
+
+    public SemanticNode getHighestParentById(List<Integer> treeId) {
+
+        if (treeId.isEmpty()) {
+            return root.node;
+        }
+        return root.children.get(treeId.get(0)).node;
+    }
+
+
+    @Builder
+    @Getter
+    @AllArgsConstructor
+    @FieldDefaults(level = AccessLevel.PRIVATE, makeFinal = true)
+    public static class Entry {
+
+        List<Integer> treeId;
+        SemanticNode node;
+        @Builder.Default
+        List<Entry> children = new LinkedList<>();
+
+
+        @Override
+        public String toString() {
+
+            return node.toString();
+        }
+
+
+        public NodeType getType() {
+
+            return node.getType();
+        }
+
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/NodeVisitor.java

@@ -0,0 +1,45 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph;
+
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Footer;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Header;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Headline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Image;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Paragraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Section;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SuperSection;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
+
+public interface NodeVisitor {
+
+    void visit(Document document);
+
+
+    void visit(SuperSection superSection);
+
+
+    void visit(Section section);
+
+
+    void visit(Headline headline);
+
+
+    void visit(Paragraph paragraph);
+
+
+    void visit(Footer footer);
+
+
+    void visit(Header header);
+
+
+    void visit(Image image);
+
+
+    void visit(Table table);
+
+
+    void visit(TableCell tableCell);
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/TextRange.java

@@ -0,0 +1,166 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph;
+
+import static java.lang.String.format;
+
+import java.util.Collection;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.stream.IntStream;
+
+import lombok.EqualsAndHashCode;
+import lombok.Setter;
+
+@Setter
+@EqualsAndHashCode
+@SuppressWarnings("PMD.AvoidFieldNameMatchingMethodName")
+public class TextRange implements Comparable<TextRange> {
+
+    private int start;
+    private int end;
+
+
+    public TextRange(int start, int end) {
+
+        if (start > end) {
+            throw new IllegalArgumentException(format("start: %d > end: %d", start, end));
+        }
+        this.start = start;
+        this.end = end;
+    }
+
+
+    public int length() {
+
+        return end - start;
+    }
+
+
+    public int start() {
+
+        return start;
+    }
+
+
+    public int end() {
+
+        return end;
+    }
+
+
+    public boolean contains(TextRange textRange) {
+
+        return start <= textRange.start() && textRange.end() <= end;
+    }
+
+
+    public boolean containedBy(TextRange textRange) {
+
+        return textRange.contains(this);
+    }
+
+
+    public boolean contains(int start, int end) {
+
+        if (start > end) {
+            throw new IllegalArgumentException(format("start: %d > end: %d", start, end));
+        }
+        return this.start <= start && end <= this.end;
+    }
+
+
+    public boolean containedBy(int start, int end) {
+
+        if (start > end) {
+            throw new IllegalArgumentException(format("start: %d > end: %d", start, end));
+        }
+        return start <= this.start && this.end <= end;
+    }
+
+
+    public boolean contains(int index) {
+
+        return start <= index && index <= end;
+    }
+
+
+    public boolean containsExclusive(int index) {
+
+        return start <= index && index < end;
+    }
+
+
+    public boolean intersects(TextRange textRange) {
+
+        return textRange.start() < this.end && this.start < textRange.end();
+    }
+
+
+    public List<TextRange> split(List<Integer> splitIndices) {
+
+        if (splitIndices.stream()
+                .anyMatch(idx -> !this.contains(idx))) {
+            throw new IndexOutOfBoundsException(format("%s splitting indices are out of range for %s",
+                                                       splitIndices.stream()
+                                                               .filter(idx -> !this.contains(idx))
+                                                               .toList(),
+                                                       this));
+        }
+        List<TextRange> splitBoundaries = new LinkedList<>();
+        int previousIndex = start;
+        for (int splitIndex : splitIndices) {
+
+            // skip split if it would produce a boundary of length 0
+            if (splitIndex == previousIndex) {
+                continue;
+            }
+            splitBoundaries.add(new TextRange(previousIndex, splitIndex));
+            previousIndex = splitIndex;
+        }
+        if (previousIndex != end) {
+            splitBoundaries.add(new TextRange(previousIndex, end));
+        }
+        return splitBoundaries;
+    }
+
+
+    public IntStream intStream() {
+
+        return IntStream.range(start, end);
+    }
+
+
+    public static TextRange merge(Collection<TextRange> boundaries) {
+
+        int minStart = boundaries.stream()
+                .mapToInt(TextRange::start)
+                .min()
+                .orElseThrow(IllegalArgumentException::new);
+        int maxEnd = boundaries.stream()
+                .mapToInt(TextRange::end)
+                .max()
+                .orElseThrow(IllegalArgumentException::new);
+        return new TextRange(minStart, maxEnd);
+    }
+
+
+    @Override
+    public String toString() {
+
+        return format("Boundary [%d|%d)", start, end);
+    }
+
+
+    @Override
+    public int compareTo(TextRange textRange) {
+
+        if (end < textRange.end() && start < textRange.start()) {
+            return -1;
+        }
+        if (start > textRange.start() && end > textRange.end()) {
+            return 1;
+        }
+
+        return 0;
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/entity/EntityType.java

@@ -0,0 +1,8 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.entity;
+
+public enum EntityType {
+    ENTITY,
+    RECOMMENDATION,
+    FALSE_POSITIVE,
+    FALSE_RECOMMENDATION
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/entity/RedactionPosition.java

@@ -0,0 +1,24 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.entity;
+
+import java.awt.geom.Rectangle2D;
+import java.util.List;
+
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class RedactionPosition {
+
+    final String id;
+    Page page;
+    // Each entry in this list corresponds to an entry in the redaction log, this means:
+    // An entity might be represented by multiple redaction log entries
+    List<Rectangle2D> rectanglePerLine;
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/entity/TextEntity.java

@@ -0,0 +1,228 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.entity;
+
+import java.awt.geom.Rectangle2D;
+import java.util.Collection;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.HashSet;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+
+import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Engine;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.utils.IdBuilder;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@Builder
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+@EqualsAndHashCode(onlyExplicitlyIncluded = true)
+public class TextEntity {
+
+    // initial values
+    @EqualsAndHashCode.Include
+    final TextRange textRange;
+    @EqualsAndHashCode.Include
+    final String type;
+    @EqualsAndHashCode.Include
+    final EntityType entityType;
+
+    // empty defaults
+    boolean redaction;
+    boolean removed;
+    boolean ignored;
+    boolean resized;
+    boolean skipRemoveEntitiesContainedInLarger;
+    boolean dictionaryEntry;
+    boolean dossierDictionaryEntry;
+    Set<Engine> engines;
+    Set<TextEntity> references;
+    @Builder.Default
+    Deque<Integer> matchedRules = new LinkedList<>();
+    String redactionReason;
+    String legalBasis;
+
+    // inferred on graph insertion
+    @EqualsAndHashCode.Include
+    String value;
+    String textBefore;
+    String textAfter;
+    @Builder.Default
+    Set<Page> pages = new HashSet<>();
+    List<RedactionPosition> redactionPositionsPerPage;
+    @Builder.Default
+    List<SemanticNode> intersectingNodes = new LinkedList<>();
+    SemanticNode deepestFullyContainingNode;
+
+
+    public static TextEntity initialEntityNode(TextRange textRange, String type, EntityType entityType) {
+
+        return TextEntity.builder().type(type).entityType(entityType).textRange(textRange).engines(new HashSet<>()).references(new HashSet<>()).build();
+    }
+
+
+    public boolean occursInNodeOfType(Class<? extends SemanticNode> clazz) {
+
+        return intersectingNodes.stream().anyMatch(clazz::isInstance);
+    }
+
+
+    public boolean occursInNode(SemanticNode semanticNode) {
+
+        return intersectingNodes.stream().anyMatch(node -> node.equals(semanticNode));
+    }
+
+
+    public boolean isType(String type) {
+
+        return this.type.equals(type);
+    }
+
+
+    public boolean isAnyType(List<String> types) {
+
+        return types.contains(type);
+    }
+
+
+    public void addIntersectingNode(SemanticNode containingNode) {
+
+        intersectingNodes.add(containingNode);
+    }
+
+
+    public void removeFromGraph() {
+
+        intersectingNodes.forEach(node -> node.getEntities().remove(this));
+        pages.forEach(page -> page.getEntities().remove(this));
+        intersectingNodes = new LinkedList<>();
+        deepestFullyContainingNode = null;
+        pages = new HashSet<>();
+        removed = true;
+        ignored = true;
+    }
+
+
+    public void addMatchedRule(int ruleNumber) {
+
+        matchedRules.add(ruleNumber);
+    }
+
+
+    public int getMatchedRule() {
+
+        if (matchedRules.isEmpty()) {
+            return 0;
+        }
+        return matchedRules.getLast();
+    }
+
+
+    public List<RedactionPosition> getRedactionPositionsPerPage() {
+
+        if (redactionPositionsPerPage == null || redactionPositionsPerPage.isEmpty()) {
+            Map<Page, List<Rectangle2D>> rectanglesPerLinePerPage = deepestFullyContainingNode.getTextBlock().getPositionsPerPage(textRange);
+
+            Page firstPage = rectanglesPerLinePerPage.keySet()
+                    .stream()
+                    .min(Comparator.comparingInt(Page::getNumber))
+                    .orElseThrow(() -> new RuntimeException("No Positions found on any page!"));
+            String id = IdBuilder.buildId(pages, rectanglesPerLinePerPage.values().stream().flatMap(Collection::stream).toList());
+            redactionPositionsPerPage = rectanglesPerLinePerPage.entrySet().stream().map(entry -> buildRedactionPosition(firstPage, id, entry)).toList();
+        }
+        return redactionPositionsPerPage;
+    }
+
+
+    private static RedactionPosition buildRedactionPosition(Page firstPage, String id, Map.Entry<Page, List<Rectangle2D>> entry) {
+
+        if (entry.getKey().equals(firstPage)) {
+            return new RedactionPosition(id, entry.getKey(), entry.getValue());
+        } else {
+            return new RedactionPosition(id + "-" + entry.getKey().getNumber(), entry.getKey(), entry.getValue());
+        }
+    }
+
+
+    public boolean containedBy(TextEntity textEntity) {
+
+        return this.textRange.containedBy(textEntity.getTextRange());
+    }
+
+
+    public boolean contains(TextEntity textEntity) {
+
+        return this.textRange.contains(textEntity.getTextRange());
+    }
+
+
+    public boolean intersects(TextEntity textEntity) {
+
+        return this.textRange.intersects(textEntity.getTextRange());
+    }
+
+
+    public void addEngine(Engine engine) {
+
+        engines.add(engine);
+    }
+
+
+    public void addEngines(Set<Engine> engines) {
+
+        this.engines.addAll(engines);
+    }
+
+
+    public void addReference(TextEntity reference) {
+
+        references.add(reference);
+    }
+
+
+    public void addReferences(List<TextEntity> references) {
+
+        this.references.addAll(references);
+    }
+
+
+    public boolean matchesAnnotationId(String manualRedactionId) {
+
+        return getRedactionPositionsPerPage().stream().anyMatch(entityPosition -> entityPosition.getId().equals(manualRedactionId));
+    }
+
+
+    @Override
+    public String toString() {
+
+        StringBuilder sb = new StringBuilder();
+        sb.append("Entity[\"");
+        sb.append(value);
+        sb.append("\", ");
+        sb.append(textRange);
+        sb.append(", pages[");
+        pages.forEach(page -> {
+            sb.append(page.getNumber());
+            sb.append(", ");
+        });
+        sb.delete(sb.length() - 2, sb.length());
+        sb.append("], type = \"");
+        sb.append(type);
+        sb.append("\", EntityType.");
+        sb.append(entityType);
+        sb.append("]");
+        return sb.toString();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/AbstractSemanticNode.java

@@ -0,0 +1,74 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import java.awt.geom.Rectangle2D;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.LayoutEngine;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.DocumentTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.entity.TextEntity;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.NoArgsConstructor;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@NoArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public abstract class AbstractSemanticNode implements GenericSemanticNode {
+
+    @Builder.Default
+    Set<LayoutEngine> engines = new HashSet<>(Set.of(LayoutEngine.ALGORITHM));
+    List<Integer> treeId;
+
+    TextBlock textBlock;
+    @EqualsAndHashCode.Exclude
+    DocumentTree documentTree;
+
+    @Builder.Default
+    @EqualsAndHashCode.Exclude
+    Set<TextEntity> entities = new HashSet<>();
+
+    @EqualsAndHashCode.Exclude
+    Map<Page, Rectangle2D> bBoxCache;
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        if (textBlock == null) {
+            textBlock = GenericSemanticNode.super.getTextBlock();
+        }
+        return textBlock;
+    }
+
+
+    @Override
+    public String toString() {
+
+        return treeId.toString() + ": " + getType() + ": " + this.getTextBlock().buildSummary();
+    }
+
+
+    @Override
+    public Map<Page, Rectangle2D> getBBox() {
+
+        if (bBoxCache == null) {
+            bBoxCache = GenericSemanticNode.super.getBBox();
+        }
+        return bBoxCache;
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Document.java

@@ -0,0 +1,173 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import java.awt.geom.Rectangle2D;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.DocumentTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.visualization.LayoutDebugLayer;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.NoArgsConstructor;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@NoArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class Document extends AbstractSemanticNode {
+
+    Set<Page> pages;
+    Integer numberOfPages;
+
+    LayoutDebugLayer layoutDebugLayer;
+
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.DOCUMENT;
+    }
+
+
+    /**
+     * Gets the sections of the document as a list.
+     *
+     * @return A list of all sections within the document.
+     */
+    public List<Section> getAllSections() {
+
+        return streamAllSubNodesOfType(NodeType.SECTION).map(node -> (Section) node)
+                .collect(Collectors.toList());
+    }
+
+
+    /**
+     * Gets the main sections of the document as a list.
+     *
+     * @return A list of main sections within the document
+     * @deprecated This method is marked for removal.
+     * Use {@link #streamChildrenOfType(NodeType)} instead,
+     * or {@link #getChildrenOfTypeSectionOrSuperSection()} which returns children of type SECTION as well as SUPER_SECTION.
+     */
+    @Deprecated(forRemoval = true)
+    public List<Section> getMainSections() {
+
+        return streamChildrenOfType(NodeType.SECTION).map(node -> (Section) node)
+                .collect(Collectors.toList());
+    }
+
+
+    /**
+     * Gets the direct children of type SECTION or SUPER_SECTION of the document as a list of SemanticNode objects.
+     *
+     * @return A list of all children of type SECTION or SUPER_SECTION.
+     */
+    public List<SemanticNode> getChildrenOfTypeSectionOrSuperSection() {
+
+        return streamChildren().filter(semanticNode -> semanticNode.getType().equals(NodeType.SECTION) || semanticNode.getType().equals(NodeType.SUPER_SECTION))
+                .toList();
+    }
+
+
+    public List<Header> getHeaders() {
+
+        return streamChildrenOfType(NodeType.HEADER).map(node -> (Header) node)
+                .collect(Collectors.toList());
+    }
+
+
+    public List<Footer> getFooters() {
+
+        return streamChildrenOfType(NodeType.FOOTER).map(node -> (Footer) node)
+                .collect(Collectors.toList());
+    }
+
+
+    @Override
+    public Headline getHeadline() {
+
+        return streamAllSubNodesOfType(NodeType.HEADLINE).map(node -> (Headline) node)
+                .findFirst().orElse(Headline.builder().build());
+    }
+
+
+    public Stream<TextBlock> streamTerminalTextBlocksInOrder() {
+
+        return streamAllNodes().filter(SemanticNode::isLeaf)
+                .map(SemanticNode::getTextBlock);
+    }
+
+
+    @Override
+    public List<Integer> getTreeId() {
+
+        return Collections.emptyList();
+    }
+
+
+    @Override
+    public void setTreeId(List<Integer> tocId) {
+
+        throw new UnsupportedOperationException("Document is always the root of the TablePageBlock of Contents");
+    }
+
+
+    private Stream<SemanticNode> streamAllNodes() {
+
+        return getDocumentTree().allEntriesInOrder()
+                .map(DocumentTree.Entry::getNode);
+    }
+
+
+    public Stream<Image> streamAllImages() {
+
+        return streamAllSubNodesOfType(NodeType.IMAGE).map(node -> (Image) node);
+    }
+
+
+    public Map<NodeType, Long> buildSemanticNodeCounts() {
+
+        return streamAllSubNodes().collect(Collectors.groupingBy(SemanticNode::getType, Collectors.counting()));
+    }
+
+
+    @Override
+    public String toString() {
+
+        return NodeType.DOCUMENT + ": " + this.getTextBlock().buildSummary();
+    }
+
+
+    @Override
+    public Map<Page, Rectangle2D> getBBox() {
+
+        Map<Page, Rectangle2D> bBox = new HashMap<>();
+        for (Page page : pages) {
+            bBox.put(page, new Rectangle2D.Double(0, 0, page.getWidth(), page.getHeight()));
+        }
+        return bBox;
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+
+        visitor.visit(this);
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/DuplicatedParagraph.java

@@ -0,0 +1,28 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import java.util.stream.Stream;
+
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlockCollector;
+
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@EqualsAndHashCode(callSuper = true)
+@SuperBuilder
+public class DuplicatedParagraph extends Paragraph {
+
+    TextBlock unsortedLeafTextBlock;
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        return Stream.of(super.getLeafTextBlock(), unsortedLeafTextBlock)
+                .collect(new TextBlockCollector());
+
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Footer.java

@@ -0,0 +1,57 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class Footer extends AbstractSemanticNode {
+
+    TextBlock leafTextBlock;
+
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.FOOTER;
+    }
+
+
+    @Override
+    public boolean isLeaf() {
+
+        return true;
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+        visitor.visit(this);
+    }
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        return leafTextBlock;
+    }
+
+
+    @Override
+    public String toString() {
+
+        return getTreeId() + ": " + NodeType.FOOTER + ": " + leafTextBlock.buildSummary();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/GenericSemanticNode.java

@@ -0,0 +1,5 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+public interface GenericSemanticNode extends SemanticNode {
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Header.java

@@ -0,0 +1,58 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class Header extends AbstractSemanticNode {
+
+    TextBlock leafTextBlock;
+
+
+    @Override
+    public boolean isLeaf() {
+
+        return true;
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+
+        visitor.visit(this);
+    }
+
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.HEADER;
+    }
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        return leafTextBlock;
+    }
+
+
+    @Override
+    public String toString() {
+
+        return getTreeId() + ": " + NodeType.HEADER + ": " + leafTextBlock.buildSummary();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Headline.java

@@ -0,0 +1,65 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class Headline extends AbstractSemanticNode {
+
+    TextBlock leafTextBlock;
+
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.HEADLINE;
+    }
+
+
+    @Override
+    public boolean isLeaf() {
+
+        return true;
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+
+        visitor.visit(this);
+    }
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        return leafTextBlock;
+    }
+
+
+    @Override
+    public String toString() {
+
+        return getTreeId() + ": " + NodeType.HEADLINE + ": " + leafTextBlock.buildSummary();
+    }
+
+
+    @Override
+    public Headline getHeadline() {
+
+        return this;
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Image.java

@@ -0,0 +1,115 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import java.awt.geom.Rectangle2D;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Map;
+import java.util.Set;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.NoArgsConstructor;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@NoArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class Image extends AbstractSemanticNode {
+
+    String id;
+
+    String representationHash;
+
+    ImageType imageType;
+    boolean transparent;
+    Rectangle2D position;
+
+    TextBlock leafTextBlock;
+
+    boolean redaction;
+    boolean ignored;
+    @Builder.Default
+    String redactionReason = "";
+    @Builder.Default
+    String legalBasis = "";
+    @Builder.Default
+    int matchedRule = -1;
+
+    @EqualsAndHashCode.Exclude
+    Page page;
+
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.IMAGE;
+    }
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        return leafTextBlock;
+    }
+
+
+    @Override
+    public Set<Page> getPages() {
+
+        return Collections.singleton(page);
+    }
+
+
+    @Override
+    public String toString() {
+
+        return getTreeId() + ": " + NodeType.IMAGE + ": " + imageType.toString() + " " + position;
+    }
+
+
+    @Override
+    public Map<Page, Rectangle2D> getBBox() {
+
+        Map<Page, Rectangle2D> bBoxPerPage = new HashMap<>();
+        bBoxPerPage.put(page, position);
+        return bBoxPerPage;
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+
+        visitor.visit(this);
+    }
+
+
+    @Override
+    public boolean isLeaf() {
+
+        return true;
+    }
+
+
+    public double getArea() {
+
+        return position.getWidth() * position.getHeight();
+    }
+
+
+    public boolean isFullPageImage() {
+
+        return imageType.equals(ImageType.OCR) || getArea() >= 0.5 * page.getArea();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/ImageType.java

@@ -0,0 +1,26 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import java.util.Locale;
+
+public enum ImageType {
+    LOGO,
+    FORMULA,
+    SIGNATURE,
+    SIGNATURE_VISUAL,
+    OTHER,
+    OCR,
+    GRAPHIC;
+
+
+    public static ImageType fromString(String imageType) {
+
+        return switch (imageType.toLowerCase(Locale.ROOT)) {
+            case "logo" -> ImageType.LOGO;
+            case "formula" -> ImageType.FORMULA;
+            case "signature" -> ImageType.SIGNATURE;
+            case "ocr" -> ImageType.OCR;
+            case "graphic" -> ImageType.GRAPHIC;
+            default -> ImageType.OTHER;
+        };
+    }
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Page.java

@@ -0,0 +1,122 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import java.util.Comparator;
+import java.util.HashSet;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.Set;
+import java.util.stream.Stream;
+
+import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationPage;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.entity.TextEntity;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlockCollector;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.EqualsAndHashCode;
+import lombok.Getter;
+import lombok.Setter;
+import lombok.experimental.FieldDefaults;
+
+@Getter
+@Setter
+@Builder
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class Page {
+
+    Integer number;
+    Integer height;
+    Integer width;
+    Integer rotation;
+    @EqualsAndHashCode.Exclude
+    List<AtomicTextBlock> textBlocksOnPage;
+    @EqualsAndHashCode.Exclude
+    Header header;
+    @EqualsAndHashCode.Exclude
+    Footer footer;
+
+    @Builder.Default
+    @EqualsAndHashCode.Exclude
+    Set<TextEntity> entities = new HashSet<>();
+
+    @Builder.Default
+    @EqualsAndHashCode.Exclude
+    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())
+                .textBlocksOnPage(new LinkedList<>())
+                .build();
+    }
+
+
+    /**
+     * Constructs and returns a {@link TextBlock} representing the concatenated text of all leaf semantic nodes in the main body.
+     *
+     * @return The main body text block.
+     */
+    public TextBlock getMainBodyTextBlock() {
+
+        return textBlocksOnPage.stream()
+                .filter(atb -> !atb.isEmpty())
+                .collect(new TextBlockCollector());
+    }
+
+
+    public List<SemanticNode> getMainBody() {
+
+        return textBlocksOnPage.stream()
+                .map(AtomicTextBlock::getParent)
+                .map(this::getHighestParentOnPage)
+                .distinct()
+                .toList();
+    }
+
+
+    private SemanticNode getHighestParentOnPage(SemanticNode node) {
+
+        SemanticNode currentNode = node;
+        while (currentNode.hasParent() && currentNode.getParent().onlyOnPage(this)) {
+            currentNode = currentNode.getParent();
+        }
+        return currentNode;
+    }
+
+
+    @Override
+    public String toString() {
+
+        return String.valueOf(number);
+    }
+
+
+    @Override
+    public int hashCode() {
+
+        return number;
+    }
+
+
+    @Override
+    public boolean equals(Object o) {
+
+        return o instanceof Page && o.hashCode() == this.hashCode();
+    }
+
+
+    public double getArea() {
+
+        return height * width;
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Paragraph.java

@@ -0,0 +1,51 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+@FieldDefaults(level = AccessLevel.PROTECTED)
+public class Paragraph extends AbstractSemanticNode {
+
+    TextBlock leafTextBlock;
+
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.PARAGRAPH;
+    }
+
+
+    @Override
+    public boolean isLeaf() {
+
+        return true;
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+
+        visitor.visit(this);
+    }
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        return leafTextBlock;
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Section.java

@@ -0,0 +1,53 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+@EqualsAndHashCode(callSuper = true)
+public class Section extends AbstractSemanticNode {
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.SECTION;
+    }
+
+
+    public Headline getHeadline() {
+
+        return streamChildrenOfType(NodeType.HEADLINE).map(node -> (Headline) node)
+                .findFirst().orElseGet(() -> getParent().getHeadline());
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+
+        visitor.visit(this);
+    }
+
+
+    public boolean hasTables() {
+
+        return streamAllSubNodesOfType(NodeType.TABLE).findAny().isPresent();
+    }
+
+
+    @Override
+    public String toString() {
+
+        return getTreeId() + ": " + NodeType.SECTION + ": " + this.getTextBlock().buildSummary();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/SemanticNode.java

@@ -0,0 +1,521 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import static java.lang.String.format;
+
+import java.awt.geom.Rectangle2D;
+import java.util.Comparator;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Locale;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.LayoutEngine;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.DocumentTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.entity.EntityType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.entity.TextEntity;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlockCollector;
+import com.knecon.fforesight.service.layoutparser.processor.utils.BBoxMergingUtility;
+import com.knecon.fforesight.service.layoutparser.processor.utils.RectangleTransformations;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+
+public interface SemanticNode {
+
+    /**
+     * Returns the type of this node, such as Section, Paragraph, etc.
+     *
+     * @return NodeType of this node
+     */
+    NodeType getType();
+
+
+    /**
+     * Searches all Nodes located underneath this Node in the DocumentTree and concatenates their AtomicTextBlocks into a single TextBlock.
+     * So, for a Section all TextBlocks of Subsections, Paragraphs, and Tables are concatenated into a single TextBlock
+     * If the Node is a Leaf, the LeafTextBlock will be returned instead.
+     *
+     * @return TextBlock containing all AtomicTextBlocks that are located under this Node.
+     */
+    default TextBlock getTextBlock() {
+
+        return streamAllSubNodes().filter(SemanticNode::isLeaf)
+                .map(SemanticNode::getTextBlock)
+                .collect(new TextBlockCollector());
+    }
+
+
+    /**
+     * Any Node maintains its own Set of Entities.
+     * This Set contains all Entities whose boundary intersects the boundary of this node.
+     *
+     * @return Set of all Entities associated with this Node
+     */
+    Set<TextEntity> getEntities();
+
+
+    /**
+     * Each AtomicTextBlock is assigned a page, so to get the pages this node appears on, it collects the PageNodes from each AtomicTextBlock belonging to this node's TextBlock.
+     *
+     * @return Set of PageNodes this node appears on.
+     */
+    default Set<Page> getPages() {
+
+        return getTextBlock().getPages();
+    }
+
+
+    default Page getFirstPage() {
+
+        return getTextBlock().getPages()
+                .stream()
+                .min(Comparator.comparingInt(Page::getNumber))
+                .orElseThrow(() -> new IllegalStateException("SemanticNode has no Page!"));
+    }
+
+
+    /**
+     * Each AtomicTextBlock is assigned a page, so to get the pages for this boundary, it collects the PageNodes from each AtomicTextBlock belonging to this node's TextBlock.
+     *
+     * @return Set of PageNodes this node appears on.
+     */
+    default Set<Page> getPages(TextRange textRange) {
+
+        if (!getBoundary().contains(textRange)) {
+            throw new IllegalArgumentException(format("%s which was used to query for pages is not contained in the %s of this node!", textRange, getBoundary()));
+        }
+        return getTextBlock().getPages(textRange);
+    }
+
+
+    default boolean isOnPage(int pageNumber) {
+
+        return getPages().stream()
+                .anyMatch(page -> page.getNumber() == pageNumber);
+    }
+
+
+    /**
+     * Returns the DocumentTree Object.
+     *
+     * @return the DocumentTree of the Document this node belongs to
+     */
+    DocumentTree getDocumentTree();
+
+
+    /**
+     * The id is a List of Integers uniquely identifying this node in the DocumentTree.
+     *
+     * @return the DocumentTree ID
+     */
+    List<Integer> getTreeId();
+
+
+    /**
+     * This should only be used during graph construction.
+     *
+     * @param tocId List of Integers
+     */
+    void setTreeId(List<Integer> tocId);
+
+
+    /**
+     * Traverses the Tree up, until it hits a Headline or hits a Section which will then return the first Headline from its children.
+     * Throws NotFoundException if no Headline is found this way
+     *
+     * @return First Headline found
+     */
+    default Headline getHeadline() {
+
+        return getParent().getHeadline();
+    }
+
+
+    /**
+     * Checks if its TocId has a length greater than zero.
+     *
+     * @return boolean indicating whether this Node has a Parent in the DocumentTree
+     */
+    default boolean hasParent() {
+
+        return getDocumentTree().hasParentById(getTreeId());
+    }
+
+
+    /**
+     * @return The SemanticNode representing the Parent in the DocumentTree
+     * throws NotFoundException, when no parent is present
+     */
+    default SemanticNode getParent() {
+
+        return getDocumentTree().getParentEntryById(getTreeId()).getNode();
+    }
+
+
+    /**
+     * @return The SemanticNode which is directly underneath the document and also under which this node is.
+     * if this is the highest child node or the document itself, it returns itself.
+     */
+    default SemanticNode getHighestParent() {
+
+        return getDocumentTree().getHighestParentById(getTreeId());
+    }
+
+
+    /**
+     * Leaf means a SemanticNode has direct access to a TextBlock, by default this is false and must be overridden.
+     * Currently only Sections, Images, and Tables are not leaves.
+     * A TableCell might be a leaf depending on its area compared to the page.
+     *
+     * @return boolean, indicating if a Node has direct access to a TextBlock
+     */
+    default boolean isLeaf() {
+
+        return false;
+    }
+
+
+    /**
+     * Leaf means a SemanticNode has direct access to a TextBlock, by default this is false and must be overridden.
+     * Currently only Sections and Tables are no leaves.
+     *
+     * @return AtomicTextBlock
+     */
+    default TextBlock getLeafTextBlock() {
+
+        throw new UnsupportedOperationException("Only leaf Nodes have access to LeafTextBlocks!");
+    }
+
+
+    /**
+     * Should only be used during construction of the Graph. Sets the LeafTextBlock of this SemanticNode.
+     *
+     * @param textBlock the TextBlock to set as the LeafTextBlock of this SemanticNode
+     */
+    default void setLeafTextBlock(TextBlock textBlock) {
+
+        throw new UnsupportedOperationException();
+    }
+
+
+    /**
+     * Checks whether this SemanticNode has any Entity with EntityType.ENTITY of the provided type.
+     *
+     * @param type string representing the type of entity to check for
+     * @return true, if this SemanticNode has at least one Entity of the provided type
+     */
+    default boolean hasEntitiesOfType(String type) {
+
+        return getEntities().stream()
+                .filter(entity -> entity.getEntityType().equals(EntityType.ENTITY))
+                .anyMatch(redactionEntity -> redactionEntity.getType().equals(type));
+    }
+
+
+    /**
+     * Returns a List of Entities in this SemanticNode which are of the provided type such as "CBI_author".
+     *
+     * @param type string representing the type of entities to return
+     * @return List of RedactionEntities of any the type
+     */
+    default List<TextEntity> getEntitiesOfType(String type) {
+
+        return getEntities().stream()
+                .filter(redactionEntity -> redactionEntity.getType().equals(type))
+                .toList();
+    }
+
+
+    /**
+     * Returns a List of Entities in this SemanticNode which have any of the provided types such as "CBI_author".
+     *
+     * @param types A list of strings representing the types of entities to return
+     * @return List of RedactionEntities of any provided type
+     */
+    default List<TextEntity> getEntitiesOfType(List<String> types) {
+
+        return getEntities().stream()
+                .filter(redactionEntity -> redactionEntity.isAnyType(types))
+                .toList();
+    }
+
+
+    /**
+     * Each AtomicTextBlock has an index on its page, this returns the number of the first AtomicTextBlock underneath this node.
+     * If this node does not have any AtomicTexBlocks underneath it, e.g. an empty TableCell. It returns -1.
+     *
+     * @return Integer representing the number on the page
+     */
+    default Integer getNumberOnPage() {
+
+        TextBlock textBlock = getTextBlock();
+        if (!textBlock.getAtomicTextBlocks().isEmpty()) {
+            return getTextBlock().getAtomicTextBlocks().get(0).getNumberOnPage();
+        } else {
+            return -1;
+        }
+    }
+
+
+    /**
+     * Checks if the SemanticNode contains any text.
+     *
+     * @return true, if this node's TextBlock is not empty
+     */
+    default boolean hasText() {
+
+        return !getTextBlock().isEmpty();
+    }
+
+
+    /**
+     * Checks whether this SemanticNode contains the provided String.
+     *
+     * @param string A String which the TextBlock might contain
+     * @return true, if this node's TextBlock contains the string
+     */
+    default boolean containsString(String string) {
+
+        return getTextBlock().getSearchText().contains(string);
+    }
+
+
+    /**
+     * Checks whether this SemanticNode contains all the provided Strings.
+     *
+     * @param strings A List of Strings which the TextBlock might contain
+     * @return true, if this node's TextBlock contains all strings
+     */
+    default boolean containsStrings(List<String> strings) {
+
+        return strings.stream()
+                .allMatch(this::containsString);
+    }
+
+
+    /**
+     * Checks whether this SemanticNode contains all the provided Strings ignoring case.
+     *
+     * @param string A String which the TextBlock might contain
+     * @return true, if this node's TextBlock contains the string ignoring case
+     */
+    default boolean containsStringIgnoreCase(String string) {
+
+        return getTextBlock().getSearchText().toLowerCase(Locale.ROOT).contains(string.toLowerCase(Locale.ROOT));
+    }
+
+
+    /**
+     * Checks whether this SemanticNode contains any of the provided Strings.
+     *
+     * @param strings A List of Strings which the TextBlock might contain
+     * @return true, if this node's TextBlock contains any of the strings
+     */
+    default boolean containsAnyString(List<String> strings) {
+
+        return strings.stream()
+                .anyMatch(this::containsString);
+    }
+
+
+    /**
+     * Checks whether this SemanticNode contains any of the provided Strings ignoring case.
+     *
+     * @param strings A List of Strings which the TextBlock might contain
+     * @return true, if this node's TextBlock contains any of the strings
+     */
+    default boolean containsAnyStringIgnoreCase(List<String> strings) {
+
+        return strings.stream()
+                .anyMatch(this::containsStringIgnoreCase);
+    }
+
+
+    /**
+     * This function is used during insertion of EntityNodes into the graph, it checks if the boundary of the RedactionEntity intersects or even contains the RedactionEntity.
+     * It sets the fields accordingly and recursively calls this function on all its children.
+     *
+     * @param textEntity RedactionEntity, which is being inserted into the graph
+     */
+    default void addThisToEntityIfIntersects(TextEntity textEntity) {
+
+        TextBlock textBlock = getTextBlock();
+        if (textBlock.getTextRange().intersects(textEntity.getTextRange())) {
+            if (textBlock.containsBoundary(textEntity.getTextRange())) {
+                textEntity.setDeepestFullyContainingNode(this);
+            }
+
+            textEntity.addIntersectingNode(this);
+            streamChildren().filter(semanticNode -> semanticNode.getBoundary().intersects(textEntity.getTextRange()))
+                    .forEach(node -> node.addThisToEntityIfIntersects(textEntity));
+        }
+    }
+
+
+    /**
+     * returns the set of layoutengines.
+     *
+     * @return set of layoutengines.
+     */
+    Set<LayoutEngine> getEngines();
+
+
+    /**
+     * adds a layoutengine to the set.
+     */
+    default void addEngine(LayoutEngine engine) {
+
+        getEngines().add(engine);
+    }
+
+
+    /**
+     * Streams all children located directly underneath this node in the DocumentTree.
+     *
+     * @return Stream of all children
+     */
+    default Stream<SemanticNode> streamChildren() {
+
+        return getDocumentTree().childNodes(getTreeId());
+    }
+
+
+    /**
+     * Streams all children located directly underneath this node in the DocumentTree of the provided type.
+     *
+     * @return Stream of all children
+     */
+    default Stream<SemanticNode> streamChildrenOfType(NodeType nodeType) {
+
+        return getDocumentTree().childNodesOfType(getTreeId(), nodeType);
+    }
+
+
+    /**
+     * Recursively streams all SemanticNodes located underneath this node in the DocumentTree in order.
+     *
+     * @return Stream of all SubNodes
+     */
+    default Stream<SemanticNode> streamAllSubNodes() {
+
+        return getDocumentTree().allSubEntriesInOrder(getTreeId())
+                .map(DocumentTree.Entry::getNode);
+    }
+
+
+    /**
+     * Recursively streams all SemanticNodes of the provided type located underneath this node in the DocumentTree in order.
+     *
+     * @return Stream of all SubNodes
+     */
+    default Stream<SemanticNode> streamAllSubNodesOfType(NodeType nodeType) {
+
+        return getDocumentTree().allSubEntriesInOrder(getTreeId())
+                .filter(entry -> entry.getType().equals(nodeType))
+                .map(DocumentTree.Entry::getNode);
+    }
+
+
+    /**
+     * The Boundary is the start and end string offsets in the reading order of the document.
+     *
+     * @return Boundary of this Node's TextBlock
+     */
+    default TextRange getBoundary() {
+
+        return getTextBlock().getTextRange();
+    }
+
+
+    /**
+     * If this Node is a Leaf it will calculate the boundingBox of its LeafTextBlock, otherwise it will calculate the Union of the BoundingBoxes of all its Children.
+     * If called on the Document, it will return the cropbox of each page
+     *
+     * @return Rectangle2D fully encapsulating this Node for each page.
+     */
+    default Map<Page, Rectangle2D> getBBox() {
+
+        if (isLeaf()) {
+            return getBBoxFromLeafTextBlock();
+        }
+
+        return getBBoxFromChildren();
+    }
+
+
+    /**
+     * Checks whether the Bounding Box of this SemanticNode contains the provided rectangle on the provided page.
+     *
+     * @param rectangle2D The rectangle to check if it is contained
+     * @param pageNumber  The Page number on which the rectangle should be checked
+     * @return boolean
+     */
+    default boolean containsRectangle(Rectangle2D rectangle2D, Integer pageNumber) {
+
+        Page helperPage = Page.builder().number(pageNumber).build();
+        if (!getPages().contains(helperPage)) {
+            return false;
+        }
+        return getBBox().get(helperPage).contains(rectangle2D);
+    }
+
+
+    /**
+     * TODO: this produces unwanted results for sections spanning multiple columns.
+     * Computes the Union of the bounding boxes of all children recursively.
+     *
+     * @return The union of the BoundingBoxes of all children
+     */
+    private Map<Page, Rectangle2D> getBBoxFromChildren() {
+
+        List<Map<Page, Rectangle2D>> childrenBBoxes = streamChildren().filter(child -> !isFullPageImage(child))
+                .map(SemanticNode::getBBox)
+                .toList();
+        return BBoxMergingUtility.mergeBBoxes(childrenBBoxes);
+    }
+
+
+    private static boolean isFullPageImage(SemanticNode child) {
+
+        if (!child.getType().equals(NodeType.IMAGE)) {
+            return false;
+        }
+        return ((Image) child).isFullPageImage();
+    }
+
+
+    /**
+     * @return The union of all BoundingBoxes of the TextBlock of this node
+     */
+    private Map<Page, Rectangle2D> getBBoxFromLeafTextBlock() {
+
+        Map<Page, Rectangle2D> bBoxPerPage = new HashMap<>();
+        Map<Page, List<AtomicTextBlock>> atomicTextBlockPerPage = getTextBlock().getAtomicTextBlocks()
+                .stream()
+                .collect(Collectors.groupingBy(AtomicTextBlock::getPage));
+        atomicTextBlockPerPage.forEach((page, atbs) -> bBoxPerPage.put(page, RectangleTransformations.bBoxUnionAtomicTextBlock(atbs)));
+        return bBoxPerPage;
+    }
+
+
+    void accept(NodeVisitor visitor);
+
+
+    /**
+     * Checks wether this SemanticNode appears on a single page only, and if that page is the provided one.
+     *
+     * @param page the page to check
+     * @return true, when SemanticNode is on a single page only and the page is the provided page. Otherwise, false.
+     */
+    default boolean onlyOnPage(Page page) {
+
+        Set<Page> pages = getPages();
+        return pages.size() == 1 && pages.contains(page);
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/SuperSection.java

@@ -0,0 +1,47 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+@EqualsAndHashCode(callSuper = true)
+public class SuperSection extends AbstractSemanticNode {
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.SUPER_SECTION;
+    }
+
+
+    public Headline getHeadline() {
+
+        return streamChildrenOfType(NodeType.HEADLINE).map(node -> (Headline) node)
+                .findFirst().orElseGet(() -> getParent().getHeadline());
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+
+        visitor.visit(this);
+    }
+
+
+    @Override
+    public String toString() {
+
+        return getTreeId() + ": " + NodeType.SUPER_SECTION + ": " + this.getTextBlock().buildSummary();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/Table.java

@@ -0,0 +1,363 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import static java.lang.String.format;
+
+import java.awt.geom.Rectangle2D;
+import java.util.Collection;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Locale;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.IntStream;
+import java.util.stream.Stream;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.LayoutEngine;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.DocumentTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.entity.TextEntity;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlockCollector;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@Builder
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class Table implements SemanticNode {
+
+    @Builder.Default
+    Set<LayoutEngine> engines = new HashSet<>(Set.of(LayoutEngine.ALGORITHM));
+    List<Integer> treeId;
+    DocumentTree documentTree;
+
+    int numberOfRows;
+    int numberOfCols;
+    TextBlock textBlock;
+
+    @Builder.Default
+    @EqualsAndHashCode.Exclude
+    Set<TextEntity> entities = new HashSet<>();
+
+    @EqualsAndHashCode.Exclude
+    Map<Page, Rectangle2D> bBoxCache;
+
+    /**
+     * Streams all entities in this table, that appear in a row, which contains any of the provided strings.
+     *
+     * @param strings Strings to check whether a row contains them
+     * @return Stream of all entities in this table, that appear in a row, which contains any of the provided strings
+     */
+    public Stream<TextEntity> streamEntitiesWhereRowContainsStringsIgnoreCase(List<String> strings) {
+
+        return IntStream.range(0, numberOfRows).boxed()
+                .filter(row -> rowContainsStringsIgnoreCase(row, strings))
+                .flatMap(this::streamRow)
+                .map(TableCell::getEntities)
+                .flatMap(Collection::stream);
+    }
+
+
+    /**
+     * Checks whether the specified row contains all the provided strings.
+     *
+     * @param row     the row to check as an Integer, must be smaller than numberOfRows
+     * @param strings a list of strings to check for
+     * @return true, if all strings appear in the provided row
+     */
+    public boolean rowContainsStringsIgnoreCase(Integer row, List<String> strings) {
+
+        String rowText = streamRow(row).map(TableCell::getTextBlock)
+                .collect(new TextBlockCollector()).getSearchText().toLowerCase(Locale.ROOT);
+        return strings.stream()
+                .map(String::toLowerCase)
+                .allMatch(rowText::contains);
+    }
+
+
+    /**
+     * Streams all entities which appear in a row where at least one cell has the provided header and the provided value.
+     *
+     * @param header the header value to search for
+     * @param value  the string which the table cell should contain
+     * @return a stream of all entities, which appear in a row where at least one cell has the provided header and the provided value.
+     */
+    public Stream<TextEntity> streamEntitiesWhereRowHasHeaderAndValue(String header, String value) {
+
+        List<Integer> vertebrateStudyCols = streamHeaders().filter(headerNode -> headerNode.containsString(header))
+                .map(TableCell::getCol)
+                .toList();
+        return streamTableCells().filter(tableCellNode -> vertebrateStudyCols.stream()
+                        .anyMatch(vertebrateStudyCol -> getCell(tableCellNode.getRow(), vertebrateStudyCol).containsString(value)))
+                .map(TableCell::getEntities)
+                .flatMap(Collection::stream);
+    }
+
+
+    /**
+     * Streams all entities which appear in a row where at least one cell has the provided header and any provided value.
+     *
+     * @param header the header value to search for
+     * @param values the strings which the table cell should contain
+     * @return a stream of all entities, which appear in a row where at least one cell has the provided header and any provided value.
+     */
+    public Stream<TextEntity> streamEntitiesWhereRowHasHeaderAndAnyValue(String header, List<String> values) {
+
+        List<Integer> colsWithHeader = streamHeaders().filter(headerNode -> headerNode.containsString(header))
+                .map(TableCell::getCol)
+                .toList();
+        return streamTableCells().filter(tableCellNode -> colsWithHeader.stream()
+                        .anyMatch(colWithHeader -> getCell(tableCellNode.getRow(), colWithHeader).containsAnyString(values)))
+                .map(TableCell::getEntities)
+                .flatMap(Collection::stream);
+    }
+
+
+    /**
+     * Streams all entities in this table, that appear in a row, which contains at least one entity with any of the provided types.
+     *
+     * @param types type strings to check whether a row contains an entity like them
+     * @return Stream of all entities in this table, that appear in a row, which contains at least one entity with any of the provided types.
+     */
+    public Stream<TextEntity> streamEntitiesWhereRowContainsEntitiesOfType(List<String> types) {
+
+        List<Integer> rowsWithEntityOfType = IntStream.range(0, numberOfRows).boxed()
+                .filter(rowNumber -> streamEntityTypesInRow(rowNumber).anyMatch(existingType -> types.stream()
+                        .anyMatch(typeToCheck -> typeToCheck.equals(existingType))))
+                .toList();
+
+        return rowsWithEntityOfType.stream()
+                .flatMap(this::streamRow)
+                .map(TableCell::getEntities)
+                .flatMap(Collection::stream);
+    }
+
+
+    /**
+     * Streams all entities in this table, that appear in a row, which does not contain any entity with any of the provided types.
+     *
+     * @param types type strings to check whether a row doesn't contain an entity like it
+     * @return Stream of all entities in this table, that appear in a row, which does not contain any entity with any of the provided types.
+     */
+    public Stream<TextEntity> streamEntitiesWhereRowContainsNoEntitiesOfType(List<String> types) {
+
+        List<Integer> rowsWithNoEntityOfType = IntStream.range(0, numberOfRows).boxed()
+                .filter(rowNumber -> streamEntityTypesInRow(rowNumber).noneMatch(existingType -> types.stream()
+                        .anyMatch(typeToCheck -> typeToCheck.equals(existingType))))
+                .toList();
+
+        return rowsWithNoEntityOfType.stream()
+                .flatMap(this::streamRow)
+                .map(TableCell::getEntities)
+                .flatMap(Collection::stream);
+    }
+
+
+    private Stream<String> streamEntityTypesInRow(Integer rowNumber) {
+
+        return streamRow(rowNumber).map(TableCell::getEntities)
+                .flatMap(Collection::stream)
+                .map(TextEntity::getType)
+                .distinct();
+    }
+
+
+    /**
+     * Returns a TableCell at the provided row and column location.
+     *
+     * @param row int representing the row, must be smaller than numberOfRows
+     * @param col int representing the col, must be smaller than numberOfCols
+     * @return TableCell at the provided location in the table
+     */
+    public TableCell getCell(int row, int col) {
+
+        if (numberOfRows - row < 0 || numberOfCols - col < 0) {
+            throw new IllegalArgumentException(format("row %d, col %d is out of bounds for number of rows of %d and number of cols %d", row, col, numberOfRows, numberOfCols));
+        }
+        int idx = row * numberOfCols + col;
+        return (TableCell) documentTree.getEntryById(treeId).getChildren()
+                .get(idx).getNode();
+    }
+
+
+    /**
+     * Streams all TableCells in this Table row-wise.
+     *
+     * @return Stream of all TableCells
+     */
+    public Stream<TableCell> streamTableCells() {
+
+        return streamChildrenOfType(NodeType.TABLE_CELL).map(node -> (TableCell) node);
+    }
+
+
+    /**
+     * Streams all TableCells in this Table which have the provided header row-wise.
+     *
+     * @return Stream of all TableCells which have the provided header
+     */
+    public Stream<TableCell> streamTableCellsWithHeader(String header) {
+
+        return streamHeaders().filter(tableCellNode -> tableCellNode.getTextBlock().getSearchText().contains(header))
+                .map(TableCell::getCol)
+                .flatMap(this::streamCol)
+                .filter(tableCellNode -> !tableCellNode.isHeader());
+    }
+
+
+    /**
+     * Streams all TableCells belonging to the provided column from top down.
+     *
+     * @param col int representing the column
+     * @return Stream of all TableCell in the provided column
+     */
+    public Stream<TableCell> streamCol(int col) {
+
+        return IntStream.range(0, numberOfRows).boxed()
+                .map(row -> getCell(row, col));
+    }
+
+
+    /**
+     * Streams all TableCells belonging to the provided row from left to right.
+     *
+     * @param row int representing the row
+     * @return Stream of all TableCell in the provided row
+     */
+    public Stream<TableCell> streamRow(int row) {
+
+        return IntStream.range(0, numberOfCols).boxed()
+                .map(col -> getCell(row, col));
+    }
+
+
+    /**
+     * Streams all TableCells row-wise and filters them with header == true.
+     *
+     * @return Stream of all TableCells with header == true
+     */
+    public Stream<TableCell> streamHeaders() {
+
+        return streamTableCells().filter(TableCell::isHeader);
+    }
+
+
+    /**
+     * Streams all TableCells of the provided row and column and filters them with header == true.
+     *
+     * @param row int representing the row
+     * @param col int representing the column
+     * @return Stream of all TableCells with header == true in the provided row or col
+     */
+    public Stream<TableCell> streamHeadersForCell(int row, int col) {
+
+        return Stream.concat(streamRow(row), streamCol(col))
+                .filter(TableCell::isHeader);
+    }
+
+
+    /**
+     * Streams all Headers and checks if any equal the provided string.
+     *
+     * @param header string to check the headers for
+     * @return true, if at least one header equals the provided string
+     */
+    public boolean hasHeader(String header) {
+
+        return streamHeaders().anyMatch(tableCellNode -> tableCellNode.getTextBlock().getSearchText().strip().equals(header));
+    }
+
+
+    /**
+     * Checks if this table has a column with the provided header and any of the table cells in that column contain the provided value.
+     *
+     * @param header string to find header cells
+     * @param value  string to check cells with provided header
+     * @return true, if this table has a column with the provided header and any of the table cells in that column contain the provided value
+     */
+    public boolean hasRowWithHeaderAndValue(String header, String value) {
+
+        return streamTableCellsWithHeader(header).anyMatch(tableCellNode -> tableCellNode.containsString(value));
+    }
+
+
+    /**
+     * Checks if this table has a column with the provided header and any of the table cells in that column contains any of the provided values.
+     *
+     * @param header string to find header cells
+     * @param values List of strings to check cells with provided header
+     * @return true, if this table has a column with the provided header and any of the table cells in that column contains any of the provided values.
+     */
+    public boolean hasRowWithHeaderAndAnyValue(String header, List<String> values) {
+
+        return streamTableCellsWithHeader(header).anyMatch(tableCellNode -> tableCellNode.containsAnyString(values));
+    }
+
+
+    /**
+     * Finds all entities of the provided type, which appear in the same row that the provided entity appears in.
+     *
+     * @param type            the type of entities to search for
+     * @param textEntity the entity, which appears in the row to search
+     * @return List of all entities of the provided type, which appear in the same row that the provided entity appears in.
+     */
+    public List<TextEntity> getEntitiesOfTypeInSameRow(String type, TextEntity textEntity) {
+
+        return textEntity.getIntersectingNodes()
+                .stream()
+                .filter(node -> node instanceof TableCell)
+                .map(node -> (TableCell) node)
+                .flatMap(tableCellNode -> streamRow(tableCellNode.getRow()))
+                .map(cell -> cell.getEntitiesOfType(type))
+                .flatMap(Collection::stream)
+                .toList();
+    }
+
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.TABLE;
+    }
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        if (textBlock == null) {
+            textBlock = SemanticNode.super.getTextBlock();
+        }
+        return textBlock;
+    }
+
+
+    @Override
+    public String toString() {
+
+        return treeId.toString() + ": " + NodeType.TABLE + ": #cols: " + numberOfCols + ", #rows: " + numberOfRows + ", " + this.getTextBlock().buildSummary();
+    }
+
+
+    @Override
+    public Map<Page, Rectangle2D> getBBox() {
+
+        if (bBoxCache == null) {
+            bBoxCache = SemanticNode.super.getBBox();
+        }
+        return bBoxCache;
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+        visitor.visit(this);
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/nodes/TableCell.java

@@ -0,0 +1,95 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes;
+
+import java.awt.geom.Rectangle2D;
+import java.util.HashMap;
+import java.util.Map;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.NodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlockCollector;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+import lombok.experimental.SuperBuilder;
+
+@Data
+@SuperBuilder
+@AllArgsConstructor
+@EqualsAndHashCode(callSuper = true)
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class TableCell extends AbstractSemanticNode {
+
+    int row;
+    int col;
+    boolean header;
+
+    Rectangle2D bBox;
+
+    TextBlock leafTextBlock;
+
+    TextBlock textBlock;
+
+
+    @Override
+    public Map<Page, Rectangle2D> getBBox() {
+
+        Map<Page, Rectangle2D> bBoxPerPage = new HashMap<>();
+        getPages().forEach(page -> bBoxPerPage.put(page, bBox));
+        return bBoxPerPage;
+    }
+
+
+    @Override
+    public void accept(NodeVisitor visitor) {
+
+        visitor.visit(this);
+    }
+
+
+    @Override
+    public NodeType getType() {
+
+        return NodeType.TABLE_CELL;
+    }
+
+
+    @Override
+    public boolean isLeaf() {
+
+        return getDocumentTree().getEntryById(getTreeId()).getChildren().isEmpty();
+    }
+
+
+    @Override
+    public TextBlock getTextBlock() {
+
+        if (isLeaf()) {
+            return leafTextBlock;
+        }
+
+        if (textBlock == null) {
+            textBlock = buildTextBlock();
+        }
+        return textBlock;
+    }
+
+
+    private TextBlock buildTextBlock() {
+
+        return streamAllSubNodes().filter(SemanticNode::isLeaf)
+                .map(SemanticNode::getLeafTextBlock)
+                .collect(new TextBlockCollector());
+    }
+
+
+    @Override
+    public String toString() {
+
+        return getTreeId() + ": " + NodeType.TABLE_CELL + ": " + this.getTextBlock().buildSummary();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/textblock/AtomicTextBlock.java

@@ -0,0 +1,275 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock;
+
+import static java.lang.String.format;
+
+import java.awt.geom.Rectangle2D;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collection;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentPositionData;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentTextData;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.utils.RectangleTransformations;
+
+import lombok.AccessLevel;
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.EqualsAndHashCode;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@Builder
+@AllArgsConstructor
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class AtomicTextBlock implements TextBlock {
+
+    Long id;
+    Integer numberOnPage;
+    Page page;
+
+    //string coordinates
+    TextRange textRange;
+    String searchText;
+    @Builder.Default
+    List<Integer> lineBreaks = new ArrayList<>();
+    @Builder.Default
+    List<TextRange> boldTextBoundaries = new ArrayList<>();
+    @Builder.Default
+    List<TextRange> italicTextBoundaries = new ArrayList<>();
+    String orientation;
+    int textDirection;
+
+    //position coordinates
+    @Builder.Default
+    List<Integer> stringIdxToPositionIdx = new ArrayList<>();
+    @Builder.Default
+    List<Rectangle2D> positions = new ArrayList<>();
+
+    @EqualsAndHashCode.Exclude
+    SemanticNode parent;
+
+
+    @Override
+    public int numberOfLines() {
+
+        return lineBreaks.size() + 1;
+    }
+
+
+    @Override
+    public String subSequenceWithLineBreaks(TextRange stringTextRange) {
+
+        if (stringTextRange.length() == 0 || !getTextRange().contains(stringTextRange)) {
+            return "";
+        }
+
+        Set<Integer> lbInBoundary = lineBreaks.stream()
+                .map(i -> i + this.textRange.start())
+                .filter(stringTextRange::contains)
+                .collect(Collectors.toSet());
+        if (stringTextRange.end() == getTextRange().end()) {
+            lbInBoundary.add(getTextRange().end());
+        }
+        StringBuilder sb = new StringBuilder();
+        for (int i = stringTextRange.start(); i < stringTextRange.end(); i++) {
+            char character = this.charAt(i);
+            if (lbInBoundary.contains(i + 1)) {
+                // always plus one, due to the linebreaks being an exclusive end index
+                if (!Character.isWhitespace(character)) {
+                    lbInBoundary.remove(i + 1);
+                    lbInBoundary.add(i + 2);
+                    sb.append(character);
+                    continue;
+                }
+                sb.append("\n");
+            } else {
+                sb.append(character);
+            }
+        }
+        return sb.toString();
+    }
+
+
+    public static AtomicTextBlock fromSearchTextWithTextPosition(String searchText,
+                                                                 List<Integer> lineBreaks,
+                                                                 List<TextRange> boldTextBoundaries,
+                                                                 List<TextRange> italicTextBoundaries,
+                                                                 List<Rectangle2D> positions,
+                                                                 List<Integer> stringIdxToPositionIdx,
+                                                                 long idx,
+                                                                 SemanticNode parent,
+                                                                 int numberOnPage,
+                                                                 Page page,
+                                                                 int offset,
+                                                                 String orientation,
+                                                                 int textDirection) {
+
+        return AtomicTextBlock.builder()
+                .id(idx)
+                .parent(parent)
+                .searchText(searchText)
+                .numberOnPage(numberOnPage)
+                .page(page)
+                .lineBreaks(lineBreaks)
+                .boldTextBoundaries(boldTextBoundaries)
+                .italicTextBoundaries(italicTextBoundaries)
+                .positions(positions)
+                .stringIdxToPositionIdx(stringIdxToPositionIdx)
+                .textRange(new TextRange(offset, offset + searchText.length()))
+                .textDirection(textDirection)
+                .orientation(orientation)
+                .build();
+    }
+
+
+    public static AtomicTextBlock empty(Long textBlockIdx, int stringOffset, Page page, int numberOnPage, SemanticNode parent) {
+
+        return AtomicTextBlock.builder()
+                .id(textBlockIdx)
+                .textRange(new TextRange(stringOffset, stringOffset))
+                .searchText("")
+                .page(page)
+                .numberOnPage(numberOnPage)
+                .parent(parent)
+                .build();
+    }
+
+
+    public static AtomicTextBlock fromAtomicTextBlockData(DocumentTextData documentTextData, DocumentPositionData documentPositionData, SemanticNode parent, Page page) {
+
+        return AtomicTextBlock.builder()
+                .id(documentTextData.getId())
+                .numberOnPage(documentTextData.getNumberOnPage())
+                .page(page)
+                .textRange(new TextRange(documentTextData.getStart(), documentTextData.getEnd()))
+                .searchText(documentTextData.getSearchText())
+                .lineBreaks(Arrays.stream(documentTextData.getLineBreaks()).boxed()
+                                    .toList())
+                .stringIdxToPositionIdx(Arrays.stream(documentPositionData.getStringIdxToPositionIdx()).boxed()
+                                                .toList())
+                .positions(toRectangle2DList(documentPositionData.getPositions()))
+                .parent(parent)
+                .build();
+    }
+
+
+    private static List<Rectangle2D> toRectangle2DList(float[][] positions) {
+
+        return Arrays.stream(positions)
+                .map(floatArr -> (Rectangle2D) new Rectangle2D.Float(floatArr[0], floatArr[1], floatArr[2], floatArr[3]))
+                .toList();
+    }
+
+
+    public CharSequence getLine(int lineNumber) {
+
+        if (lineNumber >= numberOfLines() || lineNumber < 0) {
+            throw new IndexOutOfBoundsException(format("line %d out of range for AtomicTextBlock with %d lines", lineNumber, numberOfLines()));
+        }
+        if (lineNumber == 0) {
+            if (lineBreaks.isEmpty()) {
+                return searchText;
+            }
+            return subSequence(textRange.start(), lineBreaks.get(0) + textRange.start());
+        } else if (lineNumber == numberOfLines() - 1) {
+            return subSequence(lineBreaks.get(lineBreaks.size() - 1) + textRange.start(), textRange.end());
+        }
+        return subSequence(lineBreaks.get(lineNumber - 1) + textRange.start(), lineBreaks.get(lineNumber) + textRange.start());
+    }
+
+
+    @Override
+    public List<AtomicTextBlock> getAtomicTextBlocks() {
+
+        return List.of(this);
+    }
+
+
+    @Override
+    public int getNextLinebreak(int fromIndex) {
+
+        return lineBreaks.stream()//
+                       .filter(linebreak -> linebreak > fromIndex - textRange.start()) //
+                       .findFirst() //
+                       .orElse(searchText.length()) + textRange.start();
+    }
+
+
+    @Override
+    public int getPreviousLinebreak(int fromIndex) {
+
+        return lineBreaks.stream()//
+                       .filter(linebreak -> linebreak <= fromIndex - textRange.start())//
+                       .reduce((a, b) -> b)//
+                       .orElse(0) + textRange.start();
+    }
+
+
+    @Override
+    public Rectangle2D getPosition(int stringIdx) {
+
+        return positions.get(stringIdxToPositionIdx.get(stringIdx - textRange.start()));
+    }
+
+
+    @Override
+    public List<Rectangle2D> getPositions(TextRange stringTextRange) {
+
+        if (!containsBoundary(stringTextRange)) {
+            throw new IndexOutOfBoundsException(format("%s is out of bounds for %s", stringTextRange, this.textRange));
+        }
+        if (stringTextRange.length() == 0) {
+            return Collections.emptyList();
+        }
+
+        int startPositionIdx = stringIdxToPositionIdx.get(stringTextRange.start() - this.textRange.start());
+
+        if (stringTextRange.end() == this.textRange.end()) {
+            return positions.subList(startPositionIdx, positions.size());
+        }
+
+        return positions.subList(startPositionIdx, stringIdxToPositionIdx.get(stringTextRange.end() - this.textRange.start()));
+
+    }
+
+
+    public Map<Page, List<Rectangle2D>> getPositionsPerPage(TextRange stringTextRange) {
+
+        List<Rectangle2D> rectanglesPerLine = stringTextRange.split(getAllLineBreaksInBoundary(stringTextRange))
+                .stream()
+                .map(this::getPositions)
+                .map(RectangleTransformations::rectangleBBoxWithGaps)
+                .flatMap(Collection::stream)
+                .toList();
+        Map<Page, List<Rectangle2D>> rectanglePerLinePerPage = new HashMap<>();
+        rectanglePerLinePerPage.put(page, rectanglesPerLine);
+        return rectanglePerLinePerPage;
+    }
+
+
+    protected List<Integer> getAllLineBreaksInBoundary(TextRange textRange) {
+
+        return getLineBreaks().stream()
+                .map(linebreak -> linebreak + this.textRange.start())
+                .filter(textRange::contains)
+                .toList();
+    }
+
+
+    @Override
+    public String toString() {
+
+        return searchText;
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/textblock/ConcatenatedTextBlock.java

@@ -0,0 +1,271 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock;
+
+import static java.lang.String.format;
+
+import java.awt.geom.Rectangle2D;
+import java.util.Collection;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.Map;
+import java.util.stream.Stream;
+
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+
+import lombok.AccessLevel;
+import lombok.Data;
+import lombok.experimental.FieldDefaults;
+
+@Data
+@FieldDefaults(level = AccessLevel.PRIVATE)
+public class ConcatenatedTextBlock implements TextBlock {
+
+    List<AtomicTextBlock> atomicTextBlocks;
+    String searchText;
+    TextRange textRange;
+
+
+    public static ConcatenatedTextBlock empty() {
+
+        return new ConcatenatedTextBlock(Collections.emptyList());
+    }
+
+
+    public ConcatenatedTextBlock(List<AtomicTextBlock> atomicTextBlocks) {
+
+        this.atomicTextBlocks = new LinkedList<>();
+        if (atomicTextBlocks.isEmpty()) {
+            textRange = new TextRange(-1, -1);
+            return;
+        }
+        var firstTextBlock = atomicTextBlocks.get(0);
+        this.atomicTextBlocks.add(firstTextBlock);
+        textRange = new TextRange(firstTextBlock.getTextRange().start(), firstTextBlock.getTextRange().end());
+
+        atomicTextBlocks.subList(1, atomicTextBlocks.size())
+                .forEach(this::concat);
+    }
+
+
+    public ConcatenatedTextBlock concat(TextBlock textBlock) {
+
+        int start = textBlock.getTextRange().start();
+        int end = textBlock.getTextRange().end();
+        if (this.atomicTextBlocks.isEmpty()) {
+            textRange.setStart(start);
+            textRange.setEnd(end);
+        } else if (textRange.end() != start) {
+            throw new UnsupportedOperationException(format("Can only concat consecutive TextBlocks, trying to concat %s and %s", textRange, textBlock.getTextRange()));
+        }
+        this.atomicTextBlocks.addAll(textBlock.getAtomicTextBlocks());
+        textRange.setEnd(end);
+        this.searchText = null;
+        return this;
+    }
+
+
+    private AtomicTextBlock getAtomicTextBlockByStringIndex(int stringIdx) {
+
+        return atomicTextBlocks.stream()
+                .filter(textBlock -> textBlock.getTextRange().containsExclusive(stringIdx))
+                .findAny()
+                .orElseThrow(IndexOutOfBoundsException::new);
+    }
+
+
+    private List<AtomicTextBlock> getAllAtomicTextBlocksPartiallyInStringBoundary(TextRange textRange) {
+
+        return atomicTextBlocks.stream()
+                .filter(tb -> tb.getTextRange().intersects(textRange))
+                .toList();
+    }
+
+
+    @Override
+    public String getSearchText() {
+
+        if (searchText == null) {
+            StringBuilder sb = new StringBuilder();
+            getAtomicTextBlocks().forEach(atb -> sb.append(atb.getSearchText()));
+            searchText = sb.toString();
+        }
+        return searchText;
+    }
+
+
+    @Override
+    public int numberOfLines() {
+
+        return atomicTextBlocks.stream()
+                .map(AtomicTextBlock::getLineBreaks)
+                .mapToInt(List::size).sum();
+    }
+
+
+    @Override
+    public int getNextLinebreak(int fromIndex) {
+
+        return getAtomicTextBlockByStringIndex(fromIndex).getNextLinebreak(fromIndex);
+    }
+
+
+    @Override
+    public int getPreviousLinebreak(int fromIndex) {
+
+        return getAtomicTextBlockByStringIndex(fromIndex).getPreviousLinebreak(fromIndex);
+    }
+
+
+    @Override
+    public List<Integer> getLineBreaks() {
+
+        return getAtomicTextBlocks().stream()
+                .flatMap(atomicTextBlock -> atomicTextBlock.getLineBreaks()
+                        .stream())
+                .toList();
+    }
+
+
+    @Override
+    public Rectangle2D getPosition(int stringIdx) {
+
+        return getAtomicTextBlockByStringIndex(stringIdx).getPosition(stringIdx);
+    }
+
+
+    @Override
+    public List<Rectangle2D> getPositions(TextRange stringTextRange) {
+
+        List<AtomicTextBlock> textBlocks = getAllAtomicTextBlocksPartiallyInStringBoundary(stringTextRange);
+
+        if (textBlocks.size() == 1) {
+            return textBlocks.get(0).getPositions(stringTextRange);
+        }
+
+        AtomicTextBlock firstTextBlock = textBlocks.get(0);
+        List<Rectangle2D> positions = new LinkedList<>(firstTextBlock.getPositions(new TextRange(stringTextRange.start(), firstTextBlock.getTextRange().end())));
+
+        for (AtomicTextBlock textBlock : textBlocks.subList(1, textBlocks.size() - 1)) {
+            positions.addAll(textBlock.getPositions());
+        }
+
+        var lastTextBlock = textBlocks.get(textBlocks.size() - 1);
+        positions.addAll(lastTextBlock.getPositions(new TextRange(lastTextBlock.getTextRange().start(), stringTextRange.end())));
+
+        return positions;
+    }
+
+
+    @Override
+    public Map<Page, List<Rectangle2D>> getPositionsPerPage(TextRange stringTextRange) {
+
+        List<AtomicTextBlock> textBlocks = getAllAtomicTextBlocksPartiallyInStringBoundary(stringTextRange);
+
+        if (textBlocks.size() == 1) {
+            return textBlocks.get(0).getPositionsPerPage(stringTextRange);
+        }
+
+        AtomicTextBlock firstTextBlock = textBlocks.get(0);
+        Map<Page, List<Rectangle2D>> rectanglesPerLinePerPage = firstTextBlock.getPositionsPerPage(new TextRange(stringTextRange.start(), firstTextBlock.getTextRange().end()));
+
+        for (AtomicTextBlock textBlock : textBlocks.subList(1, textBlocks.size() - 1)) {
+            rectanglesPerLinePerPage = mergeEntityPositionsWithSamePageNode(rectanglesPerLinePerPage, textBlock.getPositionsPerPage(textBlock.getTextRange()));
+        }
+
+        AtomicTextBlock lastTextBlock = textBlocks.get(textBlocks.size() - 1);
+        rectanglesPerLinePerPage = mergeEntityPositionsWithSamePageNode(rectanglesPerLinePerPage,
+                                                                        lastTextBlock.getPositionsPerPage(new TextRange(lastTextBlock.getTextRange().start(),
+                                                                                                                        stringTextRange.end())));
+
+        return rectanglesPerLinePerPage;
+    }
+
+
+    private Map<Page, List<Rectangle2D>> mergeEntityPositionsWithSamePageNode(Map<Page, List<Rectangle2D>> map1, Map<Page, List<Rectangle2D>> map2) {
+
+        Map<Page, List<Rectangle2D>> mergedMap = new HashMap<>(map1);
+        map2.forEach((pageNode, rectangles) -> mergedMap.merge(pageNode,
+                                                               rectangles,
+                                                               (l1, l2) -> Stream.concat(l1.stream(), l2.stream())
+                                                                       .toList()));
+        return mergedMap;
+    }
+
+
+    @Override
+    public String subSequenceWithLineBreaks(TextRange stringTextRange) {
+
+        if (stringTextRange.length() == 0 || !getTextRange().contains(stringTextRange)) {
+            return "";
+        }
+
+        List<AtomicTextBlock> textBlocks = getAllAtomicTextBlocksPartiallyInStringBoundary(stringTextRange);
+
+        if (textBlocks.size() == 1) {
+            return textBlocks.get(0).subSequenceWithLineBreaks(stringTextRange);
+        }
+
+        StringBuilder sb = new StringBuilder();
+        AtomicTextBlock firstTextBlock = textBlocks.get(0);
+        sb.append(firstTextBlock.subSequenceWithLineBreaks(new TextRange(stringTextRange.start(), firstTextBlock.getTextRange().end())));
+
+        for (AtomicTextBlock textBlock : textBlocks.subList(1, textBlocks.size() - 1)) {
+            sb.append(textBlock.searchTextWithLineBreaks());
+        }
+
+        var lastTextBlock = textBlocks.get(textBlocks.size() - 1);
+        sb.append(lastTextBlock.subSequenceWithLineBreaks(new TextRange(lastTextBlock.getTextRange().start(), stringTextRange.end())));
+
+        return sb.toString();
+    }
+
+
+    @Override
+    public String toString() {
+
+        return getSearchText();
+    }
+
+
+    @Override
+    public List<TextRange> getBoldTextBoundaries() {
+
+        return getAtomicTextBlocks().stream()
+                .map(AtomicTextBlock::getBoldTextBoundaries)
+                .flatMap(Collection::stream)
+                .toList();
+    }
+
+
+    @Override
+    public List<TextRange> getItalicTextBoundaries() {
+
+        return getAtomicTextBlocks().stream()
+                .map(AtomicTextBlock::getItalicTextBoundaries)
+                .flatMap(Collection::stream)
+                .toList();
+    }
+
+
+    @Override
+    public String getOrientation() {
+
+        if (atomicTextBlocks.isEmpty()) {
+            return "";
+        }
+        return atomicTextBlocks.get(0).getOrientation();
+    }
+
+
+    @Override
+    public int getTextDirection() {
+
+        if (atomicTextBlocks.isEmpty()) {
+            return 0;
+        }
+        return atomicTextBlocks.get(0).getTextDirection();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/textblock/TextBlock.java

@@ -0,0 +1,158 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock;
+
+import static java.lang.String.format;
+
+import java.awt.geom.Rectangle2D;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+
+public interface TextBlock extends CharSequence {
+
+    String getSearchText();
+
+
+    List<AtomicTextBlock> getAtomicTextBlocks();
+
+
+    List<TextRange> getBoldTextBoundaries();
+
+
+    List<TextRange> getItalicTextBoundaries();
+
+
+    String getOrientation();
+
+
+    int getTextDirection();
+
+
+    TextRange getTextRange();
+
+
+    int getNextLinebreak(int fromIndex);
+
+
+    int getPreviousLinebreak(int fromIndex);
+
+
+    List<Integer> getLineBreaks();
+
+
+    Rectangle2D getPosition(int stringIdx);
+
+
+    List<Rectangle2D> getPositions(TextRange stringTextRange);
+
+
+    Map<Page, List<Rectangle2D>> getPositionsPerPage(TextRange stringTextRange);
+
+
+    int numberOfLines();
+
+
+    String subSequenceWithLineBreaks(TextRange stringTextRange);
+
+
+    default String searchTextWithLineBreaks() {
+
+        return subSequenceWithLineBreaks(getTextRange());
+    }
+
+    default int indexOf(String searchTerm) {
+
+        return indexOf(searchTerm, getTextRange().start());
+    }
+
+
+    default Set<Page> getPages() {
+
+        return getAtomicTextBlocks().stream()
+                .map(AtomicTextBlock::getPage)
+                .collect(Collectors.toUnmodifiableSet());
+    }
+
+
+    default Set<Page> getPages(TextRange textRange) {
+
+        return getAtomicTextBlocks().stream()
+                .filter(atomicTextBlock -> atomicTextBlock.getTextRange().intersects(textRange))
+                .map(AtomicTextBlock::getPage)
+                .collect(Collectors.toUnmodifiableSet());
+    }
+
+
+    default int indexOf(String searchTerm, int startOffset) {
+
+        int start = getSearchText().indexOf(searchTerm, startOffset - getTextRange().start());
+        if (start == -1) {
+            return -1;
+        }
+        return start + getTextRange().start();
+    }
+
+
+    default CharSequence getFirstLine() {
+
+        return subSequence(getTextRange().start(), getNextLinebreak(getTextRange().start()));
+    }
+
+
+    default boolean containsBoundary(TextRange textRange) {
+
+        if (textRange.end() < textRange.start()) {
+            throw new IllegalArgumentException(format("Invalid %s, StartIndex must be smaller than EndIndex", textRange));
+        }
+        return getTextRange().contains(textRange);
+    }
+
+
+    default boolean containsIndex(int stringIndex) {
+
+        return getTextRange().containsExclusive(stringIndex);
+    }
+
+
+    default CharSequence subSequence(TextRange textRange) {
+
+        return subSequence(textRange.start(), textRange.end());
+    }
+
+
+    default String buildSummary() {
+
+        String[] words = getSearchText().split(" ");
+        int bound = Math.min(words.length, 4);
+        List<String> list = new ArrayList<>(Arrays.asList(words).subList(0, bound));
+
+        return String.join(" ", list);
+    }
+
+
+    @Override
+    default CharSequence subSequence(int start, int end) {
+
+        return getSearchText().substring(start - getTextRange().start(), end - getTextRange().start());
+    }
+
+
+    @Override
+    default int length() {
+
+        return getTextRange().length();
+    }
+
+
+    @Override
+    default char charAt(int index) {
+
+        return getSearchText().charAt(index - getTextRange().start());
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/graph/textblock/TextBlockCollector.java

@@ -0,0 +1,49 @@
+package com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock;
+
+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 lombok.NoArgsConstructor;
+
+@NoArgsConstructor
+public class TextBlockCollector implements Collector<TextBlock, ConcatenatedTextBlock, TextBlock> {
+
+    @Override
+    public Supplier<ConcatenatedTextBlock> supplier() {
+
+        return ConcatenatedTextBlock::empty;
+    }
+
+
+    @Override
+    public BiConsumer<ConcatenatedTextBlock, TextBlock> accumulator() {
+
+        return ConcatenatedTextBlock::concat;
+    }
+
+
+    @Override
+    public BinaryOperator<ConcatenatedTextBlock> combiner() {
+
+        return ConcatenatedTextBlock::concat;
+    }
+
+
+    @Override
+    public Function<ConcatenatedTextBlock, TextBlock> finisher() {
+
+        return a -> a;
+    }
+
+
+    @Override
+    public Set<Characteristics> characteristics() {
+
+        return Set.of(Characteristics.IDENTITY_FINISH, Characteristics.CONCURRENT);
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/image/ClassifiedImage.java

@@ -2,7 +2,7 @@ package com.knecon.fforesight.service.layoutparser.processor.model.image;
 
 import java.awt.geom.Rectangle2D;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.ImageType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.ImageType;
 
 import lombok.AllArgsConstructor;
 import lombok.Data;

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/outline/OutlineValidationService.java

@@ -1,6 +1,5 @@
 package com.knecon.fforesight.service.layoutparser.processor.model.outline;
 
-import static com.knecon.fforesight.service.layoutparser.processor.model.PageBlockType.TABLE_OF_CONTENTS_HEADLINE;
 import static com.knecon.fforesight.service.layoutparser.processor.model.PageBlockType.getHeadlineNumber;
 
 import java.util.ArrayList;
@@ -19,23 +18,23 @@ import lombok.extern.slf4j.Slf4j;
 
 @Service
 @Slf4j
-public class SectionTreeBuilderService {
+public class OutlineValidationService {
 
     @Observed(name = "OutlineValidationService", contextualName = "create-toc")
-    public SectionTree createSectionTree(ClassificationDocument classificationDocument) {
+    public TableOfContents createToC(ClassificationDocument classificationDocument) {
 
         List<TextPageBlock> headlines = extractHeadlines(classificationDocument);
 
-        List<SectionTreeEntry> mainSections = new ArrayList<>();
-        Map<Integer, SectionTreeEntry> lastItemsPerDepth = new HashMap<>();
-        SectionTreeEntry last = null;
+        List<TableOfContentItem> mainSections = new ArrayList<>();
+        Map<Integer, TableOfContentItem> lastItemsPerDepth = new HashMap<>();
+        TableOfContentItem last = null;
         TreeSet<Integer> depths = new TreeSet<>();
 
         for (TextPageBlock current : headlines) {
             int currentDepth = getHeadlineNumber(current.getClassification());
             Integer parentDepth = depths.floor(currentDepth - 1);
 
-            var tocItem = new SectionTreeEntry(current);
+            var tocItem = new TableOfContentItem(current);
 
             if (parentDepth == null) {
                 mainSections.add(tocItem);
@@ -45,16 +44,14 @@ public class SectionTreeBuilderService {
             } else {
                 assert last != null;
                 int lastDepth = getHeadlineNumber(last.getHeadline().getClassification());
-                if (last.getHeadline().getClassification().equals(TABLE_OF_CONTENTS_HEADLINE) && !current.getClassification().equals(TABLE_OF_CONTENTS_HEADLINE)) {
-                    // headline after toc should always start a main section
-                    parentDepth = 1;
-                } else if (lastDepth < parentDepth) {
+
+                if (lastDepth < parentDepth) {
                     parentDepth = lastDepth;
                 } else if (lastDepth == currentDepth && last.getParent() != null) {
                     parentDepth = getHeadlineNumber(last.getParent().getHeadline().getClassification());
                 }
 
-                SectionTreeEntry parent = lastItemsPerDepth.get(parentDepth);
+                TableOfContentItem parent = lastItemsPerDepth.get(parentDepth);
                 parent.addChild(tocItem);
             }
 
@@ -63,8 +60,7 @@ public class SectionTreeBuilderService {
             depths.add(currentDepth);
         }
 
-        return new SectionTree(mainSections);
-
+        return new TableOfContents(mainSections);
     }
 
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/outline/TOCEnrichmentService.java

@@ -23,28 +23,28 @@ import lombok.extern.slf4j.Slf4j;
 
 @Slf4j
 @Service
-public class SectionTreeEnhancementService {
+public class TOCEnrichmentService {
 
     public void assignSectionBlocksAndImages(ClassificationDocument document) {
 
-        SectionTree toc = document.getSectionTree();
-        Iterator<SectionTreeEntry> iterator = toc.iterator();
-        SectionTreeEntry currentTOCItem = null;
+        TableOfContents toc = document.getTableOfContents();
+        Iterator<TableOfContentItem> iterator = toc.iterator();
+        TableOfContentItem currentTOCItem = null;
         if (iterator.hasNext()) {
             currentTOCItem = iterator.next();
         }
         List<AbstractPageBlock> startBlocks = new ArrayList<>();
         List<ClassifiedImage> startImages = new ArrayList<>();
-        SectionTreeEntry currentSection = null;
+        TableOfContentItem currentSection = null;
         boolean foundFirstHeadline = false;
 
         List<ClassificationHeader> headers = new ArrayList<>();
         List<ClassificationFooter> footers = new ArrayList<>();
         TablePageBlock previousTable = null;
-        List<SectionTreeEntry> lastFoundTOCItems = new ArrayList<>();
+        List<TableOfContentItem> lastFoundTOCItems = new ArrayList<>();
 
         for (ClassificationPage page : document.getPages()) {
-            List<SectionTreeEntry> currentPageTOCItems = new ArrayList<>();
+            List<TableOfContentItem> currentPageTOCItems = new ArrayList<>();
             List<TextPageBlock> header = new ArrayList<>();
             List<TextPageBlock> footer = new ArrayList<>();
             for (AbstractPageBlock current : page.getTextBlocks()) {
@@ -101,7 +101,7 @@ public class SectionTreeEnhancementService {
                 Double xMax = null;
                 Double yMax = null;
 
-                for (SectionTreeEntry tocItem : lastFoundTOCItems) {
+                for (TableOfContentItem tocItem : lastFoundTOCItems) {
                     var headline = tocItem.getHeadline();
 
                     if (headline.getPage() != page.getPageNumber()) {
@@ -168,11 +168,11 @@ public class SectionTreeEnhancementService {
             }
         }
 
-        if (!startBlocks.isEmpty() || !startImages.isEmpty()) {
-            SectionTreeEntry unassigned = new SectionTreeEntry(null);
+        if (!startBlocks.isEmpty()) {
+            TableOfContentItem unassigned = new TableOfContentItem(null);
             unassigned.setSectionBlocks(startBlocks);
             unassigned.setImages(startImages);
-            document.getSectionTree().getMainSections().add(0, unassigned);
+            document.getTableOfContents().getMainSections().add(0, unassigned);
         }
         document.setHeaders(headers);
         document.setFooters(footers);

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/outline/TableOfContentItem.java

@@ -2,12 +2,10 @@ package com.knecon.fforesight.service.layoutparser.processor.model.outline;
 
 import java.util.ArrayList;
 import java.util.List;
-import java.util.Objects;
 import java.util.stream.Collectors;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.GenericSemanticNode;
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
-import com.knecon.fforesight.service.layoutparser.processor.model.PageBlockType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.GenericSemanticNode;
 import com.knecon.fforesight.service.layoutparser.processor.model.image.ClassifiedImage;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPageBlock;
 
@@ -16,18 +14,12 @@ import lombok.EqualsAndHashCode;
 
 @Data
 @EqualsAndHashCode(onlyExplicitlyIncluded = true)
-public class SectionTreeEntry {
-
-    public enum Type {
-        SECTION,
-        SUPER_SECTION,
-        TOC_SECTION
-    }
+public class TableOfContentItem {
 
     @EqualsAndHashCode.Include
     private TextPageBlock headline;
-    private List<SectionTreeEntry> children = new ArrayList<>();
-    private SectionTreeEntry parent;
+    private List<TableOfContentItem> children = new ArrayList<>();
+    private TableOfContentItem parent;
 
     private List<AbstractPageBlock> sectionBlocks = new ArrayList<>();
     private List<ClassifiedImage> images = new ArrayList<>();
@@ -35,32 +27,20 @@ public class SectionTreeEntry {
     private GenericSemanticNode section;
 
 
-    public SectionTreeEntry(TextPageBlock headline) {
+    public TableOfContentItem(TextPageBlock headline) {
 
         this.headline = headline;
     }
 
 
-    public Type getType() {
+    public void addChild(TableOfContentItem tableOfContentItem) {
 
-        if (!Objects.isNull(headline) && headline.getClassification().equals(PageBlockType.TABLE_OF_CONTENTS_HEADLINE)) {
-            return Type.TOC_SECTION;
-        }
-        if (children.isEmpty()) {
-            return Type.SECTION;
-        }
-        return Type.SUPER_SECTION;
+        children.add(tableOfContentItem);
+        tableOfContentItem.setParent(this);
     }
 
 
-    public void addChild(SectionTreeEntry sectionTreeEntry) {
-
-        children.add(sectionTreeEntry);
-        sectionTreeEntry.setParent(this);
-    }
-
-
-    public SectionTreeEntry getSiblingBefore() {
+    public TableOfContentItem getSiblingBefore() {
 
         if (parent != null) {
             int index = parent.getChildren().indexOf(this);
@@ -72,7 +52,7 @@ public class SectionTreeEntry {
     }
 
 
-    public SectionTreeEntry getSiblingAfter() {
+    public TableOfContentItem getSiblingAfter() {
 
         if (parent != null) {
             int index = parent.getChildren().indexOf(this);
@@ -89,7 +69,7 @@ public class SectionTreeEntry {
         if (headline.equals(block)) {
             return true;
         }
-        for (SectionTreeEntry child : children) {
+        for (TableOfContentItem child : children) {
             if (child.contains(block)) {
                 return true;
             }
@@ -98,12 +78,12 @@ public class SectionTreeEntry {
     }
 
 
-    public boolean contains(SectionTreeEntry tocItem) {
+    public boolean contains(TableOfContentItem tocItem) {
 
         if (this.equals(tocItem)) {
             return true;
         }
-        for (SectionTreeEntry child : children) {
+        for (TableOfContentItem child : children) {
             if (child.contains(tocItem)) {
                 return true;
             }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/outline/TableOfContents.java

@@ -14,12 +14,12 @@ import lombok.RequiredArgsConstructor;
 
 @Data
 @RequiredArgsConstructor
-public class SectionTree implements Iterable<SectionTreeEntry> {
+public class TableOfContents implements Iterable<TableOfContentItem> {
 
-    private List<SectionTreeEntry> mainSections = new ArrayList<>();
+    private List<TableOfContentItem> mainSections = new ArrayList<>();
 
 
-    public SectionTree(List<SectionTreeEntry> mainSections) {
+    public TableOfContents(List<TableOfContentItem> mainSections) {
 
         this.mainSections = mainSections;
     }
@@ -28,36 +28,36 @@ public class SectionTree implements Iterable<SectionTreeEntry> {
     public List<TextPageBlock> getAllTextPageBlocks() {
 
         List<TextPageBlock> allTextPageBlocks = new ArrayList<>();
-        for (SectionTreeEntry item : mainSections) {
+        for (TableOfContentItem item : mainSections) {
             collectTextPageBlocks(item, allTextPageBlocks);
         }
         return allTextPageBlocks;
     }
 
 
-    private void collectTextPageBlocks(SectionTreeEntry item, List<TextPageBlock> textPageBlocks) {
+    private void collectTextPageBlocks(TableOfContentItem item, List<TextPageBlock> textPageBlocks) {
 
         textPageBlocks.add(item.getHeadline());
-        for (SectionTreeEntry child : item.getChildren()) {
+        for (TableOfContentItem child : item.getChildren()) {
             collectTextPageBlocks(child, textPageBlocks);
         }
     }
 
 
-    public List<SectionTreeEntry> getAllTableOfContentItems() {
+    public List<TableOfContentItem> getAllTableOfContentItems() {
 
-        List<SectionTreeEntry> allItems = new ArrayList<>();
-        for (SectionTreeEntry item : mainSections) {
+        List<TableOfContentItem> allItems = new ArrayList<>();
+        for (TableOfContentItem item : mainSections) {
             collectTableOfContentItems(item, allItems);
         }
         return allItems;
     }
 
 
-    private void collectTableOfContentItems(SectionTreeEntry item, List<SectionTreeEntry> allItems) {
+    private void collectTableOfContentItems(TableOfContentItem item, List<TableOfContentItem> allItems) {
 
         allItems.add(item);
-        for (SectionTreeEntry child : item.getChildren()) {
+        for (TableOfContentItem child : item.getChildren()) {
             collectTableOfContentItems(child, allItems);
         }
     }
@@ -65,7 +65,7 @@ public class SectionTree implements Iterable<SectionTreeEntry> {
 
     private boolean containsBlock(TextPageBlock block) {
 
-        for (SectionTreeEntry existingItem : this.getMainSections()) {
+        for (TableOfContentItem existingItem : this.getMainSections()) {
             if (existingItem.getHeadline().equals(block) || existingItem.contains(block)) {
                 return true;
             }
@@ -74,9 +74,9 @@ public class SectionTree implements Iterable<SectionTreeEntry> {
     }
 
 
-    private boolean containsItem(SectionTreeEntry tocItem) {
+    private boolean containsItem(TableOfContentItem tocItem) {
 
-        for (SectionTreeEntry existingItem : this.getMainSections()) {
+        for (TableOfContentItem existingItem : this.getMainSections()) {
             if (existingItem.equals(tocItem) || existingItem.contains(tocItem)) {
                 return true;
             }
@@ -86,18 +86,18 @@ public class SectionTree implements Iterable<SectionTreeEntry> {
 
 
     @Override
-    public @NonNull Iterator<SectionTreeEntry> iterator() {
+    public @NonNull Iterator<TableOfContentItem> iterator() {
 
-        return new SectionTreeEntryIterator(mainSections);
+        return new TableOfContentItemIterator(mainSections);
     }
 
 
-    private static class SectionTreeEntryIterator implements Iterator<SectionTreeEntry> {
+    private static class TableOfContentItemIterator implements Iterator<TableOfContentItem> {
 
-        private final Stack<Iterator<SectionTreeEntry>> stack = new Stack<>();
+        private final Stack<Iterator<TableOfContentItem>> stack = new Stack<>();
 
 
-        SectionTreeEntryIterator(List<SectionTreeEntry> mainSections) {
+        TableOfContentItemIterator(List<TableOfContentItem> mainSections) {
 
             stack.push(mainSections.iterator());
         }
@@ -112,10 +112,10 @@ public class SectionTree implements Iterable<SectionTreeEntry> {
 
 
         @Override
-        public SectionTreeEntry next() {
+        public TableOfContentItem next() {
 
             ensureStackTopIsCurrent();
-            SectionTreeEntry currentItem = stack.peek().next();
+            TableOfContentItem currentItem = stack.peek().next();
             if (currentItem.getChildren() != null && !currentItem.getChildren().isEmpty()) {
                 stack.push(currentItem.getChildren()
                                    .iterator());

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/model/text/Word.java

@@ -13,7 +13,6 @@ import java.util.stream.Collectors;
 
 import org.apache.pdfbox.text.TextPosition;
 
-import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.Character;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.TextBoundingBox;
 
 import lombok.AllArgsConstructor;
@@ -39,7 +38,8 @@ public class Word extends TextBoundingBox implements CharSequence {
     private int page;
 
     @Builder.Default
-    private List<Character> characters = new ArrayList<>();
+    private List<RedTextPosition> textPositions = new ArrayList<>();
+
     private boolean isParagraphStart;
     private boolean strikethrough;
     private boolean underline;
@@ -49,9 +49,8 @@ public class Word extends TextBoundingBox implements CharSequence {
 
     public Word(List<TextPosition> textPositions, int pageNumber, boolean isParagraphStart) {
 
-        this.characters = textPositions.stream()
+        this.textPositions = textPositions.stream()
                 .map(RedTextPosition::fromTextPosition)
-                .map(Character::new)
                 .collect(Collectors.toList());
         this.page = pageNumber;
         this.isParagraphStart = isParagraphStart;
@@ -66,9 +65,9 @@ public class Word extends TextBoundingBox implements CharSequence {
     }
 
 
-    public Word(List<Character> textPositions, int page) {
+    public Word(List<RedTextPosition> textPositions, int page) {
 
-        this.characters = new ArrayList<>(textPositions);
+        this.textPositions = textPositions;
         this.page = page;
         calculateBBoxAndHashcode();
     }
@@ -77,7 +76,7 @@ public class Word extends TextBoundingBox implements CharSequence {
     @Override
     public int length() {
 
-        return characters.size();
+        return textPositions.size();
     }
 
 
@@ -102,7 +101,7 @@ public class Word extends TextBoundingBox implements CharSequence {
     public Word subSequence(int start, int end) {
 
         var textPositionSequence = new Word();
-        textPositionSequence.characters = characters.subList(start, end);
+        textPositionSequence.textPositions = textPositions.subList(start, end);
         textPositionSequence.page = page;
         textPositionSequence.dir = dir;
         textPositionSequence.setToBBoxOfComponents(getTextPositions());
@@ -123,59 +122,53 @@ public class Word extends TextBoundingBox implements CharSequence {
 
     public RedTextPosition textPositionAt(int index) {
 
-        return characters.get(index).getTextPosition();
+        return textPositions.get(index);
     }
 
 
     public void add(Word word, RedTextPosition textPosition) {
 
-        this.characters.add(new Character(textPosition));
+        this.textPositions.add(textPosition);
         this.page = word.getPage();
         calculateBBoxAndHashcode();
     }
 
 
-    public void add(Character current) {
-
-        characters.add(current);
-        calculateBBoxAndHashcode();
-    }
-
-
     public void add(TextPosition textPosition) {
 
-        add(new Character(RedTextPosition.fromTextPosition(textPosition)));
+        this.textPositions.add(RedTextPosition.fromTextPosition(textPosition));
+        calculateBBoxAndHashcode();
     }
 
 
     public double getTextHeightNoPadding() {
 
-        return characters.get(0).getTextPosition().getHeightDirAdj();
+        return textPositions.get(0).getHeightDirAdj();
     }
 
 
     public double getTextHeight() {
 
-        return characters.get(0).getTextPosition().getHeightDirAdj() + HEIGHT_PADDING;
+        return textPositions.get(0).getHeightDirAdj() + HEIGHT_PADDING;
     }
 
 
     public String getFont() {
 
-        if (characters.get(0).getTextPosition().getFontName() == null) {
+        if (textPositions.get(0).getFontName() == null) {
             return "none";
         }
 
-        return FONT_CLEANER.matcher(characters.get(0).getTextPosition().getFontName().toLowerCase(Locale.ROOT)).replaceAll("");
+        return FONT_CLEANER.matcher(textPositions.get(0).getFontName().toLowerCase(Locale.ROOT)).replaceAll("");
     }
 
 
     public String getFontStyle() {
 
-        if (characters.get(0).getTextPosition().getFontName() == null) {
+        if (textPositions.get(0).getFontName() == null) {
             return STANDARD;
         }
-        String lowercaseFontName = characters.get(0).getTextPosition().getFontName().toLowerCase(Locale.ROOT);
+        String lowercaseFontName = textPositions.get(0).getFontName().toLowerCase(Locale.ROOT);
 
         if (lowercaseFontName.contains(BOLD) && lowercaseFontName.contains(ITALIC)) {
             return BOLD_ITALIC;
@@ -191,13 +184,13 @@ public class Word extends TextBoundingBox implements CharSequence {
 
     public float getFontSize() {
 
-        return characters.get(0).getTextPosition().getFontSizeInPt();
+        return textPositions.get(0).getFontSizeInPt();
     }
 
 
     public float getSpaceWidth() {
 
-        return characters.get(0).getTextPosition().getWidthOfSpace();
+        return textPositions.get(0).getWidthOfSpace();
     }
 
 
@@ -251,14 +244,6 @@ public class Word extends TextBoundingBox implements CharSequence {
     }
 
 
-    private List<RedTextPosition> getTextPositions() {
-
-        return characters.stream()
-                .map(Character::getTextPosition)
-                .toList();
-    }
-
-
     public void transform(AffineTransform rotateInstance) {
 
         for (RedTextPosition textPosition : getTextPositions()) {

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/python_api/adapter/ImageServiceResponseAdapter.java

@@ -9,9 +9,9 @@ import java.util.Map;
 
 import org.springframework.stereotype.Service;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.ImageType;
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationPage;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.ImageType;
 import com.knecon.fforesight.service.layoutparser.processor.model.image.ClassifiedImage;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.image.ImageServiceResponse;
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/python_api/adapter/VisualLayoutParsingAdapter.java

@@ -8,7 +8,7 @@ import java.util.Map;
 
 import org.springframework.stereotype.Service;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.ImageType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.ImageType;
 import com.knecon.fforesight.service.layoutparser.processor.model.image.ClassifiedImage;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.VisualLayoutParsingBox;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.VisualLayoutParsingResponse;
@@ -79,7 +79,7 @@ public class VisualLayoutParsingAdapter {
                 ClassifiedImage signature = new ClassifiedImage(new Rectangle2D.Float(t.getBox().getX1(),
                         t.getBox().getY1(),
                         t.getBox().getX2() - t.getBox().getX1(),
-                        t.getBox().getY2() - t.getBox().getY1()), ImageType.SIGNATURE, true, false, false, pageNumber, "");
+                        t.getBox().getY2() - t.getBox().getY1()), ImageType.SIGNATURE, true, false, false, pageNumber,"");
 
                 signatures.add(signature);
             }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/SimplifiedSectionTextService.java

@@ -6,11 +6,10 @@ import java.util.stream.Stream;
 
 import org.springframework.stereotype.Service;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.SimplifiedSectionText;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.SimplifiedText;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
 
 import lombok.extern.slf4j.Slf4j;
 
@@ -24,29 +23,18 @@ public class SimplifiedSectionTextService {
                 .stream()
                 .map(this::toSimplifiedSectionText)
                 .toList();
-        List<SimplifiedSectionText> simplifiedHeadersList = document.streamAllSubNodesOfType(NodeType.HEADER)
+        List<SimplifiedSectionText> simplifiedHeadersList = document.getHeaders()
+                .stream()
                 .map(this::toSimplifiedSectionText)
                 .toList();
-        List<SimplifiedSectionText> simplifiedFootersList = document.streamAllSubNodesOfType(NodeType.FOOTER)
+        List<SimplifiedSectionText> simplifiedFootersList = document.getFooters()
+                .stream()
                 .map(this::toSimplifiedSectionText)
                 .toList();
         List<SimplifiedSectionText> simplifiedText = Stream.of(simplifiedMainSectionsList, simplifiedHeadersList, simplifiedFootersList)
                 .flatMap(List::stream)
                 .collect(Collectors.toList());
-        return SimplifiedText.builder()
-                .numberOfPages(document.getNumberOfPages())
-                .sectionTexts(simplifiedText)
-                .mainSectionNumbers(document.getAllSections()
-                                            .stream()
-                                            .map(this::getSectionNumber)
-                                            .toList())
-                .headerSectionNumbers(document.streamAllSubNodesOfType(NodeType.HEADER)
-                                              .map(this::getSectionNumber)
-                                              .toList())
-                .footerSectionNumbers(document.streamAllSubNodesOfType(NodeType.FOOTER)
-                                              .map(this::getSectionNumber)
-                                              .toList())
-                .build();
+        return SimplifiedText.builder().numberOfPages(document.getNumberOfPages()).sectionTexts(simplifiedText).build();
     }
 
 
@@ -61,13 +49,4 @@ public class SimplifiedSectionTextService {
                 .build();
     }
 
-
-    private String getSectionNumber(SemanticNode semanticNode) {
-
-        return semanticNode.getTreeId()
-                .stream()
-                .map(String::valueOf)
-                .collect(Collectors.joining("."));
-    }
-
 }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/blockification/BlockificationPostprocessingService.java

@@ -10,15 +10,13 @@ import java.util.Locale;
 import org.apache.commons.text.similarity.LevenshteinDistance;
 import org.springframework.stereotype.Service;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.LayoutEngine;
-import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationDocument;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.LayoutEngine;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationPage;
 import com.knecon.fforesight.service.layoutparser.processor.model.PageBlockType;
 import com.knecon.fforesight.service.layoutparser.processor.model.SectionIdentifier;
 import com.knecon.fforesight.service.layoutparser.processor.model.outline.OutlineObject;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.Word;
-import com.knecon.fforesight.service.layoutparser.processor.utils.PageInformation;
 import com.knecon.fforesight.service.layoutparser.processor.utils.TextNormalizationUtilities;
 
 import lombok.Data;
@@ -29,24 +27,6 @@ public class BlockificationPostprocessingService {
     private static final float STRING_SIMILARITY_THRESHOLD = 0.1f;
 
 
-    public void findHeadlinesFromOutline(ClassificationDocument classificationDocument, int pageNumber, ClassificationPage classificationPage, PageInformation pageInformation) {
-
-        OutlineObject lastProcessedOutlineObject = null;
-        List<OutlineObject> outlineObjects = classificationDocument.getOutlineObjectTree().getOutlineObjectsPerPage().getOrDefault(pageNumber, new ArrayList<>());
-
-        OutlineObject notFoundOutlineObject = null;
-        if (lastProcessedOutlineObject != null && !lastProcessedOutlineObject.isFound()) {
-            lastProcessedOutlineObject.resetPoint();
-            notFoundOutlineObject = lastProcessedOutlineObject;
-        }
-        if (!outlineObjects.isEmpty()) {
-            classificationPage.setOutlineObjects(outlineObjects);
-            lastProcessedOutlineObject = sanitizeOutlineBlocks(classificationPage, notFoundOutlineObject);
-        }
-        classificationDocument.getLayoutDebugLayer().addOutlineObjects(outlineObjects, pageInformation);
-    }
-
-
     public OutlineObject sanitizeOutlineBlocks(ClassificationPage classificationPage, OutlineObject notFoundOutlineObject) {
 
         List<OutlineObject> outlineObjects = classificationPage.getOutlineObjects();
@@ -349,8 +329,8 @@ public class BlockificationPostprocessingService {
 
         if (index > 0) {
             in = createSubSequence(sequence, 0, index);
-        } else if (endIndex < sequence.length()) {
-            in = createSubSequence(sequence, endIndex, sequence.length());
+        } else if (endIndex < sequence.getTextPositions().size()) {
+            in = createSubSequence(sequence, endIndex, sequence.getTextPositions().size());
         }
 
         return new SplitSequenceResult(in, out);
@@ -359,7 +339,7 @@ public class BlockificationPostprocessingService {
 
     private static Word createSubSequence(Word sequence, int start, int end) {
 
-        Word newSeq = new Word(new ArrayList<>(sequence.getCharacters().subList(start, end)), sequence.getPage());
+        Word newSeq = new Word(new ArrayList<>(sequence.getTextPositions().subList(start, end)), sequence.getPage());
         newSeq.setParagraphStart(sequence.isParagraphStart());
         return newSeq;
     }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/blockification/DocstrumBlockificationService.java

@@ -77,7 +77,7 @@ public class DocstrumBlockificationService {
                     .forEach(line -> {
                         line.getWords()
                                 .forEach(word -> {
-                                    words.add(new Word(word.getCharacters(), word.getPage()));
+                                    words.add(new Word(word.getTextPositions(), word.getPage()));
                                 });
                     });
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/blockification/DocuMineBlockificationService.java

@@ -9,7 +9,7 @@ import java.util.stream.Collectors;
 
 import org.springframework.stereotype.Service;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.LayoutEngine;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.LayoutEngine;
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationPage;
 import com.knecon.fforesight.service.layoutparser.processor.model.Orientation;

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/classification/ClarifyndClassificationService.java

@@ -81,7 +81,7 @@ public class ClarifyndClassificationService {
                    && (textBlock.getMostPopularWordStyle().equals("bold")
                        || !document.getFontStyleCounter().getCountPerValue().containsKey("bold")
                           && textBlock.getMostPopularWordFontSize() > document.getFontSizeCounter().getMostPopular() + 1)
-                   && textBlock.getWords().get(0).getFontSize()>= textBlock.getMostPopularWordFontSize()) {
+                   && textBlock.getWords().get(0).getTextPositions().get(0).getFontSizeInPt() >= textBlock.getMostPopularWordFontSize()) {
 
             PageBlockType headlineType = HeadlineClassificationService.headlineClassByFontSize(textBlock, headlineFontSizes);
             headlineClassificationService.classifyHeadline(textBlock, headlineType);
@@ -91,7 +91,7 @@ public class ClarifyndClassificationService {
                    && textBlock.getMostPopularWordStyle().equals("bold")
                    && !document.getFontStyleCounter().getMostPopular().equals("bold")
                    && PositionUtils.getApproxLineCount(textBlock) < 2.9
-                   && textBlock.getWords().get(0).getFontSize() >= textBlock.getMostPopularWordFontSize()) {
+                   && textBlock.getWords().get(0).getTextPositions().get(0).getFontSizeInPt() >= textBlock.getMostPopularWordFontSize()) {
 
             PageBlockType headlineType = HeadlineClassificationService.headlineClassByFontSize(textBlock, headlineFontSizes);
             headlineClassificationService.classifyHeadline(textBlock, headlineType);

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/classification/RedactManagerClassificationService.java

@@ -5,11 +5,11 @@ import java.util.regex.Pattern;
 
 import org.springframework.stereotype.Service;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.ImageType;
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationDocument;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationPage;
 import com.knecon.fforesight.service.layoutparser.processor.model.PageBlockType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.ImageType;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.utils.PositionUtils;
 
@@ -94,7 +94,7 @@ public class RedactManagerClassificationService {
                    && (textBlock.getMostPopularWordStyle().equals("bold")
                        || !document.getFontStyleCounter().getCountPerValue().containsKey("bold")
                           && textBlock.getMostPopularWordFontSize() > document.getFontSizeCounter().getMostPopular() + 1)
-                   && textBlock.getWords().get(0).getFontSize() >= textBlock.getMostPopularWordFontSize()) {
+                   && textBlock.getWords().get(0).getTextPositions().get(0).getFontSizeInPt() >= textBlock.getMostPopularWordFontSize()) {
 
             PageBlockType headlineType = HeadlineClassificationService.headlineClassByFontSize(textBlock, headlineFontSizes);
             headlineClassificationService.classifyHeadline(textBlock, headlineType);
@@ -104,7 +104,7 @@ public class RedactManagerClassificationService {
                    && textBlock.getMostPopularWordStyle().equals("bold")
                    && !document.getFontStyleCounter().getMostPopular().equals("bold")
                    && PositionUtils.getApproxLineCount(textBlock) < 2.9
-                   && textBlock.getWords().get(0).getFontSize() >= textBlock.getMostPopularWordFontSize()) {
+                   && textBlock.getWords().get(0).getTextPositions().get(0).getFontSizeInPt() >= textBlock.getMostPopularWordFontSize()) {
 
             PageBlockType headlineType = HeadlineClassificationService.headlineClassByFontSize(textBlock, headlineFontSizes);
             headlineClassificationService.classifyHeadline(textBlock, headlineType);

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/classification/TableOfContentsClassificationService.java

@@ -61,7 +61,7 @@ public class TableOfContentsClassificationService {
 
             if (end > i + 1) {
                 if (textBlock.textBlock().getClassification() == null) {
-                    textBlock.textBlock().setClassification(PageBlockType.TABLE_OF_CONTENTS_HEADLINE);
+                    textBlock.textBlock().setClassification(PageBlockType.H1);
                 }
                 i = end;
             }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/factory/DocumentGraphFactory.java

@@ -12,37 +12,33 @@ import java.util.LinkedList;
 import java.util.List;
 import java.util.Map;
 import java.util.NoSuchElementException;
-import java.util.Objects;
 import java.util.Optional;
 import java.util.Set;
 import java.util.stream.Collectors;
 
-import com.iqser.red.service.redaction.v1.server.model.document.DocumentTree;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.AbstractSemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.DuplicatedParagraph;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Footer;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.GenericSemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Header;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Headline;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Image;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Paragraph;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.TableOfContentsItem;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.AtomicTextBlock;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
 import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationDocument;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationFooter;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationHeader;
 import com.knecon.fforesight.service.layoutparser.processor.model.ClassificationPage;
-import com.knecon.fforesight.service.layoutparser.processor.model.DocumentWithVisualization;
-import com.knecon.fforesight.service.layoutparser.processor.model.PageBlockType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.DocumentTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.AbstractSemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.DuplicatedParagraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Footer;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.GenericSemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Header;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Headline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Image;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Paragraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.image.ClassifiedImage;
-import com.knecon.fforesight.service.layoutparser.processor.model.outline.SectionTreeEntry;
+import com.knecon.fforesight.service.layoutparser.processor.model.outline.TableOfContentItem;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.utils.IdBuilder;
 import com.knecon.fforesight.service.layoutparser.processor.utils.TextPositionOperations;
@@ -59,10 +55,12 @@ import lombok.extern.slf4j.Slf4j;
 @Slf4j
 public class DocumentGraphFactory {
 
-    public DocumentWithVisualization buildDocumentGraph(LayoutParsingType layoutParsingType, ClassificationDocument document) {
+    public Document buildDocumentGraph(LayoutParsingType layoutParsingType, ClassificationDocument document) {
 
         Document documentGraph = new Document();
 
+        documentGraph.setLayoutDebugLayer(document.getLayoutDebugLayer());
+
         Context context = new Context(documentGraph);
 
         document.getPages()
@@ -76,7 +74,7 @@ public class DocumentGraphFactory {
         documentGraph.setTextBlock(documentGraph.getTextBlock());
         addTextBlocksToPages(documentGraph);
 
-        return new DocumentWithVisualization(documentGraph, document.getLayoutDebugLayer());
+        return documentGraph;
     }
 
 
@@ -96,16 +94,16 @@ public class DocumentGraphFactory {
 
     private void addSections(LayoutParsingType layoutParsingType, ClassificationDocument classificationDocument, Context context, Document document) {
 
-        for (SectionTreeEntry sectionTreeEntry : classificationDocument.getSectionTree()) {
-            GenericSemanticNode parent = sectionTreeEntry.getParent() == null ? null : sectionTreeEntry.getParent().getSection();
+        for (TableOfContentItem tocItem : classificationDocument.getTableOfContents()) {
+            GenericSemanticNode parent = tocItem.getParent() == null ? null : tocItem.getParent().getSection();
             Optional<GenericSemanticNode> section = SectionNodeFactory.addSection(layoutParsingType,
                                                                                   parent,
-                                                                                  sectionTreeEntry.getType(),
-                                                                                  sectionTreeEntry.getNonEmptySectionBlocks(),
-                                                                                  sectionTreeEntry.getImages(),
+                                                                                  tocItem.getChildren().isEmpty(),
+                                                                                  tocItem.getNonEmptySectionBlocks(),
+                                                                                  tocItem.getImages(),
                                                                                   context,
                                                                                   document);
-            sectionTreeEntry.setSection(section.orElse(null));
+            tocItem.setSection(section.orElse(null));
         }
     }
 
@@ -123,8 +121,6 @@ public class DocumentGraphFactory {
             node = Headline.builder().documentTree(context.getDocumentTree()).build();
         } else if (originalTextBlock.isToDuplicate() && layoutParsingType.equals(LayoutParsingType.REDACT_MANAGER)) {
             node = DuplicatedParagraph.builder().documentTree(context.getDocumentTree()).build();
-        } else if (Objects.equals(originalTextBlock.getClassification(), PageBlockType.TABLE_OF_CONTENTS_ITEM)) {
-            node = TableOfContentsItem.builder().documentTree(context.getDocumentTree()).build();
         } else {
             node = Paragraph.builder().documentTree(context.getDocumentTree()).build();
         }
@@ -264,6 +260,7 @@ public class DocumentGraphFactory {
         DocumentTree documentTree;
         Map<Page, Integer> pages;
         List<AbstractSemanticNode> sections;
+        List<ClassifiedImage> images;
         TextBlockFactory textBlockFactory;
 
 
@@ -272,20 +269,14 @@ public class DocumentGraphFactory {
             documentTree = new DocumentTree(document);
             pages = new HashMap<>();
             sections = new LinkedList<>();
+            images = new LinkedList<>();
             textBlockFactory = new TextBlockFactory();
         }
 
 
         public void buildAndAddPageWithCounter(ClassificationPage classificationPage) {
 
-            Page page = Page.builder()
-                    .height((int) classificationPage.getPageHeight())
-                    .width((int) classificationPage.getPageWidth())
-                    .number(classificationPage.getPageNumber())
-                    .rotation(classificationPage.getRotation())
-                    .textBlocksOnPage(new LinkedList<>())
-                    .build();
-
+            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);

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/factory/SearchTextWithTextPositionDto.java

@@ -4,7 +4,7 @@ import java.awt.geom.Rectangle2D;
 import java.util.Collections;
 import java.util.List;
 
-import com.iqser.red.service.redaction.v1.server.model.document.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
 
 import lombok.AccessLevel;
 import lombok.Builder;
@@ -31,8 +31,6 @@ public class SearchTextWithTextPositionDto {
                 .lineBreaks(Collections.emptyList())
                 .positions(Collections.emptyList())
                 .stringIdxToPositionIdx(Collections.emptyList())
-                .boldTextBoundaries(Collections.emptyList())
-                .italicTextBoundaries(Collections.emptyList())
                 .build();
     }
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/factory/SearchTextWithTextPositionFactory.java

@@ -10,8 +10,7 @@ import java.util.List;
 import java.util.Locale;
 import java.util.Objects;
 
-import com.iqser.red.service.redaction.v1.server.model.document.TextRange;
-import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.Character;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.RedTextPosition;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.Word;
 
@@ -32,19 +31,19 @@ public class SearchTextWithTextPositionFactory {
     public SearchTextWithTextPositionDto buildSearchTextToTextPositionDto(List<Word> sequences) {
 
         if (sequences.isEmpty() || sequences.stream()
-                .allMatch(sequence -> sequence.getCharacters().isEmpty())) {
+                .allMatch(sequence -> sequence.getTextPositions().isEmpty())) {
             return SearchTextWithTextPositionDto.empty();
         }
 
         Context context = new Context();
 
-        RedTextPosition currentTextPosition = sequences.get(0).getCharacters().get(0).getTextPosition();
+        RedTextPosition currentTextPosition = sequences.get(0).getTextPositions().get(0);
         RedTextPosition previousTextPosition = RedTextPosition.builder().unicode(" ").bBoxDirAdj(currentTextPosition.getBBoxDirAdj()).build();
 
         for (Word word : sequences) {
-            for (int i = 0; i < word.getCharacters().size(); ++i) {
+            for (int i = 0; i < word.getTextPositions().size(); ++i) {
 
-                currentTextPosition = word.getCharacters().get(i).getTextPosition();
+                currentTextPosition = word.getTextPositions().get(i);
                 if (isLineBreak(currentTextPosition, previousTextPosition)) {
                     removeHyphenLinebreaks(context);
                     context.lineBreaksStringIdx.add(context.stringIdx);
@@ -67,9 +66,8 @@ public class SearchTextWithTextPositionFactory {
         }
 
         List<Rectangle2D> positions = sequences.stream()
-                .map(Word::getCharacters)
+                .map(Word::getTextPositions)
                 .flatMap(Collection::stream)
-                .map(Character::getTextPosition)
                 .map(RedTextPosition::getBBoxPdf)
                 .toList();
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/factory/SectionNodeFactory.java

@@ -10,16 +10,14 @@ import java.util.List;
 import java.util.Optional;
 import java.util.Set;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.AbstractSemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.GenericSemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Section;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SuperSection;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.TableOfContents;
 import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.AbstractSemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.GenericSemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Section;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SuperSection;
 import com.knecon.fforesight.service.layoutparser.processor.model.image.ClassifiedImage;
-import com.knecon.fforesight.service.layoutparser.processor.model.outline.SectionTreeEntry;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.TablePageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.TextPageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.utils.TableMergingUtility;
@@ -31,7 +29,7 @@ public class SectionNodeFactory {
 
     public Optional<GenericSemanticNode> addSection(LayoutParsingType layoutParsingType,
                                                     GenericSemanticNode parentNode,
-                                                    SectionTreeEntry.Type type,
+                                                    boolean isLeaf,
                                                     List<AbstractPageBlock> pageBlocks,
                                                     List<ClassifiedImage> images,
                                                     DocumentGraphFactory.Context context,
@@ -50,11 +48,12 @@ public class SectionNodeFactory {
             return Optional.empty();
         }
 
-        AbstractSemanticNode section = switch (type) {
-            case SECTION -> Section.builder().documentTree(context.getDocumentTree()).build();
-            case SUPER_SECTION -> SuperSection.builder().documentTree(context.getDocumentTree()).build();
-            case TOC_SECTION -> TableOfContents.builder().documentTree(context.getDocumentTree()).build();
-        };
+        AbstractSemanticNode section;
+        if (isLeaf) {
+            section = Section.builder().documentTree(context.getDocumentTree()).build();
+        } else {
+            section = SuperSection.builder().documentTree(context.getDocumentTree()).build();
+        }
 
         context.getSections().add(section);
 
@@ -65,14 +64,13 @@ public class SectionNodeFactory {
         if (containsTablesAndTextBlocks) {
             splitPageBlocksIntoSubSections(pageBlocks).forEach(subSectionPageBlocks -> addSection(layoutParsingType,
                                                                                                   section,
-                                                                                                  SectionTreeEntry.Type.SECTION,
+                                                                                                  true,
                                                                                                   subSectionPageBlocks,
                                                                                                   emptyList(),
                                                                                                   context,
                                                                                                   document));
-        } else if (type.equals(SectionTreeEntry.Type.SUPER_SECTION)) {
-            // If a SuperSection contains more blocks than just a headline, we add a Section which contains the remaining textblocks.
-            addSection(layoutParsingType, section, SectionTreeEntry.Type.SECTION, pageBlocks, emptyList(), context, document);
+        } else if (!isLeaf) {
+            addSection(layoutParsingType, section, true, pageBlocks, emptyList(), context, document);
         } else {
             addTablesAndParagraphsAndHeadlinesToSection(layoutParsingType, pageBlocks, context, section, document);
         }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/factory/TableNodeFactory.java

@@ -6,15 +6,14 @@ import java.util.Collection;
 import java.util.List;
 import java.util.stream.Collectors;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.GenericSemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Table;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.TableCell;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.TextBlock;
 import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
-import com.knecon.fforesight.service.layoutparser.processor.model.outline.SectionTreeEntry;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.GenericSemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.Cell;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.TablePageBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.Word;
@@ -121,7 +120,7 @@ public class TableNodeFactory {
         } else if (firstTextBlockIsHeadline(cell)) {
             SectionNodeFactory.addSection(layoutParsingType,
                                           tableCell,
-                                          SectionTreeEntry.Type.SECTION,
+                                          true,
                                           cell.getTextBlocks()
                                                   .stream()
                                                   .map(tb -> (AbstractPageBlock) tb)

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/factory/TextBlockFactory.java

@@ -2,11 +2,10 @@ package com.knecon.fforesight.service.layoutparser.processor.services.factory;
 
 import java.util.List;
 
-import com.iqser.red.service.redaction.v1.server.model.document.TextRange;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.AtomicTextBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.Word;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
 
 import lombok.AccessLevel;
 import lombok.experimental.FieldDefaults;
@@ -32,20 +31,29 @@ public class TextBlockFactory {
         stringOffset += searchTextWithTextPositionDto.getSearchText().length();
         long idx = textBlockIdx;
         textBlockIdx++;
-        String searchText = searchTextWithTextPositionDto.getSearchText();
-        return AtomicTextBlock.builder()
-                .id(idx)
-                .parent(parent)
-                .searchText(searchText)
-                .numberOnPage(numberOnPage)
-                .page(page)
-                .lineBreaks(searchTextWithTextPositionDto.getLineBreaks())
-                .positions(searchTextWithTextPositionDto.getPositions())
-                .stringIdxToPositionIdx(searchTextWithTextPositionDto.getStringIdxToPositionIdx())
-                .textRange(new TextRange(offset, offset + searchText.length()))
-                .boldTextRanges(searchTextWithTextPositionDto.getBoldTextBoundaries())
-                .italicTextRanges(searchTextWithTextPositionDto.getItalicTextBoundaries())
-                .build();
+        String orientation;
+        int textRotation;
+        if (sequences.isEmpty()) {
+            orientation = null;
+            textRotation = 0;
+        } else {
+            orientation = sequences.get(0).getDir().toString();
+            textRotation = sequences.get(0).getDir().getRotation();
+        }
+        var atb = AtomicTextBlock.fromSearchTextWithTextPosition(searchTextWithTextPositionDto.getSearchText(),
+                                                                 searchTextWithTextPositionDto.getLineBreaks(),
+                                                                 searchTextWithTextPositionDto.getBoldTextBoundaries(),
+                                                                 searchTextWithTextPositionDto.getItalicTextBoundaries(),
+                                                                 searchTextWithTextPositionDto.getPositions(),
+                                                                 searchTextWithTextPositionDto.getStringIdxToPositionIdx(),
+                                                                 idx,
+                                                                 parent,
+                                                                 numberOnPage,
+                                                                 page,
+                                                                 offset,
+                                                                 orientation,
+                                                                 textRotation);
+        return atb;
     }
 
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/mapper/DocumentDataMapper.java

@@ -0,0 +1,157 @@
+package com.knecon.fforesight.service.layoutparser.processor.services.mapper;
+
+import java.awt.geom.Rectangle2D;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentData;
+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.service.layoutparser.internal.api.data.redaction.LayoutEngine;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.DocumentTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.DuplicatedParagraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Image;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+
+import lombok.experimental.UtilityClass;
+
+@UtilityClass
+public class DocumentDataMapper {
+
+    public DocumentData toDocumentData(Document document) {
+
+        List<DocumentTextData> documentTextData = document.streamTerminalTextBlocksInOrder()
+                .flatMap(textBlock -> textBlock.getAtomicTextBlocks().stream())
+                .distinct()
+                .map(DocumentDataMapper::toAtomicTextBlockData)
+                .toList();
+
+        List<DocumentPositionData> atomicPositionBlockData = document.streamTerminalTextBlocksInOrder()
+                .flatMap(textBlock -> textBlock.getAtomicTextBlocks().stream())
+                .distinct()
+                .map(DocumentDataMapper::toAtomicPositionBlockData)
+                .toList();
+
+        Set<Long> nonEmptyTextBlocks = documentTextData.stream().mapToLong(DocumentTextData::getId).boxed().collect(Collectors.toSet());
+
+        List<DocumentPage> documentPageData = document.getPages().stream().map(DocumentDataMapper::toPageData).toList();
+        DocumentStructure tableOfContentsData = toDocumentTreeData(document.getDocumentTree());
+        return DocumentData.builder()
+                .documentTextData(documentTextData.toArray(new DocumentTextData[0]))
+                .documentPositions(atomicPositionBlockData.toArray(new DocumentPositionData[0]))
+                .documentPages(documentPageData.toArray(new DocumentPage[0]))
+                .documentStructure(tableOfContentsData)
+                .build();
+    }
+
+
+    private DocumentStructure toDocumentTreeData(DocumentTree documentTree) {
+
+        return new DocumentStructure(toEntryData(documentTree.getRoot()));
+    }
+
+
+    private DocumentStructure.EntryData toEntryData(DocumentTree.Entry entry) {
+
+        Long[] atomicTextBlocks;
+
+        if (entry.getNode().isLeaf()) {
+            atomicTextBlocks = toAtomicTextBlockIds(entry.getNode().getLeafTextBlock());
+        } else {
+            atomicTextBlocks = new Long[]{};
+        }
+
+        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());
+            case PARAGRAPH ->
+                    entry.getNode() instanceof DuplicatedParagraph duplicatedParagraph ? PropertiesMapper.buildDuplicateParagraphProperties(duplicatedParagraph) : new HashMap<>();
+            default -> new HashMap<>();
+        };
+
+        DocumentStructure.EntryData.EntryDataBuilder documentBuilder = DocumentStructure.EntryData.builder()
+                .treeId(toPrimitiveIntArray(entry.getTreeId()))
+                .children(entry.getChildren().stream().map(DocumentDataMapper::toEntryData).toList())
+                .type(entry.getType())
+                .atomicBlockIds(atomicTextBlocks)
+                .pageNumbers(entry.getNode().getPages().stream().map(Page::getNumber).map(Integer::longValue).toArray(Long[]::new))
+                .properties(properties);
+        if (entry.getNode() != null) {
+            documentBuilder.engines(entry.getNode().getEngines());
+        } else {
+            documentBuilder.engines(new HashSet<>(Set.of(LayoutEngine.ALGORITHM)));
+        }
+        return documentBuilder.build();
+    }
+
+
+    private Long[] toAtomicTextBlockIds(TextBlock textBlock) {
+
+        return textBlock.getAtomicTextBlocks().stream().map(AtomicTextBlock::getId).toArray(Long[]::new);
+    }
+
+
+    private DocumentPage toPageData(Page p) {
+
+        return DocumentPage.builder().rotation(p.getRotation()).height(p.getHeight()).width(p.getWidth()).number(p.getNumber()).build();
+    }
+
+
+    private DocumentTextData toAtomicTextBlockData(AtomicTextBlock atomicTextBlock) {
+
+        return DocumentTextData.builder()
+                .id(atomicTextBlock.getId())
+                .page(atomicTextBlock.getPage().getNumber().longValue())
+                .searchText(atomicTextBlock.getSearchText())
+                .numberOnPage(atomicTextBlock.getNumberOnPage())
+                .start(atomicTextBlock.getTextRange().start())
+                .end(atomicTextBlock.getTextRange().end())
+                .lineBreaks(toPrimitiveIntArray(atomicTextBlock.getLineBreaks()))
+                .build();
+    }
+
+
+    private DocumentPositionData toAtomicPositionBlockData(AtomicTextBlock atomicTextBlock) {
+
+        return DocumentPositionData.builder()
+                .id(atomicTextBlock.getId())
+                .positions(toPrimitiveFloatMatrix(atomicTextBlock.getPositions()))
+                .stringIdxToPositionIdx(toPrimitiveIntArray(atomicTextBlock.getStringIdxToPositionIdx()))
+                .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()};
+    }
+
+
+    private int[] toPrimitiveIntArray(List<Integer> list) {
+
+        return list.stream().mapToInt(Integer::intValue).toArray();
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/mapper/DocumentGraphMapper.java

@@ -0,0 +1,227 @@
+package com.knecon.fforesight.service.layoutparser.processor.services.mapper;
+
+import java.util.Arrays;
+import java.util.HashSet;
+import java.util.LinkedList;
+import java.util.List;
+import java.util.Map;
+import java.util.NoSuchElementException;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentData;
+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.service.layoutparser.processor.model.graph.DocumentTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.DuplicatedParagraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Footer;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Header;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Headline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Image;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Paragraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Section;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SuperSection;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlockCollector;
+
+import lombok.experimental.UtilityClass;
+
+@UtilityClass
+public class DocumentGraphMapper {
+
+    public Document toDocumentGraph(DocumentData documentData) {
+
+        Document document = new Document();
+        DocumentTree documentTree = new DocumentTree(document);
+        Context context = new Context(documentData, documentTree);
+
+        context.pages.addAll(Arrays.stream(documentData.getDocumentPages())
+                                     .map(DocumentGraphMapper::buildPage)
+                                     .toList());
+
+        context.documentTree.getRoot().getChildren().addAll(buildEntries(documentData.getDocumentStructure().getRoot().getChildren(), context));
+
+        document.setDocumentTree(context.documentTree);
+        document.setPages(new HashSet<>(context.pages));
+        document.setNumberOfPages(documentData.getDocumentPages().length);
+
+        document.setTextBlock(document.getTextBlock());
+        return document;
+    }
+
+
+    private List<DocumentTree.Entry> buildEntries(List<DocumentStructure.EntryData> entries, Context context) {
+
+        List<DocumentTree.Entry> newEntries = new LinkedList<>();
+        for (DocumentStructure.EntryData entryData : entries) {
+
+            List<Page> pages = Arrays.stream(entryData.getPageNumbers())
+                    .map(pageNumber -> getPage(pageNumber, context))
+                    .toList();
+
+            SemanticNode node = switch (entryData.getType()) {
+                case SECTION -> buildSection(context);
+                case SUPER_SECTION -> buildSuperSection(context);
+                case PARAGRAPH -> buildParagraph(context, entryData.getProperties());
+                case HEADLINE -> buildHeadline(context);
+                case HEADER -> buildHeader(context);
+                case FOOTER -> buildFooter(context);
+                case TABLE -> buildTable(context, entryData.getProperties());
+                case TABLE_CELL -> buildTableCell(context, entryData.getProperties());
+                case IMAGE -> buildImage(context, entryData.getProperties(), entryData.getPageNumbers());
+                default -> throw new UnsupportedOperationException("Not yet implemented for type " + entryData.getType());
+            };
+
+            if (entryData.getAtomicBlockIds().length > 0) {
+                TextBlock textBlock = toTextBlock(entryData.getAtomicBlockIds(), context, node);
+                node.setLeafTextBlock(textBlock);
+                switch (entryData.getType()) {
+                    case HEADER -> pages.forEach(page -> page.setHeader((Header) node));
+                    case FOOTER -> pages.forEach(page -> page.setFooter((Footer) node));
+                    case IMAGE -> pages.forEach(page -> page.getImages().add((Image) node));
+                    default -> textBlock.getAtomicTextBlocks()
+                            .forEach(atb -> atb.getPage().getTextBlocksOnPage().add(atb));
+                }
+            }
+            List<Integer> treeId = Arrays.stream(entryData.getTreeId()).boxed()
+                    .toList();
+            node.setTreeId(treeId);
+
+            newEntries.add(DocumentTree.Entry.builder().treeId(treeId).children(buildEntries(entryData.getChildren(), context)).node(node).build());
+        }
+        return newEntries;
+    }
+
+
+    private Headline buildHeadline(Context context) {
+
+        return Headline.builder().documentTree(context.documentTree).build();
+    }
+
+
+    private Image buildImage(Context context, Map<String, String> properties, Long[] pageNumbers) {
+
+        assert pageNumbers.length == 1;
+        Page page = getPage(pageNumbers[0], context);
+        var builder = Image.builder();
+        PropertiesMapper.parseImageProperties(properties, builder);
+        return builder.documentTree(context.documentTree).page(page).build();
+    }
+
+
+    private TableCell buildTableCell(Context context, Map<String, String> properties) {
+
+        TableCell.TableCellBuilder<?, ?> builder = TableCell.builder();
+        PropertiesMapper.parseTableCellProperties(properties, builder);
+        return builder.documentTree(context.documentTree).build();
+    }
+
+
+    private Table buildTable(Context context, Map<String, String> properties) {
+
+        Table.TableBuilder builder = Table.builder();
+        PropertiesMapper.parseTableProperties(properties, builder);
+        return builder.documentTree(context.documentTree).build();
+    }
+
+
+    private Footer buildFooter(Context context) {
+
+        return Footer.builder().documentTree(context.documentTree).build();
+    }
+
+
+    private Header buildHeader(Context context) {
+
+        return Header.builder().documentTree(context.documentTree).build();
+    }
+
+
+    private Section buildSection(Context context) {
+
+        return Section.builder().documentTree(context.documentTree).build();
+    }
+
+
+    private SuperSection buildSuperSection(Context context) {
+
+        return SuperSection.builder().documentTree(context.documentTree).build();
+    }
+
+
+    private Paragraph buildParagraph(Context context, Map<String, String> properties) {
+
+        if (PropertiesMapper.isDuplicateParagraph(properties)) {
+
+            DuplicatedParagraph duplicatedParagraph = DuplicatedParagraph.builder().documentTree(context.documentTree).build();
+
+            Long[] unsortedTextblockIds = PropertiesMapper.getUnsortedTextblockIds(properties);
+            duplicatedParagraph.setUnsortedLeafTextBlock(toTextBlock(unsortedTextblockIds, context, duplicatedParagraph));
+            return duplicatedParagraph;
+
+        }
+
+        return Paragraph.builder().documentTree(context.documentTree).build();
+    }
+
+
+    private TextBlock toTextBlock(Long[] atomicTextBlockIds, Context context, SemanticNode parent) {
+
+        return Arrays.stream(atomicTextBlockIds)
+                .map(atomicTextBlockId -> getAtomicTextBlock(context, parent, atomicTextBlockId))
+                .collect(new TextBlockCollector());
+    }
+
+
+    private AtomicTextBlock getAtomicTextBlock(Context context, SemanticNode parent, Long atomicTextBlockId) {
+
+        return AtomicTextBlock.fromAtomicTextBlockData(context.documentTextDataBlockData.get(Math.toIntExact(atomicTextBlockId)),
+                                                       context.atomicPositionBlockData.get(Math.toIntExact(atomicTextBlockId)),
+                                                       parent,
+                                                       getPage(context.documentTextDataBlockData.get(Math.toIntExact(atomicTextBlockId)).getPage(), context));
+    }
+
+
+    private Page buildPage(DocumentPage p) {
+
+        return Page.builder().rotation(p.getRotation()).height(p.getHeight()).width(p.getWidth()).number(p.getNumber()).textBlocksOnPage(new LinkedList<>()).build();
+    }
+
+
+    private Page getPage(Long pageIndex, Context context) {
+
+        return context.pages.stream()
+                .filter(page -> page.getNumber() == Math.toIntExact(pageIndex))
+                .findFirst()
+                .orElseThrow(() -> new NoSuchElementException(String.format("ClassificationPage with number %d not found", pageIndex)));
+    }
+
+
+    static final class Context {
+
+        private final DocumentTree documentTree;
+        private final List<Page> pages;
+        private final List<DocumentTextData> documentTextDataBlockData;
+        private final List<DocumentPositionData> atomicPositionBlockData;
+
+
+        Context(DocumentData documentData, DocumentTree documentTree) {
+
+            this.documentTree = documentTree;
+            this.pages = new LinkedList<>();
+            this.documentTextDataBlockData = Arrays.stream(documentData.getDocumentTextData())
+                    .toList();
+            this.atomicPositionBlockData = Arrays.stream(documentData.getDocumentPositions())
+                    .toList();
+
+        }
+
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/mapper/MarkdownMapper.java

@@ -26,12 +26,12 @@ import org.commonmark.node.StrongEmphasis;
 import org.commonmark.node.Text;
 import org.commonmark.renderer.markdown.MarkdownRenderer;
 
-import com.iqser.red.service.redaction.v1.server.model.document.AbstractNodeVisitor;
-import com.iqser.red.service.redaction.v1.server.model.document.TextRange;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Headline;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Table;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.TextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.AbstractNodeVisitor;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Headline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.text.FontStyle;
 
 public class MarkdownMapper extends AbstractNodeVisitor {
@@ -39,7 +39,7 @@ public class MarkdownMapper extends AbstractNodeVisitor {
     Document markdownDocument = new Document();
 
 
-    public String toMarkdownContent(com.iqser.red.service.redaction.v1.server.model.document.nodes.Document document) {
+    public String toMarkdownContent(com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document document) {
 
         visit(document);
 
@@ -55,7 +55,7 @@ public class MarkdownMapper extends AbstractNodeVisitor {
 
 
     @Override
-    public void visit(com.iqser.red.service.redaction.v1.server.model.document.nodes.Paragraph paragraph) {
+    public void visit(com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Paragraph paragraph) {
 
         markdownDocument.appendChild(parseParagraph(paragraph));
     }
@@ -108,7 +108,7 @@ public class MarkdownMapper extends AbstractNodeVisitor {
     }
 
 
-    private Node createTableCell(com.iqser.red.service.redaction.v1.server.model.document.nodes.TableCell tc) {
+    private Node createTableCell(com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell tc) {
 
         var cell = new TableCell();
         List<SemanticNode> childNodes = tc.streamChildren()
@@ -122,9 +122,9 @@ public class MarkdownMapper extends AbstractNodeVisitor {
     }
 
 
-    private Paragraph parseParagraph(com.iqser.red.service.redaction.v1.server.model.document.nodes.Paragraph paragraph) {
+    private Paragraph parseParagraph(com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Paragraph paragraph) {
 
-        Paragraph markdownParagraph = new org.commonmark.node.Paragraph();
+        org.commonmark.node.Paragraph markdownParagraph = new org.commonmark.node.Paragraph();
         parseTextBlock(paragraph.getTextBlock(), true).forEach(markdownParagraph::appendChild);
         return markdownParagraph;
     }
@@ -230,12 +230,12 @@ public class MarkdownMapper extends AbstractNodeVisitor {
         int start = textBlock.getTextRange().start();
         int end = textBlock.getTextRange().end();
 
-        for (TextRange bold : textBlock.getBoldTextRanges()) {
+        for (TextRange bold : textBlock.getBoldTextBoundaries()) {
             styleChanges.computeIfAbsent(bold.start() + start, k -> new HashSet<>()).add(FontStyleChange.enter(FontStyle.BOLD));
             styleChanges.computeIfAbsent(bold.end() + start, k -> new HashSet<>()).add(FontStyleChange.leave(FontStyle.BOLD));
         }
 
-        for (TextRange italic : textBlock.getItalicTextRanges()) {
+        for (TextRange italic : textBlock.getItalicTextBoundaries()) {
             styleChanges.computeIfAbsent(italic.start() + start, k -> new HashSet<>()).add(FontStyleChange.enter(FontStyle.ITALIC));
             styleChanges.computeIfAbsent(italic.end() + start, k -> new HashSet<>()).add(FontStyleChange.leave(FontStyle.ITALIC));
         }
@@ -298,6 +298,7 @@ public class MarkdownMapper extends AbstractNodeVisitor {
     }
 
 
+
     record FontStyleChange(boolean enter, FontStyle style) {
 
         public static FontStyleChange enter(FontStyle style) {
@@ -322,5 +323,4 @@ public class MarkdownMapper extends AbstractNodeVisitor {
     record TextRangeWithTextType(TextRange textRange, FontStyle fontStyle) {
 
     }
-
 }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/mapper/OutlineMapper.java

@@ -7,12 +7,12 @@ import java.util.LinkedList;
 import java.util.Map;
 import java.util.Optional;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Headline;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Section;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SuperSection;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Headline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Section;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SuperSection;
 import com.knecon.fforesight.service.layoutparser.processor.utils.CoordinateTransforms;
 import com.knecon.fforesight.service.layoutparser.processor.utils.PageInformation;
 import com.knecon.fforesight.service.viewerdoc.model.Outline;

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/mapper/PropertiesMapper.java

@@ -0,0 +1,139 @@
+package com.knecon.fforesight.service.layoutparser.processor.services.mapper;
+
+import java.awt.geom.Rectangle2D;
+import java.util.Arrays;
+import java.util.HashMap;
+import java.util.Locale;
+import java.util.Map;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.DocumentStructure;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.DuplicatedParagraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Image;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.ImageType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
+
+public class PropertiesMapper {
+
+    public static Map<String, String> buildImageProperties(Image image) {
+
+        Map<String, String> properties = new HashMap<>();
+        properties.put(DocumentStructure.ImageProperties.IMAGE_TYPE, image.getImageType().toString());
+        properties.put(DocumentStructure.ImageProperties.TRANSPARENT, String.valueOf(image.isTransparent()));
+        properties.put(DocumentStructure.ImageProperties.POSITION, toString(image.getPosition()));
+        properties.put(DocumentStructure.ImageProperties.ID, image.getId());
+        properties.put(DocumentStructure.ImageProperties.REPRESENTATION_HASH, image.getRepresentationHash());
+        return properties;
+    }
+
+
+    public static Map<String, String> buildTableCellProperties(TableCell tableCell) {
+
+        Map<String, String> properties = new HashMap<>();
+        properties.put(DocumentStructure.TableCellProperties.ROW, String.valueOf(tableCell.getRow()));
+        properties.put(DocumentStructure.TableCellProperties.COL, String.valueOf(tableCell.getCol()));
+        properties.put(DocumentStructure.TableCellProperties.HEADER, String.valueOf(tableCell.isHeader()));
+
+        if (tableCell.getPages().size() > 1 || tableCell.getBBox().keySet().size() > 1) {
+            throw new IllegalArgumentException("TableCell can only occur on a single page!");
+        }
+        String bBoxString = toString(tableCell.getBBox().get(tableCell.getPages().stream().findFirst().get()));
+        properties.put(DocumentStructure.TableCellProperties.B_BOX, bBoxString);
+
+        return properties;
+    }
+
+
+    public static Map<String, String> buildTableProperties(Table table) {
+
+        Map<String, String> properties = new HashMap<>();
+        properties.put(DocumentStructure.TableProperties.NUMBER_OF_ROWS, String.valueOf(table.getNumberOfRows()));
+        properties.put(DocumentStructure.TableProperties.NUMBER_OF_COLS, String.valueOf(table.getNumberOfCols()));
+        return properties;
+    }
+
+
+    public static void parseImageProperties(Map<String, String> properties, Image.ImageBuilder builder) {
+
+        builder.imageType(parseImageType(properties.get(DocumentStructure.ImageProperties.IMAGE_TYPE)));
+        builder.transparent(Boolean.parseBoolean(properties.get(DocumentStructure.ImageProperties.TRANSPARENT)));
+        builder.position(DocumentStructure.parseRectangle2D(properties.get(DocumentStructure.ImageProperties.POSITION)));
+    }
+
+
+    public static void parseTableCellProperties(Map<String, String> properties, TableCell.TableCellBuilder builder) {
+
+        builder.row(Integer.parseInt(properties.get(DocumentStructure.TableCellProperties.ROW)));
+        builder.col(Integer.parseInt(properties.get(DocumentStructure.TableCellProperties.COL)));
+        builder.header(Boolean.parseBoolean(properties.get(DocumentStructure.TableCellProperties.HEADER)));
+        builder.bBox(DocumentStructure.parseRectangle2D(properties.get(DocumentStructure.TableCellProperties.B_BOX)));
+    }
+
+
+    public static void parseTableProperties(Map<String, String> properties, Table.TableBuilder builder) {
+
+        builder.numberOfRows(Integer.parseInt(properties.get(DocumentStructure.TableProperties.NUMBER_OF_ROWS)));
+        builder.numberOfCols(Integer.parseInt(properties.get(DocumentStructure.TableProperties.NUMBER_OF_COLS)));
+    }
+
+
+    public static Map<String, String> buildDuplicateParagraphProperties(DuplicatedParagraph duplicatedParagraph) {
+
+        Map<String, String> properties = new HashMap<>();
+        properties.put(DocumentStructure.DuplicateParagraphProperties.UNSORTED_TEXTBLOCK_ID, Arrays.toString(toAtomicTextBlockIds(duplicatedParagraph.getUnsortedLeafTextBlock())));
+        return properties;
+    }
+
+
+    public static boolean isDuplicateParagraph(Map<String, String> properties) {
+
+        return properties.containsKey(DocumentStructure.DuplicateParagraphProperties.UNSORTED_TEXTBLOCK_ID);
+    }
+
+
+    public static Long[] getUnsortedTextblockIds(Map<String, String> properties) {
+
+        return toLongArray(properties.get(DocumentStructure.DuplicateParagraphProperties.UNSORTED_TEXTBLOCK_ID));
+    }
+
+
+    public static Long[] toLongArray(String ids) {
+
+        return Arrays.stream(ids.substring(1, ids.length() - 1).trim().split(",")).map(Long::valueOf).toArray(Long[]::new);
+    }
+
+
+    private static ImageType parseImageType(String imageType) {
+
+        return switch (imageType) {
+            case "LOGO" -> ImageType.LOGO;
+            case "FORMULA" -> ImageType.FORMULA;
+            case "SIGNATURE" -> ImageType.SIGNATURE;
+            case "OCR" -> ImageType.OCR;
+            default -> ImageType.OTHER;
+        };
+    }
+
+
+    public static String toString(Rectangle2D rectangle2D) {
+
+        return String.format(Locale.US,
+                "%f%s%f%s%f%s%f",
+                rectangle2D.getX(),
+                DocumentStructure.RECTANGLE_DELIMITER,
+                rectangle2D.getY(),
+                DocumentStructure.RECTANGLE_DELIMITER,
+                rectangle2D.getWidth(),
+                DocumentStructure.RECTANGLE_DELIMITER,
+                rectangle2D.getHeight());
+    }
+
+
+    private static Long[] toAtomicTextBlockIds(TextBlock textBlock) {
+
+        return textBlock.getAtomicTextBlocks().stream().map(AtomicTextBlock::getId).toArray(Long[]::new);
+    }
+
+}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/mapper/TaasDocumentDataMapper.java

@@ -6,21 +6,20 @@ import java.util.Locale;
 import java.util.concurrent.atomic.AtomicInteger;
 import java.util.stream.IntStream;
 
-import com.iqser.red.service.redaction.v1.server.model.document.TextRange;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Table;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.TableCell;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.TextBlock;
-
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.taas.ParagraphData;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.taas.Range;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.taas.ResearchDocumentData;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.taas.RowData;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.taas.StructureObject;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.taas.TableData;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
 
 public class TaasDocumentDataMapper {
 
@@ -45,26 +44,22 @@ public class TaasDocumentDataMapper {
     public static ParagraphData fromTextBlock(String classification, TextBlock textBlock) {
 
         return ParagraphData.builder()
-                .boldTextBoundaries(textBlock.getBoldTextRanges()
-                                            .stream()
-                                            .map(b -> new Range(b.start(), b.end()))
-                                            .toList())
-                .italicTextBoundaries(textBlock.getItalicTextRanges()
-                                              .stream()
-                                              .map(b -> new Range(b.start(), b.end()))
-                                              .toList())
+                .boldTextBoundaries(textBlock.getBoldTextBoundaries().stream().map(b -> new Range(b.start(), b.end())).toList())
+                .italicTextBoundaries(textBlock.getItalicTextBoundaries().stream().map(b -> new Range(b.start(), b.end())).toList())
                 .text(textBlock.getSearchText())
                 .linebreaks(textBlock.getLineBreaks())
                 .classification(classification)
+                .orientation(textBlock.getOrientation())
+                .textDirection(textBlock.getTextDirection())
                 .build();
     }
 
 
     public static TableData fromTable(Table table) {
 
-        List<RowData> rowData = IntStream.range(0, table.getNumberOfRows()).boxed()
-                .map(rowIdx -> table.streamRow(rowIdx)
-                        .toList())
+        List<RowData> rowData = IntStream.range(0, table.getNumberOfRows())
+                .boxed()
+                .map(rowIdx -> table.streamRow(rowIdx).toList())
                 .map(TaasDocumentDataMapper::fromTableCells)
                 .toList();
         return new TableData(rowData, table.getNumberOfCols(), table.getNumberOfRows());
@@ -76,24 +71,14 @@ public class TaasDocumentDataMapper {
         if (tableCells.isEmpty()) {
             throw new IllegalArgumentException("no table cells provided");
         }
-        boolean header = tableCells.stream()
-                .allMatch(TableCell::isHeader);
+        boolean header = tableCells.stream().allMatch(TableCell::isHeader);
         Page firstPage = tableCells.get(0).getFirstPage();
-        Rectangle2D bBox = tableCells.stream()
-                .map(TableCell::getBBox)
-                .reduce((map1, map2) -> {
-                    map1.forEach((page, rectangle) -> map2.merge(page, rectangle, (rect1, rect2) -> rect1.createUnion(rect2).getBounds2D()));
-                    return map2;
-                })
-                .orElseThrow().get(firstPage);
-        List<TextBlock> textBlocks = tableCells.stream()
-                .map(TableCell::getTextBlock)
-                .toList();
-        return new RowData(header,
-                           textBlocks.stream()
-                                   .map(textBlock -> TaasDocumentDataMapper.fromTextBlock("table_cell", textBlock))
-                                   .toList(),
-                           toFloatArray(bBox));
+        Rectangle2D bBox = tableCells.stream().map(TableCell::getBBox).reduce((map1, map2) -> {
+            map1.forEach((page, rectangle) -> map2.merge(page, rectangle, (rect1, rect2) -> rect1.createUnion(rect2).getBounds2D()));
+            return map2;
+        }).orElseThrow().get(firstPage);
+        List<TextBlock> textBlocks = tableCells.stream().map(TableCell::getTextBlock).toList();
+        return new RowData(header, textBlocks.stream().map(textBlock -> TaasDocumentDataMapper.fromTextBlock("table_cell", textBlock)).toList(), toFloatArray(bBox));
     }
 
 
@@ -105,9 +90,7 @@ public class TaasDocumentDataMapper {
 
     private static List<Range> toRange(List<TextRange> textRange) {
 
-        return textRange.stream()
-                .map(TaasDocumentDataMapper::toRange)
-                .toList();
+        return textRange.stream().map(TaasDocumentDataMapper::toRange).toList();
     }
 
 
@@ -119,7 +102,7 @@ public class TaasDocumentDataMapper {
                 .treeId(semanticNode.getTreeId())
                 .structureObjectNumber(structureObjectNumber)
                 .boundingBox(toFloatArray(bBox))
-                .stringOffset(semanticNode.getTextRange().start())
+                .stringOffset(semanticNode.getBoundary().start())
                 .page(page.getNumber())
                 .paragraph(TaasDocumentDataMapper.fromTextBlock(semanticNode.getType().toString().toLowerCase(Locale.ROOT), semanticNode.getTextBlock()))
                 .table(null)
@@ -135,7 +118,7 @@ public class TaasDocumentDataMapper {
                 .treeId(table.getTreeId())
                 .structureObjectNumber(structureObjectNumber)
                 .boundingBox(toFloatArray(bBox))
-                .stringOffset(table.getTextRange().start())
+                .stringOffset(table.getBoundary().start())
                 .page(page.getNumber())
                 .paragraph(null)
                 .table(TaasDocumentDataMapper.fromTable(table))

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/parsing/PDFLinesTextStripper.java

@@ -248,8 +248,8 @@ public class PDFLinesTextStripper extends PDFTextStripper {
 
             if (!words.isEmpty()) {
                 previous = words.get(words.size() - 1)
-                        .getCharacters()
-                        .get(words.get(words.size() - 1).getCharacters().size() - 1).getTextPosition();
+                        .getTextPositions()
+                        .get(words.get(words.size() - 1).getTextPositions().size() - 1);
             }
 
             if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0") || textPositions.get(i).getUnicode().equals("\t"))) {

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/visualization/LayoutGridService.java

@@ -5,13 +5,11 @@ import java.util.List;
 
 import org.springframework.stereotype.Service;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Headline;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Image;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Paragraph;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Table;
-import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
-import com.knecon.fforesight.service.layoutparser.processor.model.DocumentWithVisualization;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Headline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Image;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Paragraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
 import com.knecon.fforesight.service.layoutparser.processor.services.mapper.OutlineMapper;
 import com.knecon.fforesight.service.layoutparser.processor.visualization.LayoutGrid;
 import com.knecon.fforesight.service.viewerdoc.model.Outline;
@@ -33,50 +31,35 @@ public class LayoutGridService {
 
     @SneakyThrows
     @Observed(name = "ViewerDocumentService", contextualName = "create-viewer-document")
-    public void addLayoutGrid(File originFile,
-                              DocumentWithVisualization document,
-                              File destinationFile,
-                              LayoutParsingType layoutParsingType,
-                              String layoutParserVersion,
-                              boolean layerVisibilityDefaultValue) {
+    public void addLayoutGrid(File originFile, Document document, File destinationFile, boolean layerVisibilityDefaultValue) {
 
-        String layoutParsingTypeName = layoutParsingType.name();
-        LayoutGrid layoutGrid = createLayoutGrid(document.document(), layoutParserVersion, layoutParsingTypeName);
-        Outline outline = OutlineMapper.createOutline(document.document());
+        LayoutGrid layoutGrid = createLayoutGrid(document);
+        Outline outline = OutlineMapper.createOutline(document);
         layoutGrid.setVisibleByDefault(layerVisibilityDefaultValue);
 
-        document.layoutDebugLayer().addSentenceVisualization(document.document().getTextBlock());
-        document.layoutDebugLayer().addOutlineHeadlines(document.document());
+        document.getLayoutDebugLayer().addSentenceVisualization(document.getTextBlock());
 
-        if (document.layoutDebugLayer().isActive()) {
-            viewerDocumentService.addLayerGroups(originFile,
-                                                 destinationFile,
-                                                 List.of(layoutGrid, document.layoutDebugLayer()),
-                                                 layoutParserVersion,
-                                                 layoutParsingTypeName,
-                                                 outline);
+        if (document.getLayoutDebugLayer().isActive()) {
+            viewerDocumentService.addLayerGroups(originFile, destinationFile, List.of(layoutGrid, document.getLayoutDebugLayer()), outline);
         } else {
-            viewerDocumentService.addLayerGroups(originFile, destinationFile, List.of(layoutGrid), layoutParserVersion, layoutParsingTypeName, outline);
+            viewerDocumentService.addLayerGroups(originFile, destinationFile, List.of(layoutGrid), outline);
         }
     }
 
 
-    private LayoutGrid createLayoutGrid(Document document, String layoutParserVersion, String layoutParsingType) {
+    private LayoutGrid createLayoutGrid(Document document) {
 
         LayoutGrid layoutGrid = new LayoutGrid();
-        layoutGrid.addVersionAndLayoutParsingType(layoutParserVersion, layoutParsingType, document.getFirstPage());
-
         document.streamAllSubNodes()
                 .peek(layoutGrid::addTreeId)
                 .forEach(semanticNode -> {
                     switch (semanticNode.getType()) {
-                        case SECTION, SUPER_SECTION, TABLE_OF_CONTENTS -> layoutGrid.addSection(semanticNode);
+                        case SECTION, SUPER_SECTION -> layoutGrid.addSection(semanticNode);
                         case HEADLINE -> layoutGrid.addHeadline((Headline) semanticNode);
                         case PARAGRAPH -> layoutGrid.addParagraph((Paragraph) semanticNode);
                         case TABLE -> layoutGrid.addTable((Table) semanticNode);
                         case IMAGE -> layoutGrid.addImage((Image) semanticNode);
                         case HEADER, FOOTER -> layoutGrid.addHeaderOrFooter(semanticNode);
-                        case TABLE_OF_CONTENTS_ITEM -> layoutGrid.addTableOfContentsItem(semanticNode);
                     }
                 });
         return layoutGrid;

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/utils/BBoxMergingUtility.java

@@ -7,7 +7,7 @@ import java.util.Map;
 import java.util.Set;
 import java.util.stream.Collectors;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
 
 import lombok.experimental.UtilityClass;
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/utils/IdBuilder.java

@@ -9,7 +9,7 @@ import java.util.stream.Collectors;
 
 import com.google.common.hash.HashFunction;
 import com.google.common.hash.Hashing;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
 
 import lombok.experimental.UtilityClass;
 
@@ -21,18 +21,14 @@ public final class IdBuilder {
 
     public String buildId(Set<Page> pages, List<Rectangle2D> rectanglesPerLine) {
 
-        return buildId(pages.stream()
-                               .map(Page::getNumber)
-                               .collect(Collectors.toList()), rectanglesPerLine);
+        return buildId(pages.stream().map(Page::getNumber).collect(Collectors.toList()), rectanglesPerLine);
     }
 
 
     public String buildId(List<Integer> pageNumbers, List<Rectangle2D> rectanglesPerLine) {
 
         StringBuilder sb = new StringBuilder();
-        List<Integer> sortedPageNumbers = pageNumbers.stream()
-                .sorted(Comparator.comparingInt(Integer::intValue))
-                .toList();
+        List<Integer> sortedPageNumbers = pageNumbers.stream().sorted(Comparator.comparingInt(Integer::intValue)).toList();
         sortedPageNumbers.forEach(sb::append);
         rectanglesPerLine.forEach(rectangle2D -> sb.append(Math.round(rectangle2D.getX()))
                 .append(Math.round(rectangle2D.getY()))

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/utils/PageInformation.java

@@ -5,7 +5,7 @@ import java.awt.geom.Rectangle2D;
 import org.apache.pdfbox.pdmodel.PDPage;
 import org.apache.pdfbox.pdmodel.common.PDRectangle;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
 
 public record PageInformation(Rectangle2D mediabox, int number, int rotationDegrees) {
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/utils/PdfVisualisationUtility.java

@@ -13,12 +13,13 @@ import org.apache.pdfbox.pdmodel.PDPage;
 import org.apache.pdfbox.pdmodel.PDPageContentStream;
 import org.apache.pdfbox.pdmodel.font.PDType1Font;
 import org.apache.pdfbox.pdmodel.font.Standard14Fonts;
-import com.iqser.red.service.redaction.v1.server.model.document.DocumentTree;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.AtomicTextBlock;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.TextBlock;
+
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.DocumentTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
 
 import lombok.SneakyThrows;
 import lombok.experimental.UtilityClass;
@@ -110,8 +111,8 @@ public class PdfVisualisationUtility {
         return DrawingOptions.builder().stroke(true).strokeColor(switch (entry.getType()) {
             case DOCUMENT -> Color.LIGHT_GRAY;
             case HEADER, FOOTER -> Color.GREEN;
-            case PARAGRAPH, TABLE_OF_CONTENTS_ITEM -> Color.BLUE;
-            case SUPER_SECTION, SECTION, TABLE_OF_CONTENTS -> Color.BLACK;
+            case PARAGRAPH -> Color.BLUE;
+            case SUPER_SECTION, SECTION -> Color.BLACK;
             case HEADLINE -> Color.RED;
             case TABLE -> Color.ORANGE;
             case TABLE_CELL -> Color.GRAY;

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/utils/ProtobufUtil.java

@@ -1,42 +0,0 @@
-package com.knecon.fforesight.service.layoutparser.processor.utils;
-
-import java.io.BufferedOutputStream;
-import java.io.File;
-import java.io.FileOutputStream;
-import java.io.IOException;
-import java.util.zip.GZIPOutputStream;
-
-import com.google.protobuf.Message;
-import com.google.protobuf.MessageOrBuilder;
-import com.google.protobuf.Struct;
-import com.google.protobuf.util.JsonFormat;
-
-import lombok.SneakyThrows;
-import lombok.experimental.UtilityClass;
-
-@UtilityClass
-public class ProtobufUtil {
-
-    public static String toJson(MessageOrBuilder messageOrBuilder) throws IOException {
-        return JsonFormat.printer().print(messageOrBuilder);
-    }
-
-    @SuppressWarnings("unchecked")
-    public static Message fromJson(String json) throws IOException {
-        Message.Builder structBuilder = Struct.newBuilder();
-        JsonFormat.parser().ignoringUnknownFields().merge(json, structBuilder);
-        return structBuilder.build();
-    }
-
-
-    @SneakyThrows
-    public <T extends Message> File serializeToTempFile(T any) {
-        var tempFile = File.createTempFile("storage-protobuf", ".data");
-
-        try (var fos = new GZIPOutputStream(new BufferedOutputStream(new FileOutputStream(tempFile)))) {
-            any.writeTo(fos);
-            return tempFile;
-        }
-    }
-
-}

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/utils/RectangleTransformations.java

@@ -20,7 +20,7 @@ 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.model.document.textblock.AtomicTextBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.AtomicTextBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.CleanRulings;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.Ruling;
 

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/utils/TextPositionOperations.java

@@ -11,7 +11,6 @@ import java.util.Set;
 import java.util.stream.Collectors;
 
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.AngleFilter;
-import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.Line;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.TextBoundingBox;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.UnionFind;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.utils.DoubleUtils;
@@ -55,20 +54,11 @@ public class TextPositionOperations {
 
     private List<Word> sortUsingLineDetection(Set<Word> sequences) {
 
-        return sortWords(groupByLine(sequences));
+        return sortLines(groupByLine(sequences));
     }
 
 
-    public List<Word> sortWords(Collection<Set<Word>> lines) {
-
-        return sortLines(lines).stream()
-                .map(Line::getWords)
-                .flatMap(Collection::stream)
-                .collect(Collectors.toList());
-    }
-
-
-    public List<Line> sortLines(Collection<Set<Word>> lines) {
+    public List<Word> sortLines(Collection<Set<Word>> lines) {
 
         List<List<Word>> lineBlocks = new ArrayList<>();
         for (Set<Word> line : lines) {
@@ -80,9 +70,9 @@ public class TextPositionOperations {
         // need to use old sorting, since COMPARATOR_DIR_ADJ is not transitive
         QuickSort.sort(lineBlocks, Comparator.comparing(line -> line.get(0), COMPARATOR_DIR_ADJ));
 
-        List<Line> list = new ArrayList<>();
-        for (List<Word> lineBlock : lineBlocks) {
-            list.add(new Line(lineBlock));
+        List<Word> list = new ArrayList<>();
+        for (List<Word> words : lineBlocks) {
+            list.addAll(words);
         }
         return list;
     }
@@ -105,12 +95,6 @@ public class TextPositionOperations {
                                  .map(Word::getBBoxDirAdj)
                                  .mapToDouble(RectangularShape::getWidth).average().orElse(75) * MAX_WORD_DISTANCE_FACTOR;
 
-        return groupByLine(sequences, maxLineDistance, maxXGap);
-    }
-
-
-    public Collection<Set<Word>> groupByLine(Set<Word> sequences, double maxLineDistance, double maxXGap) {
-
         UnionFind<Word> unionFind = new UnionFind<>(sequences);
 
         for (Word sequence : sequences) {

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/visualization/LayoutDebugLayer.java

@@ -17,16 +17,12 @@ import java.util.concurrent.atomic.AtomicInteger;
 import org.apache.pdfbox.pdmodel.documentinterchange.markedcontent.PDMarkedContent;
 
 import com.iqser.red.service.persistence.service.v1.api.shared.model.redactionlog.Rectangle;
-import com.iqser.red.service.redaction.v1.server.data.LayoutEngineProto;
-import com.iqser.red.service.redaction.v1.server.model.document.TextRange;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.LayoutEngine;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.TextBlock;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.BoundingBox;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.Line;
 import com.knecon.fforesight.service.layoutparser.processor.docstrum.model.Zone;
 import com.knecon.fforesight.service.layoutparser.processor.model.AbstractPageBlock;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.TextRange;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
 import com.knecon.fforesight.service.layoutparser.processor.model.outline.OutlineObject;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.CleanRulings;
 import com.knecon.fforesight.service.layoutparser.processor.model.table.Ruling;
@@ -279,9 +275,7 @@ public class LayoutDebugLayer extends LayoutDebugLayerConfig {
         AtomicInteger index = new AtomicInteger(0);
         zones.forEach(zone -> zone.getLines()
                 .stream()
-                .map(Line::getWords)
-                .flatMap(Collection::stream)
-                .map(Word::getCharacters)
+                .map(Line::getCharacters)
                 .flatMap(Collection::stream)
                 .forEach(character -> {
                     Color color = getRotatingColor(index);
@@ -383,19 +377,4 @@ public class LayoutDebugLayer extends LayoutDebugLayerConfig {
         }
     }
 
-
-    public void addOutlineHeadlines(Document document) {
-
-        if (!active) {
-            return;
-        }
-
-        document.streamAllSubNodes()
-                .filter(node -> node.getType().equals(NodeType.HEADLINE))
-                .filter(node -> node.getEngines().contains(LayoutEngine.OUTLINE))
-                .forEach(headline -> headline.getBBox()
-                        .forEach((page, bbox) -> getOrCreateVisualizationsOnPage(page.getNumber(), this.outlineHeadlines).getColoredRectangles()
-                                .add(new ColoredRectangle(bbox, HEADLINE_COLOR, LINE_WIDTH))));
-    }
-
 }

layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/visualization/LayoutGrid.java

@@ -1,7 +1,6 @@
 package com.knecon.fforesight.service.layoutparser.processor.visualization;
 
 import java.awt.Color;
-import java.awt.geom.AffineTransform;
 import java.awt.geom.Line2D;
 import java.awt.geom.Point2D;
 import java.awt.geom.Rectangle2D;
@@ -16,19 +15,17 @@ import java.util.Optional;
 import java.util.stream.Collectors;
 import java.util.stream.Stream;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.DuplicatedParagraph;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Headline;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Image;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.NodeType;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Page;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Paragraph;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SuperSection;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Table;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.TableCell;
-import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingType;
-import com.knecon.fforesight.service.layoutparser.processor.utils.CoordinateTransforms;
-import com.knecon.fforesight.service.layoutparser.processor.utils.PageInformation;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.LayoutEngine;
+import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.NodeType;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.DuplicatedParagraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Headline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Image;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Page;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Paragraph;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SuperSection;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Table;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.TableCell;
 import com.knecon.fforesight.service.layoutparser.processor.utils.RectangleTransformations;
 import com.knecon.fforesight.service.viewerdoc.layers.LayoutGridLayerConfig;
 import com.knecon.fforesight.service.viewerdoc.model.ColoredLine;
@@ -75,12 +72,10 @@ public class LayoutGrid extends LayoutGridLayerConfig {
 
     public void addHeadline(Headline headline) {
 
-        if (headline.getParent().getType().equals(NodeType.TABLE_OF_CONTENTS)) {
-            addAsRectangle(headline, toc, HEADLINE_COLOR);
-        } else {
-            addAsRectangle(headline, headlines, HEADLINE_COLOR);
+        addAsRectangle(headline, headlines, HEADLINE_COLOR);
+        if (headline.getEngines().contains(LayoutEngine.OUTLINE)) {
+            addAsRectangle(headline, outlineHeadlines, HEADLINE_COLOR);
         }
-
     }
 
 
@@ -129,7 +124,7 @@ public class LayoutGrid extends LayoutGridLayerConfig {
     public void addSection(SemanticNode section) {
 
         Map<Page, Rectangle2D> bBoxMap = section.getBBox();
-        Color color = section.getType().equals(NodeType.TABLE_OF_CONTENTS) ? TOC_COLOR : SECTION_COLOR;
+
         List<SemanticNode> subSections = section.streamAllSubNodesOfType(NodeType.SECTION)
                 .toList();
         Integer maxChildDepth = subSections.stream()
@@ -142,7 +137,7 @@ public class LayoutGrid extends LayoutGridLayerConfig {
         String treeIdString = buildTreeIdString(section);
 
         if (bBoxMap.values().size() == 1) {
-            handleSinglePage(section, firstPage, bBoxMap.get(firstPage), treeIdString, maxChildDepth, ownDepth, color);
+            handleSinglePage(section, firstPage, bBoxMap.get(firstPage), treeIdString, maxChildDepth, ownDepth);
             return;
         }
         List<Page> pagesInOrder = bBoxMap.keySet()
@@ -150,12 +145,12 @@ public class LayoutGrid extends LayoutGridLayerConfig {
                 .sorted(Comparator.comparingInt(Page::getNumber))
                 .collect(Collectors.toList());
         pagesInOrder.remove(0);
-        handleFirstPageOfSection(section, firstPage, bBoxMap.get(firstPage), treeIdString, maxChildDepth, ownDepth, color);
+        handleFirstPageOfSection(section, firstPage, bBoxMap.get(firstPage), treeIdString, maxChildDepth, ownDepth);
         for (Page middlePage : pagesInOrder.subList(0, pagesInOrder.size() - 1)) {
-            handleForMiddlePageOfSection(section, middlePage, bBoxMap.get(middlePage), treeIdString, maxChildDepth, ownDepth, color);
+            handleForMiddlePageOfSection(section, middlePage, bBoxMap.get(middlePage), treeIdString, maxChildDepth, ownDepth);
         }
         var lastPage = pagesInOrder.remove(pagesInOrder.size() - 1);
-        handleLastPageOfSection(section, lastPage, bBoxMap.get(lastPage), treeIdString, maxChildDepth, ownDepth, color);
+        handleLastPageOfSection(section, lastPage, bBoxMap.get(lastPage), treeIdString, maxChildDepth, ownDepth);
     }
 
 
@@ -237,45 +232,33 @@ public class LayoutGrid extends LayoutGridLayerConfig {
     }
 
 
-    private void handleSinglePage(SemanticNode semanticNode, Page page, Rectangle2D rectangle2D, String treeIdString, Integer maxChildDepth, Integer ownDepth, Color color) {
+    private void handleSinglePage(SemanticNode semanticNode, Page page, Rectangle2D rectangle2D, String treeIdString, Integer maxChildDepth, Integer ownDepth) {
 
         RectangleAndLinesResult result = createLinesAndPlaceText(semanticNode, page, rectangle2D, treeIdString, maxChildDepth, ownDepth);
         // add string to top line
         var firstLine = result.pageLines().remove(0);
-        result.coloredLines().add(new ColoredLine(firstLine, color, LINE_WIDTH));
+        result.coloredLines().add(new ColoredLine(firstLine, SECTION_COLOR, LINE_WIDTH));
         for (Line2D line : result.pageLines()) {
-            result.coloredLines().add(new ColoredLine(line, color, LINE_WIDTH));
+            result.coloredLines().add(new ColoredLine(line, SECTION_COLOR, LINE_WIDTH));
         }
     }
 
 
-    private void handleFirstPageOfSection(SemanticNode semanticNode,
-                                          Page firstPage,
-                                          Rectangle2D rectangle2D,
-                                          String treeIdString,
-                                          Integer maxChildDepth,
-                                          Integer ownDepth,
-                                          Color color) {
+    private void handleFirstPageOfSection(SemanticNode semanticNode, Page firstPage, Rectangle2D rectangle2D, String treeIdString, Integer maxChildDepth, Integer ownDepth) {
 
         RectangleAndLinesResult result = createLinesAndPlaceText(semanticNode, firstPage, rectangle2D, treeIdString, maxChildDepth, ownDepth);
         // remove bottom line
         result.pageLines().remove(2);
         // add string to top line
         var firstLine = result.pageLines().remove(0);
-        result.coloredLines().add(new ColoredLine(firstLine, color, LINE_WIDTH));
+        result.coloredLines().add(new ColoredLine(firstLine, SECTION_COLOR, LINE_WIDTH));
         for (Line2D line : result.pageLines()) {
-            result.coloredLines().add(new ColoredLine(line, color, LINE_WIDTH));
+            result.coloredLines().add(new ColoredLine(line, SECTION_COLOR, LINE_WIDTH));
         }
     }
 
 
-    private void handleForMiddlePageOfSection(SemanticNode semanticNode,
-                                              Page middlePage,
-                                              Rectangle2D rectangle2D,
-                                              String treeIdString,
-                                              Integer maxChildDepth,
-                                              Integer ownDepth,
-                                              Color color) {
+    private void handleForMiddlePageOfSection(SemanticNode semanticNode, Page middlePage, Rectangle2D rectangle2D, String treeIdString, Integer maxChildDepth, Integer ownDepth) {
 
         RectangleAndLinesResult result = createLinesAndPlaceText(semanticNode, middlePage, rectangle2D, treeIdString, maxChildDepth, ownDepth);
         // remove top line
@@ -284,29 +267,23 @@ public class LayoutGrid extends LayoutGridLayerConfig {
         result.pageLines().remove(1);
         // add string to left line
         var leftLine = result.pageLines().remove(1);
-        result.coloredLines().add(new ColoredLine(leftLine, color, LINE_WIDTH));
+        result.coloredLines().add(new ColoredLine(leftLine, SECTION_COLOR, LINE_WIDTH));
         for (Line2D line : result.pageLines()) {
-            result.coloredLines().add(new ColoredLine(line, color, LINE_WIDTH));
+            result.coloredLines().add(new ColoredLine(line, SECTION_COLOR, LINE_WIDTH));
         }
     }
 
 
-    private void handleLastPageOfSection(SemanticNode semanticNode,
-                                         Page lastPage,
-                                         Rectangle2D rectangle2D,
-                                         String treeIdString,
-                                         Integer maxChildDepth,
-                                         Integer ownDepth,
-                                         Color color) {
+    private void handleLastPageOfSection(SemanticNode semanticNode, Page lastPage, Rectangle2D rectangle2D, String treeIdString, Integer maxChildDepth, Integer ownDepth) {
 
         RectangleAndLinesResult result = createLinesAndPlaceText(semanticNode, lastPage, rectangle2D, treeIdString, maxChildDepth, ownDepth);
         // remove top line
         result.pageLines().remove(0);
         // add string to left line
         var leftLine = result.pageLines().remove(2);
-        result.coloredLines().add(new ColoredLine(leftLine, color, LINE_WIDTH));
+        result.coloredLines().add(new ColoredLine(leftLine, SECTION_COLOR, LINE_WIDTH));
         for (Line2D line : result.pageLines()) {
-            result.coloredLines().add(new ColoredLine(line, color, LINE_WIDTH));
+            result.coloredLines().add(new ColoredLine(line, SECTION_COLOR, LINE_WIDTH));
         }
     }
 
@@ -318,15 +295,14 @@ public class LayoutGrid extends LayoutGridLayerConfig {
                                                             Integer maxChildDepth,
                                                             Integer ownDepth) {
 
-        Visualizations visualizations = semanticNode.getType().equals(NodeType.TABLE_OF_CONTENTS) ? toc : sections;
-        List<ColoredLine> coloredLines = getOrCreateVisualizationsOnPage(page.getNumber(), visualizations).getColoredLines();
+        List<ColoredLine> coloredLines = getOrCreateVisualizationsOnPage(page.getNumber(), sections).getColoredLines();
         int lineWidthModifier = maxChildDepth - ownDepth;
         Rectangle2D r = RectangleTransformations.pad(semanticNode.getBBox()
                                                              .get(page), LINE_WIDTH * (1 + lineWidthModifier), LINE_WIDTH * (1 + lineWidthModifier));
 
         SemanticNode highestParent = semanticNode.getHighestParent();
         Rectangle2D highestParentRect = rectangleMap.get(new RectangleIdentifier(highestParent.getTreeId(), page.getNumber()));
-        addPlacedText(page, rectangle2D, highestParentRect, treeIdString, maxChildDepth, visualizations, SECTION_COLOR);
+        addPlacedText(page, rectangle2D, highestParentRect, treeIdString, maxChildDepth, sections, SECTION_COLOR);
         var lastPageLines = createLinesFromRectangle(r, page.getRotation());
 
         if (semanticNode instanceof SuperSection) {
@@ -345,7 +321,7 @@ public class LayoutGrid extends LayoutGridLayerConfig {
         for (Page page : table.getPages()) {
 
             Optional<Integer> optionalFirstRowOnPage = table.streamCol(0)
-                    .filter(tableCell -> tableCell.onPage(page.getNumber()))
+                    .filter(tableCell -> tableCell.isOnPage(page.getNumber()))
                     .map(TableCell::getRow)
                     .findFirst();
             if (optionalFirstRowOnPage.isEmpty()) {
@@ -389,7 +365,7 @@ public class LayoutGrid extends LayoutGridLayerConfig {
 
     private static Stream<Rectangle2D> streamBBoxOfCellsOnPage(Stream<TableCell> table, Page page) {
 
-        return table.filter(tableCell -> tableCell.onPage(page.getNumber()))
+        return table.filter(tableCell -> tableCell.isOnPage(page.getNumber()))
                 .map(TableCell::getBBox)
                 .map(bBoxMap -> bBoxMap.get(page));
     }
@@ -408,27 +384,6 @@ public class LayoutGrid extends LayoutGridLayerConfig {
     }
 
 
-    public void addTableOfContentsItem(SemanticNode semanticNode) {
-
-        addAsRectangle(semanticNode, toc, PARAGRAPH_COLOR);
-    }
-
-
-    public void addVersionAndLayoutParsingType(String version, String layoutParsingType, Page page) {
-
-        PageInformation pageInformation = PageInformation.fromPage(page);
-        double startHeight = pageInformation.heightRot() - 5;
-        Point2D point1 = new Point2D.Double(0, startHeight);
-        Point2D point2 = new Point2D.Double(0, startHeight - FONT_SIZE * 1.5);
-        AffineTransform affineTransform = CoordinateTransforms.calculatePageCoordsToInitialUserSpaceCoords(pageInformation);
-        affineTransform.transform(point1, point1);
-        affineTransform.transform(point2, point2);
-        getOrCreateVisualizationsOnPage(page.getNumber(), this.versionAndType).getPlacedTexts()
-                .addAll(List.of(PlacedText.textFacingUp(String.valueOf(version), point1, FONT_SIZE, Color.BLACK, FONT),
-                                PlacedText.textFacingUp(String.valueOf(layoutParsingType), point2, FONT_SIZE, Color.BLACK, FONT)));
-    }
-
-
     private record RectangleAndLinesResult(List<ColoredLine> coloredLines, Rectangle2D rectangle, List<Line2D> pageLines) {
 
     }

layoutparser-service/layoutparser-service-server/build.gradle.kts

@@ -29,8 +29,8 @@ dependencies {
     implementation(project(":layoutparser-service-processor"))
     implementation(project(":layoutparser-service-internal-api"))
 
-    implementation("com.iqser.red.commons:storage-commons:2.50.0")
-    implementation("com.knecon.fforesight:tenant-commons:0.31.0")
+    implementation("com.iqser.red.commons:storage-commons:2.45.0")
+    implementation("com.knecon.fforesight:tenant-commons:0.21.0")
     implementation("com.knecon.fforesight:tracing-commons:0.5.0")
     implementation("com.knecon.fforesight:lifecycle-commons:0.6.0")
     implementation("org.springframework.boot:spring-boot-starter-actuator:${springBootStarterVersion}")
@@ -38,20 +38,15 @@ dependencies {
     implementation("com.amazonaws:aws-java-sdk-s3:1.12.536")
     implementation("org.springframework.cloud:spring-cloud-starter-openfeign:4.0.4")
     implementation("net.logstash.logback:logstash-logback-encoder:7.4")
-    implementation("ch.qos.logback:logback-classic")
-    api("com.iqser.red.commons:metric-commons:2.3.0")
-
     implementation("com.pdftron:PDFNet:10.11.0")
 
     // for integration testing only
     testImplementation(project(":viewer-doc-processor"))
     testImplementation(project(":layoutparser-service-internal-api"))
 
-    testImplementation("com.knecon.fforesight:document:${rootProject.extra.get("documentVersion")}")
-
     testImplementation("org.springframework.boot:spring-boot-starter-amqp:${springBootStarterVersion}")
     testImplementation("org.springframework.boot:spring-boot-starter-test:${springBootStarterVersion}")
-    testImplementation("com.iqser.red.service:persistence-service-shared-api-v1:2.564.0-RED9010.0")
+    testImplementation("com.iqser.red.service:persistence-service-shared-api-v1:2.144.0")
     testImplementation("com.iqser.red.commons:jackson-commons:1.3.0")
     testImplementation("com.fasterxml.jackson.module:jackson-module-afterburner:${jacksonVersion}")
     testImplementation("com.fasterxml.jackson.datatype:jackson-datatype-jsr310:${jacksonVersion}")

layoutparser-service/layoutparser-service-server/src/main/java/com/knecon/fforesight/service/layoutparser/server/configuration/MessagingConfiguration.java

@@ -1,29 +0,0 @@
-package com.knecon.fforesight.service.layoutparser.server.configuration;
-
-import static com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingQueueNames.LAYOUT_PARSING_DLQ;
-
-import org.springframework.amqp.core.DirectExchange;
-import org.springframework.amqp.core.Queue;
-import org.springframework.amqp.core.QueueBuilder;
-import org.springframework.context.annotation.Bean;
-import org.springframework.context.annotation.Configuration;
-
-import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingQueueNames;
-
-@Configuration
-public class MessagingConfiguration {
-
-    @Bean
-    public DirectExchange layoutParsingRequestExchange() {
-
-        return new DirectExchange(LayoutParsingQueueNames.LAYOUT_PARSING_REQUEST_EXCHANGE);
-    }
-
-
-    @Bean
-    public Queue layoutParsingDLQ() {
-
-        return QueueBuilder.durable(LAYOUT_PARSING_DLQ).build();
-    }
-
-}

layoutparser-service/layoutparser-service-server/src/main/java/com/knecon/fforesight/service/layoutparser/server/configuration/TenantQueueProviderConfig.java

@@ -1,27 +0,0 @@
-package com.knecon.fforesight.service.layoutparser.server.configuration;
-
-import java.util.Set;
-
-import org.springframework.context.annotation.Bean;
-import org.springframework.context.annotation.Configuration;
-
-import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsingQueueNames;
-import com.knecon.fforesight.service.layoutparser.server.queue.MessageHandler;
-import com.knecon.fforesight.tenantcommons.model.TenantQueueConfiguration;
-import com.knecon.fforesight.tenantcommons.model.TenantQueueProvider;
-
-@Configuration
-public class TenantQueueProviderConfig {
-
-    @Bean
-    public TenantQueueProvider getTenantQueueConfigs() {
-
-        return new TenantQueueProvider(Set.of(TenantQueueConfiguration.builder()
-                                                      .listenerId(MessageHandler.LAYOUT_PARSING_REQUEST_LISTENER_ID)
-                                                      .exchangeName(LayoutParsingQueueNames.LAYOUT_PARSING_REQUEST_EXCHANGE)
-                                                      .queuePrefix(LayoutParsingQueueNames.LAYOUT_PARSING_REQUEST_QUEUE_PREFIX)
-                                                      .dlqName(LayoutParsingQueueNames.LAYOUT_PARSING_DLQ)
-                                                      .build()));
-    }
-
-}

layoutparser-service/layoutparser-service-server/src/main/java/com/knecon/fforesight/service/layoutparser/server/queue/MessageHandler.java

@@ -18,7 +18,6 @@ import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsi
 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 lombok.RequiredArgsConstructor;
 import lombok.SneakyThrows;
@@ -29,17 +28,15 @@ import lombok.extern.slf4j.Slf4j;
 @RequiredArgsConstructor
 public class MessageHandler {
 
-    public static final String LAYOUT_PARSING_REQUEST_LISTENER_ID = "layout-parsing-request-listener";
-
     private final LayoutParsingPipeline layoutParsingPipeline;
     private final ObjectMapper objectMapper;
     private final RabbitTemplate rabbitTemplate;
     private final static String X_PIPELINE_PREFIX = "X-PIPE-";
 
 
-    @SneakyThrows
     @RabbitHandler
-    @RabbitListener(id = LAYOUT_PARSING_REQUEST_LISTENER_ID)
+    @RabbitListener(queues = LayoutParsingQueueNames.LAYOUT_PARSING_REQUEST_QUEUE)
+    @SneakyThrows
     public void receiveLayoutParsingRequest(Message message) {
 
         LayoutParsingRequest layoutParsingRequest = objectMapper.readValue(message.getBody(), LayoutParsingRequest.class);
@@ -52,7 +49,6 @@ public class MessageHandler {
             throw new AmqpRejectAndDontRequeueException(String.format("Error during last layout parsing of request with identifier: %s, do not retry.",
                     layoutParsingRequest.identifier()));
         }
-        log.info("--------------------------------- Starting layoutparsing ---------------------------------");
         LayoutParsingFinishedEvent layoutParsingFinishedEvent = layoutParsingPipeline.parseLayoutAndSaveFilesToStorage(layoutParsingRequest);
         sendLayoutParsingFinishedEvent(layoutParsingFinishedEvent, message);
     }
@@ -61,7 +57,7 @@ public class MessageHandler {
     public void sendLayoutParsingFinishedEvent(LayoutParsingFinishedEvent layoutParsingFinishedEvent, Message message) {
 
         Arrays.stream(layoutParsingFinishedEvent.message().split("\n")).forEach(log::info);
-        rabbitTemplate.convertAndSend(LayoutParsingQueueNames.LAYOUT_PARSING_RESPONSE_EXCHANGE, TenantContext.getTenantId(), layoutParsingFinishedEvent, m -> {
+        rabbitTemplate.convertAndSend(LayoutParsingQueueNames.LAYOUT_PARSING_FINISHED_EVENT_QUEUE, layoutParsingFinishedEvent, m -> {
             var forwardHeaders = message.getMessageProperties()
                     .getHeaders()
                     .entrySet()
@@ -72,7 +68,6 @@ public class MessageHandler {
             m.getMessageProperties().getHeaders().putAll(forwardHeaders);
             return m;
         });
-        log.info("--------------------------------- Finished layoutparsing ---------------------------------");
     }
 
 }

layoutparser-service/layoutparser-service-server/src/test/java/com/knecon/fforesight/service/layoutparser/server/ApplicationTests.java

@@ -2,6 +2,8 @@ package com.knecon.fforesight.service.layoutparser.server;
 
 import org.junit.jupiter.api.Test;
 
+import com.knecon.fforesight.service.layoutparser.server.utils.AbstractTest;
+
 class ApplicationTests extends AbstractTest {
 
     @Test

layoutparser-service/layoutparser-service-server/src/test/java/com/knecon/fforesight/service/layoutparser/server/BdrJsonBuildTest.java

@@ -21,18 +21,18 @@ import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.core.io.ClassPathResource;
 
 import com.fasterxml.jackson.databind.ObjectMapper;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.textblock.TextBlock;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.taas.ResearchDocumentData;
 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.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.textblock.TextBlock;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.image.ImageServiceResponse;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.TableServiceResponse;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.VisualLayoutParsingResponse;
 import com.knecon.fforesight.service.layoutparser.processor.services.factory.DocumentGraphFactory;
 import com.knecon.fforesight.service.layoutparser.processor.services.mapper.TaasDocumentDataMapper;
 import com.knecon.fforesight.service.layoutparser.processor.utils.DrawingOptions;
+import com.knecon.fforesight.service.layoutparser.server.utils.AbstractTest;
 import com.knecon.fforesight.service.layoutparser.server.utils.visualizations.PdfDraw;
 
 import lombok.SneakyThrows;
@@ -55,7 +55,7 @@ public class BdrJsonBuildTest extends AbstractTest {
                         new ImageServiceResponse(),
                         new TableServiceResponse(),
                         new VisualLayoutParsingResponse(),
-                        Map.of("file",file.toString()))).document();
+                        Map.of("file",file.toString())));
     }
 
 

layoutparser-service/layoutparser-service-server/src/test/java/com/knecon/fforesight/service/layoutparser/server/HeadlinesGoldStandardIntegrationTest.java

@@ -8,24 +8,30 @@ import java.util.Set;
 import java.util.stream.Collectors;
 
 import org.assertj.core.api.Assertions;
-import org.junit.jupiter.api.Disabled;
 import org.junit.jupiter.api.Test;
 import org.junit.jupiter.api.extension.ExtendWith;
-import org.springframework.amqp.rabbit.core.RabbitAdmin;
 import org.springframework.amqp.rabbit.core.RabbitTemplate;
 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.ComponentScan;
+import org.springframework.context.annotation.Configuration;
+import org.springframework.context.annotation.FilterType;
 import org.springframework.context.annotation.Import;
 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.redaction.v1.server.model.document.nodes.Document;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
+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.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.LayoutParsingPipeline;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.image.ImageServiceResponse;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.TableServiceResponse;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.VisualLayoutParsingResponse;
@@ -38,10 +44,9 @@ import lombok.EqualsAndHashCode;
 import lombok.SneakyThrows;
 import lombok.ToString;
 
-@Disabled
 @ExtendWith(SpringExtension.class)
 @SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT)
-@Import(AbstractTest.TestConfiguration.class)
+@Import(HeadlinesGoldStandardIntegrationTest.RedactionIntegrationTestConfiguration.class)
 public class HeadlinesGoldStandardIntegrationTest {
 
     @MockBean
@@ -50,9 +55,6 @@ public class HeadlinesGoldStandardIntegrationTest {
     @MockBean
     private RabbitTemplate rabbitTemplate;
 
-    @MockBean
-    private RabbitAdmin rabbitAdmin;
-
     @Autowired
     private ObjectMapper objectMapper;
 
@@ -93,13 +95,10 @@ public class HeadlinesGoldStandardIntegrationTest {
         ClassPathResource pdfFileResource = new ClassPathResource(filePath);
 
         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())));
+        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())));
 
         Document documentGraph = DocumentGraphFactory.buildDocumentGraph(LayoutParsingType.DOCUMINE,
                                                                          layoutParsingPipeline.parseLayout(LayoutParsingType.DOCUMINE,
@@ -107,7 +106,8 @@ public class HeadlinesGoldStandardIntegrationTest {
                                                                                                            new ImageServiceResponse(),
                                                                                                            new TableServiceResponse(),
                                                                                                            new VisualLayoutParsingResponse(),
-                                                                                                           Map.of("file", filePath))).document();
+                                                                                                           Map.of("file", filePath)));
+
         var foundHeadlines = documentGraph.streamAllSubNodes()
                 .map(SemanticNode::getHeadline)
                 .distinct()
@@ -138,6 +138,13 @@ public class HeadlinesGoldStandardIntegrationTest {
     }
 
 
+    @Configuration
+    @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class/*, StorageAutoConfiguration.class*/})
+    @ComponentScan(excludeFilters = {@ComponentScan.Filter(type = FilterType.ASSIGNABLE_TYPE, value = StorageAutoConfiguration.class)})
+    public static class RedactionIntegrationTestConfiguration {
+
+    }
+
     @Data
     @EqualsAndHashCode
     @AllArgsConstructor

layoutparser-service/layoutparser-service-server/src/test/java/com/knecon/fforesight/service/layoutparser/server/LayoutparserEnd2EndTest.java

@@ -19,6 +19,7 @@ import com.knecon.fforesight.service.layoutparser.internal.api.queue.LayoutParsi
 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.server.utils.AbstractTest;
 
 import lombok.SneakyThrows;
 import lombok.extern.slf4j.Slf4j;

layoutparser-service/layoutparser-service-server/src/test/java/com/knecon/fforesight/service/layoutparser/server/OutlineDetectionTest.java

@@ -10,6 +10,7 @@ import java.util.Collection;
 import java.util.List;
 import java.util.Locale;
 import java.util.Map;
+import java.util.function.Predicate;
 
 import org.apache.commons.lang3.StringUtils;
 import org.junit.jupiter.api.BeforeEach;
@@ -17,21 +18,22 @@ import org.junit.jupiter.api.Test;
 import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.core.io.ClassPathResource;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Section;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SemanticNode;
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.SuperSection;
 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.model.ClassificationDocument;
-import com.knecon.fforesight.service.layoutparser.processor.model.DocumentWithVisualization;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Document;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.Section;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SemanticNode;
+import com.knecon.fforesight.service.layoutparser.processor.model.graph.nodes.SuperSection;
 import com.knecon.fforesight.service.layoutparser.processor.model.outline.OutlineObject;
 import com.knecon.fforesight.service.layoutparser.processor.model.outline.OutlineObjectTree;
-import com.knecon.fforesight.service.layoutparser.processor.model.outline.SectionTree;
+import com.knecon.fforesight.service.layoutparser.processor.model.outline.TableOfContents;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.TableServiceResponse;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.VisualLayoutParsingResponse;
 import com.knecon.fforesight.service.layoutparser.processor.services.factory.DocumentGraphFactory;
 import com.knecon.fforesight.service.layoutparser.processor.services.visualization.LayoutGridService;
+import com.knecon.fforesight.service.layoutparser.server.utils.AbstractTest;
 import com.knecon.fforesight.service.viewerdoc.service.PDFTronViewerDocumentService;
 
 import lombok.SneakyThrows;
@@ -78,8 +80,8 @@ public class OutlineDetectionTest extends AbstractTest {
 
         long start = System.currentTimeMillis();
         ClassificationDocument classificationDocument = parseLayout(fileName, LayoutParsingType.DOCUMINE_OLD);
-        var document = buildGraph(fileName, classificationDocument);
-        layoutGridService.addLayoutGrid(documentFile, document, new File(tmpFileName), LayoutParsingType.DOCUMINE_OLD, "TEST_VERSION", true);
+        Document document = buildGraph(fileName, classificationDocument);
+        layoutGridService.addLayoutGrid(documentFile, document, new File(tmpFileName), true);
         OutlineObjectTree outlineObjectTree = classificationDocument.getOutlineObjectTree();
         assertEquals(outlineObjectTree.getRootNodes().size(), 8);
         assertEquals(outlineObjectTree.getOutlineObjectsPerPage().get(2).size(), 1);
@@ -98,10 +100,10 @@ public class OutlineDetectionTest extends AbstractTest {
                            .flatMap(Collection::stream)
                            .allMatch(OutlineObject::isFound));
 
-        SectionTree sectionTree = classificationDocument.getSectionTree();
+        TableOfContents tableOfContents = classificationDocument.getTableOfContents();
 
-        assertEquals(sectionTree.getMainSections().size(), 9);
-        assertEquals(sectionTree.getMainSections().subList(1, 9)
+        assertEquals(tableOfContents.getMainSections().size(), 9);
+        assertEquals(tableOfContents.getMainSections().subList(1, 9)
                              .stream()
                              .map(tableOfContentItem -> sanitizeString(tableOfContentItem.getHeadline().toString()))
                              .toList(),
@@ -119,19 +121,19 @@ public class OutlineDetectionTest extends AbstractTest {
 //        assertEquals(tableOfContents.getMainSections().get(6).getImages().size(), 1);
 //        assertEquals(tableOfContents.getMainSections().get(8).getChildren().get(2).getChildren().get(0).getChildren().get(2).getImages().size(), 1);
 
-        assertTrue(sectionTree.getAllTableOfContentItems()
+        assertTrue(tableOfContents.getAllTableOfContentItems()
                            .stream()
                            .allMatch(tableOfContentItem -> tableOfContentItem.getSection() != null));
-        assertTrue(sectionTree.getAllTableOfContentItems()
+        assertTrue(tableOfContents.getAllTableOfContentItems()
                            .stream()
                            .filter(tableOfContentItem -> tableOfContentItem.getChildren().isEmpty())
                            .allMatch(tableOfContentItem -> tableOfContentItem.getSection() instanceof Section));
-        assertTrue(sectionTree.getAllTableOfContentItems()
+        assertTrue(tableOfContents.getAllTableOfContentItems()
                            .stream()
                            .filter(tableOfContentItem -> !tableOfContentItem.getChildren().isEmpty())
                            .allMatch(tableOfContentItem -> tableOfContentItem.getSection() instanceof SuperSection));
 
-        List<SemanticNode> childrenOfTypeSectionOrSuperSection = document.document().getChildrenOfTypeSectionOrSuperSection();
+        List<SemanticNode> childrenOfTypeSectionOrSuperSection = document.getChildrenOfTypeSectionOrSuperSection();
 
         assertEquals(childrenOfTypeSectionOrSuperSection.size(), 9);
         assertEquals(childrenOfTypeSectionOrSuperSection.subList(1, 9)
@@ -198,7 +200,7 @@ public class OutlineDetectionTest extends AbstractTest {
 
 
     @SneakyThrows
-    protected DocumentWithVisualization buildGraph(String filename, ClassificationDocument classificationDocument) {
+    protected Document buildGraph(String filename, ClassificationDocument classificationDocument) {
 
         if (!filename.startsWith("files") && filename.startsWith("/")) {
             LayoutParsingRequest layoutParsingRequest = buildDefaultLayoutParsingRequest(Path.of(filename).getFileName().toString(), LayoutParsingType.REDACT_MANAGER, true);

layoutparser-service/layoutparser-service-server/src/test/java/com/knecon/fforesight/service/layoutparser/server/SimplifiedTextServiceTest.java

@@ -5,27 +5,30 @@ import static org.assertj.core.api.AssertionsForClassTypes.assertThat;
 import java.io.File;
 import java.util.List;
 import java.util.Map;
+import java.util.stream.Collectors;
 
 import org.junit.jupiter.api.Test;
 import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.core.io.ClassPathResource;
 
-import com.iqser.red.service.redaction.v1.server.model.document.nodes.Document;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.SimplifiedSectionText;
 import com.knecon.fforesight.service.layoutparser.internal.api.data.redaction.SimplifiedText;
 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.model.graph.nodes.Document;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.image.ImageServiceResponse;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.TableServiceResponse;
 import com.knecon.fforesight.service.layoutparser.processor.python_api.model.table.VisualLayoutParsingResponse;
 import com.knecon.fforesight.service.layoutparser.processor.services.SimplifiedSectionTextService;
 import com.knecon.fforesight.service.layoutparser.processor.services.factory.DocumentGraphFactory;
+import com.knecon.fforesight.service.layoutparser.server.utils.AbstractTest;
 
 import lombok.SneakyThrows;
 import lombok.extern.slf4j.Slf4j;
 
 @Slf4j
-public class SimplifiedTextServiceTest extends AbstractTest {
+public class SimplifiedTextServiceTest
+        extends AbstractTest {
 
     @Autowired
     private LayoutParsingPipeline layoutParsingPipeline;
@@ -39,20 +42,17 @@ public class SimplifiedTextServiceTest extends AbstractTest {
 
         File file = new ClassPathResource("files/headerFooterTest3Pages.pdf").getFile();
         String footerExample = "Footer to search for ";
-        String headerExample = "Header to search for ";
+        String headerExample ="Header to search for ";
         Document document = buildGraph(file);
         SimplifiedText simplifiedText = simplifiedSectionTextService.toSimplifiedText(document);
         List<SimplifiedSectionText> sectionTexts = simplifiedText.getSectionTexts();
-        assertThat(sectionTexts.stream()
-                           .filter(section -> section.getText().contains(footerExample))
-                           .toList().size()).isGreaterThan(0);
-        assertThat(sectionTexts.stream()
-                           .filter(section -> section.getText().contains(headerExample))
-                           .toList().size()).isGreaterThan(0);
+        assertThat(sectionTexts.stream().filter(section -> section.getText().contains(footerExample)).toList().size()).isGreaterThan(0);
+        assertThat(sectionTexts.stream().filter(section -> section.getText().contains(headerExample)).toList().size()).isGreaterThan(0);
+
+
 
     }
 
-
     @SneakyThrows
     protected Document buildGraph(File file) {
 
@@ -62,7 +62,9 @@ public class SimplifiedTextServiceTest extends AbstractTest {
                                                                                          new ImageServiceResponse(),
                                                                                          new TableServiceResponse(),
                                                                                          new VisualLayoutParsingResponse(),
-                                                                                         Map.of("file", file.toString()))).document();
+                                                                                         Map.of("file",file.toString())));
     }
 
+
+
 }