First steps for incremental analysis

Merge in RED/redaction-service from RED-1090 to master

* commit '0dd47111c5294d9a9a0a10e174c2832436983205':
  RED-1090: Return text with areas and sectionNumber

bamboo-specs/src/main/java/buildjob/PlanSpec.java

@@ -72,14 +72,6 @@ public class PlanSpec {
                         .jobs(new Job("Default Job",
                                 new BambooKey("JOB1"))
                                 .tasks(
-                                        new ScriptTask()
-                                                .description("Clean")
-                                                .inlineBody("#!/bin/bash\n" +
-                                                        "set -e\n" +
-                                                        "rm -rf ./*"),
-                                        new VcsCheckoutTask()
-                                                .description("Checkout Default Repository")
-                                                .checkoutItems(new CheckoutItem().defaultRepository()),
                                         new ScriptTask()
                                                 .description("Clean")
                                                 .inlineBody("#!/bin/bash\n" +

redaction-service-v1/pom.xml

@@ -5,7 +5,7 @@
     <parent>
         <artifactId>platform-dependency</artifactId>
         <groupId>com.iqser.red</groupId>
-        <version>1.0.2</version>
+        <version>1.0.3</version>
     </parent>
     <modelVersion>4.0.0</modelVersion>
 

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/AnalyzeRequest.java

@@ -0,0 +1,17 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class AnalyzeRequest {
+
+    private byte[] document;
+    private String ruleSetId;
+    private ManualRedactions manualRedactions;
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/AnalyzeResult.java

@@ -0,0 +1,19 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class AnalyzeResult {
+
+    private int numberOfPages;
+    private RedactionLog redactionLog;
+    private SectionGrid sectionGrid;
+    private Text text;
+
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/AnnotateRequest.java

@@ -0,0 +1,17 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class AnnotateRequest {
+
+    private byte[] document;
+    private RedactionLog redactionLog;
+    private SectionGrid sectionGrid;
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/AnnotateResponse.java

@@ -0,0 +1,15 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class AnnotateResponse {
+
+    private byte[] document;
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/ManualForceRedact.java

@@ -0,0 +1,19 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class ManualForceRedact {
+
+    private String id;
+    private String user;
+    private Status status;
+    private String legalBasis;
+
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/ManualRedactionType.java

@@ -1,5 +1,5 @@
 package com.iqser.red.service.redaction.v1.model;
 
 public enum ManualRedactionType {
-    ADD, REMOVE
+    ADD, REMOVE, FORCE_REDACT
 }

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/ManualRedactions.java

@@ -20,6 +20,9 @@ public class ManualRedactions {
     @Builder.Default
     private Set<IdRemoval> idsToRemove = new HashSet<>();
 
+    @Builder.Default
+    private Set<ManualForceRedact> forceRedacts = new HashSet<>();
+
     @Builder.Default
     private Set<ManualRedactionEntry> entriesToAdd = new HashSet<>();
 

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/ReanalyzeResult.java

@@ -0,0 +1,15 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class ReanalyzeResult {
+
+    private RedactionLog redactionLog;
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/RedactionLogEntry.java

@@ -37,4 +37,10 @@ public class RedactionLogEntry {
     private String textBefore;
     private String textAfter;
 
+    @Builder.Default
+    private List<Comment> comments = new ArrayList<>();
+
+    private int startOffset;
+    private int endOffset;
+
 }

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/RedactionRequest.java

@@ -13,6 +13,5 @@ public class RedactionRequest {
 
     private byte[] document;
     private String ruleSetId;
-    private boolean flatRedaction;
     private ManualRedactions manualRedactions;
 }

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/RenalyzeRequest.java

@@ -0,0 +1,22 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import java.time.OffsetDateTime;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class RenalyzeRequest {
+
+    private byte[] document;
+    private String ruleSetId;
+    private ManualRedactions manualRedactions;
+    private Text text;
+    private RedactionLog redactionLog;
+    private OffsetDateTime lastProcessed;
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/SectionArea.java

@@ -0,0 +1,33 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+import lombok.NonNull;
+import lombok.RequiredArgsConstructor;
+
+@Data
+@RequiredArgsConstructor
+@NoArgsConstructor
+@AllArgsConstructor
+public class SectionArea {
+
+    @NonNull
+    private Point topLeft;
+
+    @NonNull
+    private float width;
+
+    @NonNull
+    private float height;
+
+    @NonNull
+    private int page;
+
+    private String header;
+
+    public boolean contains(Rectangle other) {
+        return page == other.getPage() && this.topLeft.getX() <= other.getTopLeft().getX() && this.topLeft.getX() + this.getWidth() >= other.getTopLeft().getX() + other.getWidth() && this.getTopLeft().getY() <= other.getTopLeft().getY()  && this.getTopLeft().getY() + this.getHeight() >= other.getTopLeft().getY() + other.getHeight();
+    }
+
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/SectionText.java

@@ -0,0 +1,25 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class SectionText {
+
+    private int sectionNumber;
+    private String text;
+
+    private boolean isTable;
+    private String headline;
+
+    private List<SectionArea> sectionAreas = new ArrayList<>();
+
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/model/Text.java

@@ -0,0 +1,17 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@NoArgsConstructor
+@AllArgsConstructor
+public class Text {
+
+    private List<SectionText> sectionTexts = new ArrayList<>();
+
+}

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/resources/RedactionResource.java

@@ -1,7 +1,14 @@
 package com.iqser.red.service.redaction.v1.resources;
 
+import com.iqser.red.service.redaction.v1.model.AnalyzeRequest;
+import com.iqser.red.service.redaction.v1.model.AnalyzeResult;
+import com.iqser.red.service.redaction.v1.model.AnnotateRequest;
+import com.iqser.red.service.redaction.v1.model.AnnotateResponse;
+import com.iqser.red.service.redaction.v1.model.ReanalyzeResult;
 import com.iqser.red.service.redaction.v1.model.RedactionRequest;
 import com.iqser.red.service.redaction.v1.model.RedactionResult;
+import com.iqser.red.service.redaction.v1.model.RenalyzeRequest;
+
 import org.springframework.http.MediaType;
 import org.springframework.web.bind.annotation.PathVariable;
 import org.springframework.web.bind.annotation.PostMapping;
@@ -14,8 +21,14 @@ public interface RedactionResource {
     String RULE_SET_PARAMETER_NAME = "ruleSetId";
     String RULE_SET_PATH_VARIABLE = "/{" + RULE_SET_PARAMETER_NAME + "}";
 
-    @PostMapping(value = "/redact", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
-    RedactionResult redact(@RequestBody RedactionRequest redactionRequest);
+    @PostMapping(value = "/analyze", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
+    AnalyzeResult analyze(@RequestBody AnalyzeRequest analyzeRequest);
+
+    @PostMapping(value = "/reanalyze", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
+    ReanalyzeResult reanalyze(@RequestBody RenalyzeRequest renalyzeRequest);
+
+    @PostMapping(value = "/annotate", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
+    AnnotateResponse annotate(@RequestBody AnnotateRequest annotateRequest);
 
     @PostMapping(value = "/debug/classifications", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
     RedactionResult classify(@RequestBody RedactionRequest redactionRequest);

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

@@ -8,6 +8,7 @@ import java.util.Map;
 import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
 import com.iqser.red.service.redaction.v1.model.SectionGrid;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
+import com.iqser.red.service.redaction.v1.model.SectionText;
 
 import lombok.Data;
 import lombok.NoArgsConstructor;
@@ -18,6 +19,9 @@ public class Document {
 
     private List<Page> pages = new ArrayList<>();
     private List<Paragraph> paragraphs = new ArrayList<>();
+    private List<Header> headers = new ArrayList<>();
+    private List<Footer> footers = new ArrayList<>();
+    private List<UnclassifiedText> unclassifiedTexts = new ArrayList<>();
     private Map<Integer, List<Entity>> entities = new HashMap<>();
     private FloatFrequencyCounter textHeightCounter = new FloatFrequencyCounter();
     private FloatFrequencyCounter fontSizeCounter= new FloatFrequencyCounter();
@@ -29,4 +33,6 @@ public class Document {
     private SectionGrid sectionGrid = new SectionGrid();
     private long dictionaryVersion;
     private long rulesVersion;
+
+    private List<SectionText> sectionText = new ArrayList<>();
 }

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

@@ -0,0 +1,24 @@
+package com.iqser.red.service.redaction.v1.server.classification.model;
+
+import java.util.List;
+
+import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+
+@Data
+@AllArgsConstructor
+public class Footer {
+
+    private List<TextBlock> textBlocks;
+
+
+    public SearchableText getSearchableText() {
+
+        SearchableText searchableText = new SearchableText();
+        textBlocks.forEach(block -> searchableText.addAll(block.getSequences()));
+        return searchableText;
+    }
+
+}

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

@@ -0,0 +1,24 @@
+package com.iqser.red.service.redaction.v1.server.classification.model;
+
+import java.util.List;
+
+import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+
+@Data
+@AllArgsConstructor
+public class Header {
+
+    private List<TextBlock> textBlocks;
+
+
+    public SearchableText getSearchableText() {
+
+        SearchableText searchableText = new SearchableText();
+        textBlocks.forEach(block -> searchableText.addAll(block.getSequences()));
+        return searchableText;
+    }
+
+}

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

@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.classification.model;
 
+import java.awt.geom.Rectangle2D;
 import java.util.List;
 
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
@@ -16,6 +17,8 @@ public class Page {
     @NonNull
     private List<AbstractTextContainer> textBlocks;
 
+    private List<Rectangle2D> imageBounds;
+
     private Rectangle bodyTextFrame;
 
     private boolean landscape;

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

@@ -0,0 +1,24 @@
+package com.iqser.red.service.redaction.v1.server.classification.model;
+
+import java.util.List;
+
+import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+
+@Data
+@AllArgsConstructor
+public class UnclassifiedText {
+
+    private List<TextBlock> textBlocks;
+
+
+    public SearchableText getSearchableText() {
+
+        SearchableText searchableText = new SearchableText();
+        textBlocks.forEach(block -> searchableText.addAll(block.getSequences()));
+        return searchableText;
+    }
+
+}

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

@@ -111,6 +111,8 @@ public class ClassificationService {
             textBlock.setClassification("TextBlock Italic");
         } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock)) {
             textBlock.setClassification("TextBlock Unknown");
+        } else {
+            textBlock.setClassification("Other");
         }
     }
 

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

@@ -1,23 +1,35 @@
 package com.iqser.red.service.redaction.v1.server.controller;
 
+import com.iqser.red.service.redaction.v1.model.AnalyzeRequest;
+import com.iqser.red.service.redaction.v1.model.AnalyzeResult;
+import com.iqser.red.service.redaction.v1.model.AnnotateRequest;
+import com.iqser.red.service.redaction.v1.model.AnnotateResponse;
+import com.iqser.red.service.redaction.v1.model.ReanalyzeResult;
 import com.iqser.red.service.redaction.v1.model.RedactionLog;
 import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
 import com.iqser.red.service.redaction.v1.model.RedactionRequest;
 import com.iqser.red.service.redaction.v1.model.RedactionResult;
+import com.iqser.red.service.redaction.v1.model.RenalyzeRequest;
 import com.iqser.red.service.redaction.v1.model.SectionGrid;
+import com.iqser.red.service.redaction.v1.model.Text;
 import com.iqser.red.service.redaction.v1.resources.RedactionResource;
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
 import com.iqser.red.service.redaction.v1.server.classification.model.Page;
 import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
+import com.iqser.red.service.redaction.v1.server.redaction.service.AnnotationService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.DictionaryService;
 import com.iqser.red.service.redaction.v1.server.redaction.service.DroolsExecutionService;
 import com.iqser.red.service.redaction.v1.server.redaction.service.EntityRedactionService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.ReanalyzeService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.RedactionLogCreatorService;
 import com.iqser.red.service.redaction.v1.server.segmentation.PdfSegmentationService;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
-import com.iqser.red.service.redaction.v1.server.visualization.service.AnnotationHighlightService;
 import com.iqser.red.service.redaction.v1.server.visualization.service.PdfVisualisationService;
+
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
+
 import org.apache.pdfbox.pdmodel.PDDocument;
 import org.springframework.web.bind.annotation.PathVariable;
 import org.springframework.web.bind.annotation.RequestBody;
@@ -35,34 +47,37 @@ public class RedactionController implements RedactionResource {
 
     private final PdfVisualisationService pdfVisualisationService;
     private final PdfSegmentationService pdfSegmentationService;
-    private final AnnotationHighlightService annotationHighlightService;
+    private final RedactionLogCreatorService redactionLogCreatorService;
     private final EntityRedactionService entityRedactionService;
     private final DroolsExecutionService droolsExecutionService;
+    private final DictionaryService dictionaryService;
+    private final AnnotationService annotationService;
+    private final ReanalyzeService reanalyzeService;
 
 
     @Override
-    public RedactionResult redact(@RequestBody RedactionRequest redactionRequest) {
+    public AnalyzeResult analyze(@RequestBody AnalyzeRequest analyzeRequest) {
 
-        try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(redactionRequest.getDocument()))) {
+        try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(analyzeRequest.getDocument()))) {
             pdDocument.setAllSecurityToBeRemoved(true);
 
             Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
 
             log.info("Document structure analysis successful, starting redaction analysis...");
 
-            entityRedactionService.processDocument(classifiedDoc, redactionRequest.getRuleSetId(), redactionRequest.getManualRedactions());
-            annotationHighlightService.highlight(pdDocument, classifiedDoc, redactionRequest.isFlatRedaction(), redactionRequest
-                    .getManualRedactions(), redactionRequest.getRuleSetId());
+            entityRedactionService.processDocument(classifiedDoc, analyzeRequest.getRuleSetId(), analyzeRequest.getManualRedactions());
+            redactionLogCreatorService.createRedactionLog(classifiedDoc, pdDocument.getNumberOfPages(), analyzeRequest.getManualRedactions(), analyzeRequest
+                    .getRuleSetId());
 
             log.info("Redaction analysis successful...");
 
-            return convert(pdDocument,
-                    classifiedDoc.getPages().size(),
-                    classifiedDoc.getRedactionLogEntities(),
-                    classifiedDoc.getSectionGrid(),
-                    classifiedDoc.getDictionaryVersion(),
-                    classifiedDoc.getRulesVersion(),
-                    redactionRequest.getRuleSetId());
+            return AnalyzeResult.builder()
+                    .sectionGrid(classifiedDoc.getSectionGrid())
+                    .redactionLog(new RedactionLog(classifiedDoc.getRedactionLogEntities(), classifiedDoc.getDictionaryVersion(), classifiedDoc
+                            .getRulesVersion(), analyzeRequest.getRuleSetId()))
+                    .numberOfPages(classifiedDoc.getPages().size())
+                    .text(new Text(classifiedDoc.getSectionText()))
+                    .build();
 
         } catch (Exception e) {
             throw new RedactionException(e);
@@ -71,6 +86,31 @@ public class RedactionController implements RedactionResource {
     }
 
 
+    public ReanalyzeResult reanalyze(@RequestBody RenalyzeRequest renalyzeRequest) {
+
+        return reanalyzeService.reanalyze(renalyzeRequest);
+    }
+
+
+    public AnnotateResponse annotate(@RequestBody AnnotateRequest annotateRequest) {
+
+        try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(annotateRequest.getDocument()))) {
+
+            pdDocument.setAllSecurityToBeRemoved(true);
+            dictionaryService.updateDictionary(annotateRequest.getRedactionLog().getRuleSetId());
+            annotationService.annotate(pdDocument, annotateRequest.getRedactionLog(), annotateRequest.getSectionGrid());
+
+            try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream()) {
+                pdDocument.save(byteArrayOutputStream);
+                return AnnotateResponse.builder().document(byteArrayOutputStream.toByteArray()).build();
+            }
+
+        } catch (Exception e) {
+            throw new RedactionException(e);
+        }
+    }
+
+
     @Override
     public RedactionResult classify(@RequestBody RedactionRequest pdfSegmentationRequest) {
 
@@ -136,11 +176,14 @@ public class RedactionController implements RedactionResource {
 
     @Override
     public void updateRules(@PathVariable(RULE_SET_PARAMETER_NAME) String ruleSetId) {
+
         droolsExecutionService.updateRules(ruleSetId);
     }
 
+
     @Override
     public void testRules(@RequestBody String rules) {
+
         droolsExecutionService.testRules(rules);
     }
 
@@ -152,8 +195,8 @@ public class RedactionController implements RedactionResource {
 
 
     private RedactionResult convert(PDDocument document, int numberOfPages,
-                                    List<RedactionLogEntry> redactionLogEntities,
-                                    SectionGrid sectionGrid, long dictionaryVersion, long rulesVersion, String ruleSetId) throws IOException {
+                                    List<RedactionLogEntry> redactionLogEntities, SectionGrid sectionGrid,
+                                    long dictionaryVersion, long rulesVersion, String ruleSetId) throws IOException {
 
         try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream()) {
             document.save(byteArrayOutputStream);

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

@@ -0,0 +1,83 @@
+package com.iqser.red.service.redaction.v1.server.parsing;
+
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.List;
+
+import org.apache.pdfbox.text.PDFTextStripperByArea;
+import org.apache.pdfbox.text.TextPosition;
+
+import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
+
+import lombok.Getter;
+import lombok.Setter;
+
+public class PDFAreaTextStripper extends PDFTextStripperByArea {
+
+    @Getter
+    private List<TextPositionSequence> textPositionSequences = new ArrayList<>();
+
+    @Setter
+    private int pageNumber;
+
+    public PDFAreaTextStripper() throws IOException {
+
+    }
+
+    @Override
+    public void writeString(String text, List<TextPosition> textPositions) throws IOException {
+
+        int startIndex = 0;
+        for (int i = 0; i <= textPositions.size() - 1; i++) {
+
+            if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
+                    .getUnicode()
+                    .equals("\u00A0"))) {
+                startIndex++;
+                continue;
+            }
+
+            // Strange but sometimes this is happening, for example: Metolachlor2.pdf
+            if (i > 0 && textPositions.get(i).getX() < textPositions.get(i - 1).getX()) {
+                List<TextPosition> sublist = textPositions.subList(startIndex, i);
+                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0)
+                        .getUnicode()
+                        .equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+                    textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
+                }
+                startIndex = i;
+            }
+
+            if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
+                    .getUnicode()
+                    .equals("\u00A0")) && i <= textPositions.size() - 2) {
+                List<TextPosition> sublist = textPositions.subList(startIndex, i);
+                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0)
+                        .getUnicode()
+                        .equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+                    textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
+                }
+                startIndex = i + 1;
+            }
+        }
+
+        List<TextPosition> sublist = textPositions.subList(startIndex, textPositions.size());
+        if (!sublist.isEmpty() && (sublist.get(sublist.size() - 1)
+                .getUnicode()
+                .equals(" ") || sublist.get(sublist.size() - 1).getUnicode().equals("\u00A0"))) {
+            sublist = sublist.subList(0, sublist.size() - 1);
+        }
+        if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0)
+                .getUnicode()
+                .equals("\u00A0")))) {
+            textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
+        }
+        super.writeString(text);
+    }
+
+
+    public void clearPositions(){
+        textPositionSequences = new ArrayList<>();
+    }
+
+}

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

@@ -1,6 +1,8 @@
 package com.iqser.red.service.redaction.v1.server.parsing;
 
+import java.awt.geom.AffineTransform;
 import java.awt.geom.Point2D;
+import java.awt.geom.Rectangle2D;
 import java.io.IOException;
 import java.util.ArrayList;
 import java.util.List;
@@ -28,10 +30,15 @@ import org.apache.pdfbox.contentstream.operator.state.SetLineWidth;
 import org.apache.pdfbox.contentstream.operator.state.SetRenderingIntent;
 import org.apache.pdfbox.contentstream.operator.text.SetFontAndSize;
 import org.apache.pdfbox.cos.COSBase;
+import org.apache.pdfbox.cos.COSName;
 import org.apache.pdfbox.cos.COSNumber;
 import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.pdmodel.graphics.PDXObject;
+import org.apache.pdfbox.pdmodel.graphics.image.PDImageXObject;
 import org.apache.pdfbox.text.PDFTextStripper;
 import org.apache.pdfbox.text.TextPosition;
+import org.apache.pdfbox.util.Matrix;
 
 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Ruling;
@@ -43,8 +50,17 @@ import lombok.extern.slf4j.Slf4j;
 @Slf4j
 public class PDFLinesTextStripper extends PDFTextStripper {
 
+    @Setter
+    protected PDPage pdpage;
+
     @Getter
-    private int maxCharWidths;
+    private int minCharWidth;
+
+    @Getter
+    private int maxCharWidth;
+    
+    @Getter
+    private int minCharHeight;
 
     @Getter
     private int maxCharHeight;
@@ -57,13 +73,18 @@ public class PDFLinesTextStripper extends PDFTextStripper {
 
     private final List<Ruling> graphicsPath = new ArrayList<>();
 
+    @Getter
+    private List<Rectangle2D> imageBounds = new ArrayList<>();
+
     private float path_x;
     private float path_y;
 
     @Setter
     private int pageNumber;
 
+
     public PDFLinesTextStripper() throws IOException {
+
         super();
         this.addOperator(new SetStrokingColorSpace());
         this.addOperator(new SetNonStrokingColorSpace());
@@ -87,9 +108,9 @@ public class PDFLinesTextStripper extends PDFTextStripper {
         this.addOperator(new SetLineWidth());
     }
 
+
     @Override
-    protected void processOperator(Operator operator, List<COSBase> arguments)
-            throws IOException {
+    protected void processOperator(Operator operator, List<COSBase> arguments) throws IOException {
 
         String operation = operator.getName();
 
@@ -110,9 +131,11 @@ public class PDFLinesTextStripper extends PDFTextStripper {
 
                     // The direction of vertical lines must always be from bottom to top for the table extraction algorithm.
                     if (pos.getY() > path_y) {
-                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, path_y), new Point2D.Float((float) pos.getX(), (float) pos.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, path_y), new Point2D.Float((float) pos.getX(), (float) pos
+                                .getY())));
                     } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, (float) pos.getY()), new Point2D.Float((float) pos.getX(), path_y)));
+                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, (float) pos.getY()), new Point2D.Float((float) pos
+                                .getX(), path_y)));
                     }
 
                     path_x = (float) pos.getX();
@@ -133,19 +156,25 @@ public class PDFLinesTextStripper extends PDFTextStripper {
                     Point2D p2 = transformPosition(x + width, y + height);
 
                     // Horizontal lines
-                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p2.getX(), (float) p1.getY())));
-                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p2.getX(), (float) p2.getY())));
+                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p2
+                            .getX(), (float) p1.getY())));
+                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p2
+                            .getX(), (float) p2.getY())));
 
                     // Vertical lines, direction must always be from bottom to top for the table extraction algorithm.
                     if (p2.getY() > p1.getY()) {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p1.getY()), new Point2D.Float((float) p2.getX(), (float) p2.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p1.getY()), new Point2D.Float((float) p2
+                                .getX(), (float) p2.getY())));
                     } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p2.getY()), new Point2D.Float((float) p2.getX(), (float) p1.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p2.getY()), new Point2D.Float((float) p2
+                                .getX(), (float) p1.getY())));
                     }
                     if (p2.getY() > p1.getY()) {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p1.getX(), (float) p2.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p1
+                                .getX(), (float) p2.getY())));
                     } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p1.getX(), (float) p1.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p1
+                                .getX(), (float) p1.getY())));
                     }
                 }
                 break;
@@ -168,12 +197,59 @@ public class PDFLinesTextStripper extends PDFTextStripper {
             case OperatorName.ENDPATH:
                 graphicsPath.clear();
                 break;
+
+            case OperatorName.DRAW_OBJECT:
+                processImageOperation(arguments);
+                break;
+
         }
 
         super.processOperator(operator, arguments);
     }
 
+
+    protected void processImageOperation(List<COSBase> arguments) {
+
+        try {
+            COSName objectName = (COSName) arguments.get(0);
+            PDXObject xobject = getResources().getXObject(objectName);
+            if (xobject instanceof PDImageXObject) {
+                PDImageXObject pdfImage = (PDImageXObject) xobject;
+
+                Rectangle2D imageBounds = calculateImagePosition(pdfImage);
+
+                Rectangle2D rect = new Rectangle2D.Float((float) imageBounds.getX(), (float) imageBounds.getY(), (float) imageBounds
+                        .getWidth(), (float) imageBounds.getHeight());
+
+                if (rect.getHeight() > 2 && rect.getWidth() > 2) {
+                    this.imageBounds.add(rect);
+                }
+            }
+        } catch (Exception e) {
+            log.warn("Problem during image extraction: {}", e.getMessage());
+        }
+    }
+
+
+    private Rectangle2D calculateImagePosition(PDImageXObject pdfImage) throws IOException {
+
+        Matrix ctm = getGraphicsState().getCurrentTransformationMatrix();
+
+        Rectangle2D imageBounds = pdfImage.getImage().getRaster().getBounds();
+
+        AffineTransform imageTransform = new AffineTransform(ctm.createAffineTransform());
+        imageTransform.scale(1.0 / pdfImage.getWidth(), -1.0 / pdfImage.getHeight());
+        imageTransform.translate(0, -pdfImage.getHeight());
+
+        AffineTransform pageTransform = new AffineTransform();
+        pageTransform.concatenate(imageTransform);
+
+        return pageTransform.createTransformedShape(imageBounds).getBounds2D();
+    }
+
+
     private float floatValue(COSBase value) {
+
         if (value instanceof COSNumber) {
             return ((COSNumber) value).floatValue();
         } else {
@@ -181,38 +257,56 @@ public class PDFLinesTextStripper extends PDFTextStripper {
         }
     }
 
+
     private Point2D.Float transformPosition(float x, float y) {
+
         return super.transformedPoint(x, y);
     }
 
+
     private void addVisibleRulings(List<Ruling> path, boolean stroke) throws IOException {
 
         try {
-            if (stroke && !getGraphicsState().getStrokingColor().isPattern() && getGraphicsState().getStrokingColor().toRGB() == 0 || !stroke && !getGraphicsState().getNonStrokingColor().isPattern() && getGraphicsState().getNonStrokingColor().toRGB() == 0) {
+            if (stroke && !getGraphicsState().getStrokingColor().isPattern() && getGraphicsState().getStrokingColor()
+                    .toRGB() == 0 || !stroke && !getGraphicsState().getNonStrokingColor()
+                    .isPattern() && getGraphicsState().getNonStrokingColor().toRGB() == 0) {
                 rulings.addAll(path);
             }
         } catch (UnsupportedOperationException e) {
-            log.error("UnsupportedOperationException: " + getGraphicsState().getStrokingColor().getColorSpace().getName() + " or " + getGraphicsState().getNonStrokingColor().getColorSpace().getName() + " does not support toRGB");
+            log.error("UnsupportedOperationException: " + getGraphicsState().getStrokingColor()
+                    .getColorSpace()
+                    .getName() + " or " + getGraphicsState().getNonStrokingColor()
+                    .getColorSpace()
+                    .getName() + " does not support toRGB");
         }
     }
 
+
     @Override
     public void writeString(String text, List<TextPosition> textPositions) throws IOException {
 
         int startIndex = 0;
         for (int i = 0; i <= textPositions.size() - 1; i++) {
 
+            int charWidth = (int) textPositions.get(i).getWidthDirAdj();
+            if (charWidth < minCharWidth) {
+                minCharWidth = charWidth;
+            }
+            if (charWidth > maxCharWidth) {
+                maxCharWidth = charWidth;
+            }
+
             int charHeight = (int) textPositions.get(i).getHeightDir();
-            if(charHeight > maxCharHeight){
+            if (charHeight < minCharHeight) {
+                minCharHeight = charHeight;
+            }
+            if (charWidth > maxCharHeight) {
                 maxCharHeight = charHeight;
             }
 
-            int charWidth = (int) textPositions.get(i).getWidthDirAdj();
-            if(charWidth > maxCharWidths){
-                maxCharWidths = charWidth;
-            }
-
-            if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0"))) {
+            if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
+                    .getUnicode()
+                    .equals("\u00A0"))) {
                 startIndex++;
                 continue;
             }
@@ -220,15 +314,21 @@ public class PDFLinesTextStripper extends PDFTextStripper {
             // Strange but sometimes this is happening, for example: Metolachlor2.pdf
             if (i > 0 && textPositions.get(i).getX() < textPositions.get(i - 1).getX()) {
                 List<TextPosition> sublist = textPositions.subList(startIndex, i);
-                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0)
+                        .getUnicode()
+                        .equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
                     textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
                 }
                 startIndex = i;
             }
 
-            if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0")) && i <= textPositions.size() - 2) {
+            if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
+                    .getUnicode()
+                    .equals("\u00A0")) && i <= textPositions.size() - 2) {
                 List<TextPosition> sublist = textPositions.subList(startIndex, i);
-                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0)
+                        .getUnicode()
+                        .equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
                     textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
                 }
                 startIndex = i + 1;
@@ -236,21 +336,29 @@ public class PDFLinesTextStripper extends PDFTextStripper {
         }
 
         List<TextPosition> sublist = textPositions.subList(startIndex, textPositions.size());
-        if (!sublist.isEmpty() && (sublist.get(sublist.size() - 1).getUnicode().equals(" ") || sublist.get(sublist.size() - 1).getUnicode().equals("\u00A0"))) {
+        if (!sublist.isEmpty() && (sublist.get(sublist.size() - 1)
+                .getUnicode()
+                .equals(" ") || sublist.get(sublist.size() - 1).getUnicode().equals("\u00A0"))) {
             sublist = sublist.subList(0, sublist.size() - 1);
         }
-        if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+        if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0)
+                .getUnicode()
+                .equals("\u00A0")))) {
             textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
         }
         super.writeString(text);
     }
 
+
     @Override
     public String getText(PDDocument doc) throws IOException {
 
-        maxCharWidths = 0;
-        maxCharWidths = 0;
+        minCharWidth = Integer.MAX_VALUE;
+        maxCharWidth = 0;
+        minCharHeight = Integer.MAX_VALUE;
+        maxCharHeight = 0;
         textPositionSequences.clear();
+        imageBounds = new ArrayList<>();
         rulings.clear();
         graphicsPath.clear();
         path_x = 0.0f;

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/parsing/model/ParsedElements.java

@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.parsing.model;
 
+import java.awt.geom.Rectangle2D;
 import java.util.List;
 
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Ruling;
@@ -13,10 +14,11 @@ public class ParsedElements {
 
     private List<TextPositionSequence> sequences;
     private List<Ruling> rulings;
+    private List<Rectangle2D> imageBounds;
 
     private boolean landscape;
     private boolean rotated;
 
+    private float minCharWidth;
     private float maxCharWidth;
-    private float maxCharHeight;
 }

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

@@ -28,6 +28,14 @@ public class Dictionary {
     }
 
 
+    public int getDictionaryRank(String type){
+        if(!localAccessMap.containsKey(type)){
+            return 0;
+        }
+        return localAccessMap.get(type).getRank();
+    }
+
+
     public boolean isRecommendation(String type) {
 
         DictionaryModel model = localAccessMap.get(type);

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

@@ -0,0 +1,15 @@
+package com.iqser.red.service.redaction.v1.server.redaction.model;
+
+import java.util.Set;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+
+@Data
+@AllArgsConstructor
+public class DictionaryIncrement {
+
+    private Set<String> values;
+    private long dictionaryVersion;
+
+}

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

@@ -39,7 +39,7 @@ public class Entity {
     private String textAfter;
 
 
-    public Entity(String word, String type, boolean redaction, String redactionReason, List<EntityPositionSequence> positionSequences, String headline, int matchedRule, int sectionNumber, String legalBasis, boolean isDictionaryEntry, String textBefore, String textAfter) {
+    public Entity(String word, String type, boolean redaction, String redactionReason, List<EntityPositionSequence> positionSequences, String headline, int matchedRule, int sectionNumber, String legalBasis, boolean isDictionaryEntry, String textBefore, String textAfter, Integer start, Integer end) {
 
         this.word = word;
         this.type = type;
@@ -53,6 +53,8 @@ public class Entity {
         this.isDictionaryEntry = isDictionaryEntry;
         this.textBefore = textBefore;
         this.textAfter = textAfter;
+        this.start = start;
+        this.end = end;
     }
 
 

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

@@ -0,0 +1,34 @@
+package com.iqser.red.service.redaction.v1.server.redaction.model;
+
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+
+import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@NoArgsConstructor
+public class ReanalysisSection {
+
+    private int sectionNumber;
+    private String headline;
+    private List<TextBlock> textBlocks;
+    private Map<String, CellValue> tabularData = new HashMap<>();
+    private List<Integer> cellStarts;
+
+
+    public SearchableText getSearchableText() {
+
+        SearchableText searchableText = new SearchableText();
+        textBlocks.forEach(block -> {
+            if (block instanceof TextBlock) {
+                searchableText.addAll(block.getSequences());
+            }
+        });
+        return searchableText;
+    }
+
+}

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

@@ -32,6 +32,7 @@ public class SearchableText {
 
     }
 
+
     @SuppressWarnings("checkstyle:ModifiedControlVariable")
     public List<EntityPositionSequence> getSequences(String searchString, boolean caseInsensitive,
                                                      List<TextPositionSequence> sequencesSubList) {
@@ -66,9 +67,12 @@ public class SearchableText {
             for (int j = 0; j < searchSpace.get(i).length(); j++) {
 
                 if (i > 0 && j == 0 && searchSpace.get(i).charAt(0, caseInsensitive) == ' ' && searchSpace.get(i - 1)
-                        .charAt(searchSpace.get(i - 1).length() - 1, caseInsensitive) == ' ' || j > 0 && searchSpace.get(i)
-                        .charAt(j, caseInsensitive) == ' ' && searchSpace.get(i).charAt(j - 1, caseInsensitive) == ' ') {
-                    if (j == searchSpace.get(i).length() - 1 && counter != 0 && !partMatch.getTextPositions().isEmpty()) {
+                        .charAt(searchSpace.get(i - 1)
+                                .length() - 1, caseInsensitive) == ' ' || j > 0 && searchSpace.get(i)
+                        .charAt(j, caseInsensitive) == ' ' && searchSpace.get(i)
+                        .charAt(j - 1, caseInsensitive) == ' ') {
+                    if (j == searchSpace.get(i).length() - 1 && counter != 0 && !partMatch.getTextPositions()
+                            .isEmpty()) {
                         crossSequenceParts.add(partMatch);
                     }
                     continue;
@@ -80,8 +84,8 @@ public class SearchableText {
                     counter++;
                 }
 
-                if (searchSpace.get(i)
-                        .charAt(j, caseInsensitive) == searchChars[counter] || counter != 0 && searchSpace.get(i)
+                if (searchSpace.get(i).charAt(j, caseInsensitive) == searchChars[counter] || counter != 0 && searchSpace
+                        .get(i)
                         .charAt(j, caseInsensitive) == '-') {
 
                     if (counter != 0 || i == 0 && j == 0 || j != 0 && isSeparator(searchSpace.get(i)
@@ -100,14 +104,15 @@ public class SearchableText {
                     if (counter == searchString.length()) {
                         crossSequenceParts.add(partMatch);
 
-                        if (i == searchSpace.size() - 1 && j == searchSpace.get(i).length() - 1 || j != searchSpace.get(i)
-                                .length() - 1 && isSeparator(searchSpace.get(i)
+                        if (i == searchSpace.size() - 1 && j == searchSpace.get(i)
+                                .length() - 1 || j != searchSpace.get(i).length() - 1 && isSeparator(searchSpace.get(i)
                                 .charAt(j + 1, caseInsensitive)) || j == searchSpace.get(i)
                                 .length() - 1 && isSeparator(searchSpace.get(i + 1)
-                                .charAt(0, caseInsensitive)) || j == searchSpace.get(i).length() - 1 && searchSpace.get(i)
+                                .charAt(0, caseInsensitive)) || j == searchSpace.get(i)
+                                .length() - 1 && searchSpace.get(i)
                                 .charAt(j, caseInsensitive) != ' ' && searchSpace.get(i + 1)
                                 .charAt(0, caseInsensitive) != ' ') {
-                            finalMatches.addAll(buildEntityPositionSequence(crossSequenceParts));
+                            finalMatches.addAll(buildEntityPositionSequence(crossSequenceParts, normalizedSearchString));
                         }
 
                         counter = 0;
@@ -130,15 +135,21 @@ public class SearchableText {
         }
 
         return finalMatches;
-
     }
 
 
-    private List<EntityPositionSequence> buildEntityPositionSequence(List<TextPositionSequence> crossSequenceParts) {
+    private List<EntityPositionSequence> buildEntityPositionSequence(List<TextPositionSequence> crossSequenceParts,
+                                                                     String searchString) {
+
+        List<EntityPositionSequence> result = new ArrayList<>();
+        String asString = buildString(crossSequenceParts);
+        if (!asString.equalsIgnoreCase(searchString)) {
+            return result;
+        }
 
         String plainId = IdBuilder.buildId(crossSequenceParts);
         String id = plainId;
-        List<EntityPositionSequence> result = new ArrayList<>();
+
         int currentPage = -1;
         int idDiffentPageSuffix = 1;
         EntityPositionSequence entityPositionSequence = new EntityPositionSequence(id);
@@ -173,6 +184,12 @@ public class SearchableText {
     @Override
     public String toString() {
 
+        return buildString(sequences);
+    }
+
+
+    public String buildString(List<TextPositionSequence> sequences) {
+
         StringBuilder sb = new StringBuilder();
 
         TextPositionSequence previous = null;

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

@@ -62,6 +62,13 @@ public class Section {
     }
 
 
+    public boolean hasTableHeader(String headerName) {
+
+        String cleanHeaderName = headerName.replaceAll("\n", "").replaceAll(" ", "").replaceAll("-", "");
+        return tabularData != null && tabularData.containsKey(cleanHeaderName);
+    }
+
+
     public boolean matchesType(String type) {
 
         return entities.stream().anyMatch(entity -> entity.getType().equals(type));
@@ -74,6 +81,34 @@ public class Section {
     }
 
 
+    public void expandByRegEx(String type, String pattern, boolean patternCaseInsensitive, int group) {
+
+        Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
+
+        Set<Entity> expanded = new HashSet<>();
+        for (Entity entity : entities) {
+
+            if (!entity.getType().equals(type) || entity.getTextAfter() == null) {
+                continue;
+            }
+
+            Matcher matcher = compiledPattern.matcher(entity.getTextAfter());
+
+            while (matcher.find()) {
+                String match = matcher.group(group);
+                if (StringUtils.isNotBlank(match)) {
+                    expanded.addAll(findEntities(entity.getWord() + match, type, false, entity.isRedaction(), entity.getMatchedRule(), entity
+                            .getRedactionReason(), entity.getLegalBasis()));
+
+                }
+            }
+        }
+
+        EntitySearchUtils.addEntitiesWithHigherRank(entities, expanded, dictionary);
+        EntitySearchUtils.removeEntitiesContainedInLarger(entities);
+    }
+
+
     public void redact(String type, int ruleNumber, String reason, String legalBasis) {
 
         boolean hasRecommendationDictionary = dictionaryTypes.contains(RECOMMENDATION_PREFIX + type);
@@ -105,6 +140,48 @@ public class Section {
     }
 
 
+    public void expandToHintAnnotationByRegEx(String type, String pattern, boolean patternCaseInsensitive, int group, String asType) {
+
+        Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
+
+        Set<Entity> expanded = new HashSet<>();
+        for (Entity entity : entities) {
+
+            if (!entity.getType().equals(type) || entity.getTextAfter() == null) {
+                continue;
+            }
+
+            Matcher matcher = compiledPattern.matcher(entity.getTextAfter());
+
+            while (matcher.find()) {
+                String match = matcher.group(group);
+                if (StringUtils.isNotBlank(match)) {
+                    expanded.addAll(findEntities(entity.getWord() + match, asType, false, false, 0, null, null));
+                }
+            }
+        }
+
+        EntitySearchUtils.addEntitiesWithHigherRank(entities, expanded, dictionary);
+        EntitySearchUtils.removeEntitiesContainedInLarger(entities);
+    }
+
+
+    public void addHintAnnotationByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType) {
+
+        Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
+
+        Matcher matcher = compiledPattern.matcher(searchText);
+
+        while (matcher.find()) {
+            String match = matcher.group(group);
+            if (StringUtils.isNotBlank(match)) {
+                Set<Entity> found = findEntities(match.trim(), asType, false, false, 0, null, null);
+                EntitySearchUtils.addEntitiesWithHigherRank(entities, found, dictionary);
+            }
+        }
+    }
+
+
     public void redactIfPrecededBy(String prefix, String type, int ruleNumber, String reason, String legalBasis) {
 
         entities.forEach(entity -> {
@@ -121,7 +198,14 @@ public class Section {
     public void addHintAnnotation(String value, String asType) {
 
         Set<Entity> found = findEntities(value.trim(), asType, true, false, 0, null, null);
-        addNewerToEntities(found);
+        EntitySearchUtils.addEntitiesIgnoreRank(entities, found);
+    }
+
+
+    public void addRedaction(String value, String asType, int ruleNumber, String reason, String legalBasis) {
+
+        Set<Entity> found = findEntities(value.trim(), asType, true, true, ruleNumber, reason, legalBasis);
+        EntitySearchUtils.addEntitiesIgnoreRank(entities, found);
     }
 
 
@@ -134,14 +218,44 @@ public class Section {
             for (String value : values) {
                 if (StringUtils.isNotBlank(value)) {
                     Set<Entity> found = findEntities(value.trim(), asType, false, true, ruleNumber, reason, legalBasis);
-                    addNewerToEntities(found);
+                    EntitySearchUtils.addEntitiesWithHigherRank(entities, found, dictionary);
+
+                    if (redactEverywhere && !isLocal()) {
+                        localDictionaryAdds.computeIfAbsent(asType, (x) -> new HashSet<>()).add(value.trim());
+                    }
                 }
             }
         }
     }
 
 
-    public void redactByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType, int ruleNumber, String reason, String legalBasis) {
+    public void recommendLineAfter(String start, String asType) {
+
+        String[] values = StringUtils.substringsBetween(text, start, "\n");
+
+        if (values != null) {
+            for (String value : values) {
+
+                String trimmedValue = value.trim();
+                String cleanValue;
+                if (trimmedValue.startsWith(":")) {
+                    cleanValue = trimmedValue.substring(1).trim();
+                } else {
+                    cleanValue = trimmedValue;
+                }
+
+                if (StringUtils.isNotBlank(cleanValue) && cleanValue.length() >= 3) {
+                    localDictionaryAdds.computeIfAbsent(RECOMMENDATION_PREFIX + asType, (x) -> new HashSet<>())
+                            .add(cleanValue);
+                }
+            }
+        }
+    }
+
+
+    public void redactByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType, int ruleNumber,
+                              String reason, String legalBasis) {
+
         Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
 
         Matcher matcher = compiledPattern.matcher(searchText);
@@ -150,13 +264,14 @@ public class Section {
             String match = matcher.group(group);
             if (StringUtils.isNotBlank(match)) {
                 Set<Entity> found = findEntities(match.trim(), asType, false, true, ruleNumber, reason, legalBasis);
-                addNewerToEntities(found);
+                EntitySearchUtils.addEntitiesWithHigherRank(entities, found, dictionary);
             }
         }
     }
 
 
     public void addRecommendationByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType) {
+
         Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
 
         Matcher matcher = compiledPattern.matcher(text);
@@ -164,8 +279,24 @@ public class Section {
         while (matcher.find()) {
             String match = matcher.group(group);
             if (StringUtils.isNotBlank(match) && match.length() >= 3) {
-                localDictionaryAdds.computeIfAbsent(RECOMMENDATION_PREFIX + asType, (x) -> new HashSet<>())
-                        .add(match);
+                localDictionaryAdds.computeIfAbsent(RECOMMENDATION_PREFIX + asType, (x) -> new HashSet<>()).add(match);
+            }
+        }
+    }
+
+
+    public void redactAndRecommendByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType,
+                                          int ruleNumber, String reason, String legalBasis) {
+
+        Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
+
+        Matcher matcher = compiledPattern.matcher(searchText);
+
+        while (matcher.find()) {
+            String match = matcher.group(group);
+            if (StringUtils.isNotBlank(match) && match.length() >= 3) {
+                localDictionaryAdds.computeIfAbsent(RECOMMENDATION_PREFIX + asType, (x) -> new HashSet<>()).add(match);
+                localDictionaryAdds.computeIfAbsent(asType, (x) -> new HashSet<>()).add(match);
             }
         }
     }
@@ -181,7 +312,7 @@ public class Section {
                 if (StringUtils.isNotBlank(value)) {
 
                     Set<Entity> found = findEntities(value.trim(), asType, false, true, ruleNumber, reason, legalBasis);
-                    addNewerToEntities(found);
+                    EntitySearchUtils.addEntitiesWithHigherRank(entities, found, dictionary);
 
                     if (redactEverywhere && !isLocal()) {
                         localDictionaryAdds.computeIfAbsent(asType, (x) -> new HashSet<>()).add(value.trim());
@@ -192,8 +323,7 @@ public class Section {
     }
 
 
-    public void redactLinesBetween(String start, String stop, String asType, int ruleNumber,
-                                   boolean redactEverywhere,
+    public void redactLinesBetween(String start, String stop, String asType, int ruleNumber, boolean redactEverywhere,
                                    String reason, String legalBasis) {
 
         String[] values = StringUtils.substringsBetween(text, start, stop);
@@ -210,7 +340,7 @@ public class Section {
                         }
 
                         Set<Entity> found = findEntities(line.trim(), asType, false, true, ruleNumber, reason, legalBasis);
-                        addNewerToEntities(found);
+                        EntitySearchUtils.addEntitiesWithHigherRank(entities, found, dictionary);
 
                         if (redactEverywhere && !isLocal()) {
                             localDictionaryAdds.computeIfAbsent(asType, (x) -> new HashSet<>()).add(line.trim());
@@ -228,8 +358,7 @@ public class Section {
     }
 
 
-    public void redactCell(String cellHeader, int ruleNumber, String type, boolean addAsRecommendations, String
-            reason,
+    public void redactCell(String cellHeader, int ruleNumber, String type, boolean addAsRecommendations, String reason,
                            String legalBasis) {
 
         annotateCell(cellHeader, ruleNumber, type, true, addAsRecommendations, reason, legalBasis);
@@ -290,7 +419,7 @@ public class Section {
             singleEntitySet.add(entity);
             EntitySearchUtils.clearAndFindPositions(singleEntitySet, searchableText, dictionary);
 
-            addNewerToEntities(entity);
+            EntitySearchUtils.addEntitiesWithHigherRank(entities, entity, dictionary);
 
             EntitySearchUtils.removeEntitiesContainedInLarger(entities);
 
@@ -313,19 +442,6 @@ public class Section {
         }
     }
 
-
-    private void addNewerToEntities(Set<Entity> found) {
-        // HashSet keeps the older value, but we want the new only.
-        entities.removeAll(found);
-        entities.addAll(found);
-    }
-
-    private void addNewerToEntities(Entity found) {
-        // HashSet keeps the older value, but we want the new only.
-        entities.remove(found);
-        entities.add(found);
-    }
-
 }
 
 

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

@@ -0,0 +1,268 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import java.awt.Color;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.GregorianCalendar;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.stream.Collectors;
+
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.pdmodel.PDPageContentStream;
+import org.apache.pdfbox.pdmodel.common.PDRectangle;
+import org.apache.pdfbox.pdmodel.font.PDType1Font;
+import org.apache.pdfbox.pdmodel.graphics.color.PDColor;
+import org.apache.pdfbox.pdmodel.graphics.color.PDDeviceRGB;
+import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotation;
+import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotationText;
+import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotationTextMarkup;
+import org.springframework.stereotype.Service;
+
+import com.iqser.red.service.redaction.v1.model.CellRectangle;
+import com.iqser.red.service.redaction.v1.model.Comment;
+import com.iqser.red.service.redaction.v1.model.Rectangle;
+import com.iqser.red.service.redaction.v1.model.RedactionLog;
+import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import com.iqser.red.service.redaction.v1.model.SectionGrid;
+import com.iqser.red.service.redaction.v1.model.SectionRectangle;
+
+import lombok.RequiredArgsConstructor;
+
+@Service
+@RequiredArgsConstructor
+public class AnnotationService {
+
+    private final DictionaryService dictionaryService;
+
+
+    public void annotate(PDDocument document, RedactionLog redactionLog, SectionGrid sectionGrid) throws IOException {
+
+        Map<Integer, List<RedactionLogEntry>> redactionLogPerPage = convertRedactionLog(redactionLog);
+
+        for (int page = 1; page <= document.getNumberOfPages(); page++) {
+
+            PDPage pdPage = document.getPage(page - 1);
+
+            List<SectionRectangle> sectionRectangles = sectionGrid.getRectanglesPerPage().get(page);
+            if (sectionRectangles != null && !sectionRectangles.isEmpty()) {
+                drawSectionGrid(document, pdPage, sectionRectangles);
+            }
+
+            List<RedactionLogEntry> logEntries = redactionLogPerPage.get(page);
+            if (logEntries != null && !logEntries.isEmpty()) {
+                addAnnotations(logEntries, pdPage, page);
+            }
+        }
+    }
+
+
+    private void addAnnotations(List<RedactionLogEntry> logEntries, PDPage pdPage, int page) throws IOException {
+
+        List<PDAnnotation> annotations = pdPage.getAnnotations();
+
+        for (RedactionLogEntry entry : logEntries) {
+            annotations.addAll(createAnnotation(entry, page, pdPage.getMediaBox(), pdPage.getCropBox()));
+        }
+    }
+
+
+    private List<PDAnnotation> createAnnotation(RedactionLogEntry redactionLogEntry, int page, PDRectangle mediaBox,
+                                                PDRectangle cropBox) {
+
+        List<PDAnnotation> annotations = new ArrayList<>();
+
+        List<Rectangle> rectangles = redactionLogEntry.getPositions()
+                .stream()
+                .filter(pos -> pos.getPage() == page)
+                .collect(Collectors.toList());
+
+        if (rectangles.isEmpty()) {
+            return annotations;
+        }
+
+        PDAnnotationTextMarkup annotation = new PDAnnotationTextMarkup(PDAnnotationTextMarkup.SUB_TYPE_HIGHLIGHT);
+        annotation.constructAppearances();
+        PDRectangle pdRectangle = toPDRectangle(rectangles, mediaBox, cropBox);
+        annotation.setRectangle(pdRectangle);
+        annotation.setQuadPoints(toQuadPoints(rectangles, mediaBox, cropBox));
+        if (!redactionLogEntry.isHint()) {
+            annotation.setContents(createAnnotationContent(redactionLogEntry));
+        }
+        annotation.setTitlePopup(redactionLogEntry.getId());
+        annotation.setAnnotationName(redactionLogEntry.getId());
+        annotation.setColor(new PDColor(redactionLogEntry.getColor(), PDDeviceRGB.INSTANCE));
+        annotations.add(annotation);
+
+        if (redactionLogEntry.getComments() != null) {
+            for (Comment comment : redactionLogEntry.getComments()) {
+                PDAnnotationText txtAnnot = new PDAnnotationText();
+                txtAnnot.setAnnotationName(comment.getId());
+                txtAnnot.setInReplyTo(annotation); // Reference to highlight annotation
+                txtAnnot.setName(PDAnnotationText.NAME_COMMENT);
+                txtAnnot.setCreationDate(GregorianCalendar.from(comment.getDate().toZonedDateTime()));
+                txtAnnot.setTitlePopup(comment.getUser());
+                txtAnnot.setContents(comment.getText());
+                txtAnnot.setRectangle(pdRectangle);
+                annotations.add(txtAnnot);
+            }
+        }
+
+        return annotations;
+    }
+
+
+    private String createAnnotationContent(RedactionLogEntry redactionLogEntry) {
+
+        if (redactionLogEntry.isManual()) {
+            return "\nManual Redaction\n\nIn Section : \"" + redactionLogEntry.getSection() + "\"";
+        }
+        return "\nRule " + redactionLogEntry.getMatchedRule() + " matched\n\n" + redactionLogEntry.getReason() + "\n\nLegal basis:" + redactionLogEntry
+                .getLegalBasis() + "\n\nIn section: \"" + redactionLogEntry.getSection() + "\"";
+    }
+
+
+    private PDRectangle toPDRectangle(List<Rectangle> rectangles, PDRectangle mediaBox, PDRectangle cropBox) {
+
+        float lowerLeftX = Float.MAX_VALUE;
+        float upperRightX = 0;
+        float lowerLeftY = 0;
+        float upperRightY = Float.MAX_VALUE;
+
+        for (Rectangle rectangle : rectangles) {
+            if (rectangle.getTopLeft().getX() < lowerLeftX) {
+                lowerLeftX = rectangle.getTopLeft().getX();
+            }
+            if (rectangle.getTopLeft().getX() + rectangle.getWidth() > upperRightX) {
+                upperRightX = rectangle.getTopLeft().getX() + rectangle.getWidth();
+            }
+            if (rectangle.getTopLeft().getY() + rectangle.getHeight() > lowerLeftY) {
+                lowerLeftY = rectangle.getTopLeft().getY() + rectangle.getHeight();
+            }
+            if (rectangle.getTopLeft().getY() < upperRightY) {
+                upperRightY = rectangle.getTopLeft().getY();
+            }
+        }
+
+        var x1 = lowerLeftX + cropBox.getLowerLeftX() - mediaBox.getLowerLeftY();
+        var y1 = lowerLeftY + (mediaBox.getLowerLeftY() - cropBox.getLowerLeftY());
+        var x2 = upperRightX + cropBox.getLowerLeftX() - mediaBox.getLowerLeftY();
+        var y2 = upperRightY - (mediaBox.getLowerLeftY() - cropBox.getLowerLeftY());
+
+        PDRectangle annotationPosition = new PDRectangle();
+        annotationPosition.setLowerLeftX(x1);
+        annotationPosition.setLowerLeftY(y1);
+        annotationPosition.setUpperRightX(x2);
+        annotationPosition.setUpperRightY(y2);
+        return annotationPosition;
+    }
+
+
+    private float[] toQuadPoints(List<Rectangle> rectangles, PDRectangle mediaBox, PDRectangle cropBox) {
+
+        float[] quadPoints = new float[rectangles.size() * 8];
+        int i = 0;
+
+        for (Rectangle rectangle : rectangles) {
+            float[] quadPoint = toQuadPoint(rectangle, mediaBox, cropBox);
+            for (int j = 0; j <= 7; j++) {
+                quadPoints[i + j] = quadPoint[j];
+            }
+            i += 8;
+        }
+        return quadPoints;
+    }
+
+
+    private float[] toQuadPoint(Rectangle rectangle, PDRectangle mediaBox, PDRectangle cropBox) {
+
+        var x1 = rectangle.getTopLeft().getX() + cropBox.getLowerLeftX() - mediaBox.getLowerLeftY();
+        var y1 = rectangle.getTopLeft().getY() + (mediaBox.getLowerLeftY() - cropBox.getLowerLeftY());
+        var x2 = rectangle.getTopLeft()
+                .getX() + rectangle.getWidth() + cropBox.getLowerLeftX() - mediaBox.getLowerLeftY();
+        var y2 = rectangle.getTopLeft()
+                .getY() + rectangle.getHeight() - (mediaBox.getLowerLeftY() - cropBox.getLowerLeftY());
+
+        // quadPoints is array of x,y coordinates in Z-like order (top-left, top-right, bottom-left,bottom-right)
+        // of the area to be highlighted
+        return new float[]{x1, y1, x2, y2, x1, y2 - rectangle.getHeight(), x2, y1 - rectangle.getHeight()};
+    }
+
+
+    private void drawSectionGrid(PDDocument document, PDPage pdPage,
+                                 List<SectionRectangle> sectionRectangles) throws IOException {
+
+        PDPageContentStream contentStream = new PDPageContentStream(document, pdPage, PDPageContentStream.AppendMode.APPEND, true);
+        for (SectionRectangle sectionRectangle : sectionRectangles) {
+            drawSectionRectangle(contentStream, sectionRectangle);
+            drawSectionPartNumberText(contentStream, sectionRectangle);
+            drawTableCells(contentStream, sectionRectangle);
+        }
+        contentStream.close();
+    }
+
+
+    private void drawSectionRectangle(PDPageContentStream contentStream,
+                                      SectionRectangle sectionRectangle) throws IOException {
+
+        contentStream.setStrokingColor(Color.LIGHT_GRAY);
+        contentStream.setLineWidth(0.5f);
+        contentStream.addRect(sectionRectangle.getTopLeft().getX(), sectionRectangle.getTopLeft()
+                .getY(), sectionRectangle.getWidth(), sectionRectangle.getHeight());
+        contentStream.stroke();
+    }
+
+
+    private void drawSectionPartNumberText(PDPageContentStream contentStream,
+                                           SectionRectangle sectionRectangle) throws IOException {
+
+        contentStream.beginText();
+        contentStream.setNonStrokingColor(Color.DARK_GRAY);
+        contentStream.setFont(PDType1Font.TIMES_ROMAN, 8f);
+        if (sectionRectangle.getTableCells() == null) {
+            contentStream.newLineAtOffset(sectionRectangle.getTopLeft().getX(), sectionRectangle.getTopLeft()
+                    .getY() + sectionRectangle.getHeight());
+        } else {
+            contentStream.newLineAtOffset(sectionRectangle.getTopLeft().getX(), sectionRectangle.getTopLeft().getY());
+        }
+        contentStream.showText(sectionRectangle.getPart() + "/" + sectionRectangle.getNumberOfParts());
+        contentStream.endText();
+    }
+
+
+    private void drawTableCells(PDPageContentStream contentStream,
+                                SectionRectangle sectionRectangle) throws IOException {
+
+        if (sectionRectangle.getTableCells() != null) {
+            for (CellRectangle cell : sectionRectangle.getTableCells()) {
+                contentStream.setLineWidth(0.5f);
+                contentStream.setStrokingColor(Color.CYAN);
+                contentStream.addRect(cell.getTopLeft().getX(), cell.getTopLeft()
+                        .getY(), cell.getWidth(), cell.getHeight());
+                contentStream.stroke();
+            }
+        }
+    }
+
+
+    private Map<Integer, List<RedactionLogEntry>> convertRedactionLog(RedactionLog redactionLog) {
+
+        Map<Integer, List<RedactionLogEntry>> redactionLogPerPage = new HashMap<>();
+        if (redactionLog == null) {
+            return redactionLogPerPage;
+        }
+        for (RedactionLogEntry entry : redactionLog.getRedactionLogEntry()) {
+            int page = 0;
+            for (Rectangle position : entry.getPositions()) {
+                if (position.getPage() != page) {
+                    redactionLogPerPage.computeIfAbsent(position.getPage(), x -> new ArrayList<>()).add(entry);
+                    page = position.getPage();
+                }
+            }
+        }
+        return redactionLogPerPage;
+    }
+
+}

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

@@ -5,6 +5,7 @@ import com.iqser.red.service.configuration.v1.api.model.TypeResponse;
 import com.iqser.red.service.configuration.v1.api.model.TypeResult;
 import com.iqser.red.service.redaction.v1.server.client.DictionaryClient;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
+import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryIncrement;
 import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryModel;
 import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryRepresentation;
 import feign.FeignException;
@@ -48,6 +49,14 @@ public class DictionaryService {
     }
 
 
+    public DictionaryIncrement getDictionaryIncrements(String ruleSetId, long fromVersion){
+
+        updateDictionary(ruleSetId);
+
+        return new DictionaryIncrement(Set.of("Properties"), 1L);
+    }
+
+
     private void updateDictionaryEntry(String ruleSetId, long version) {
 
         try {

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

@@ -1,11 +1,11 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;
 
-import com.iqser.red.service.configuration.v1.api.model.RulesResponse;
-import com.iqser.red.service.redaction.v1.server.client.RulesClient;
-import com.iqser.red.service.redaction.v1.server.exception.RulesValidationException;
-import com.iqser.red.service.redaction.v1.server.redaction.model.Section;
-import lombok.Getter;
-import lombok.RequiredArgsConstructor;
+import java.io.ByteArrayInputStream;
+import java.io.InputStream;
+import java.nio.charset.StandardCharsets;
+import java.util.HashMap;
+import java.util.Map;
+
 import org.apache.commons.lang3.StringUtils;
 import org.kie.api.KieServices;
 import org.kie.api.builder.KieBuilder;
@@ -15,11 +15,12 @@ import org.kie.api.runtime.KieContainer;
 import org.kie.api.runtime.KieSession;
 import org.springframework.stereotype.Service;
 
-import java.io.ByteArrayInputStream;
-import java.io.InputStream;
-import java.nio.charset.StandardCharsets;
-import java.util.HashMap;
-import java.util.Map;
+import com.iqser.red.service.configuration.v1.api.model.RulesResponse;
+import com.iqser.red.service.redaction.v1.server.client.RulesClient;
+import com.iqser.red.service.redaction.v1.server.exception.RulesValidationException;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Section;
+
+import lombok.RequiredArgsConstructor;
 
 @Service
 @RequiredArgsConstructor
@@ -29,7 +30,11 @@ public class DroolsExecutionService {
 
     private Map<String, KieContainer> kieContainers = new HashMap<>();
 
+    private Map<String, Long> rulesVersionPerRuleSetId = new HashMap<>();
+
+
     public KieContainer getKieContainer(String ruleSetId) {
+
         KieContainer container = kieContainers.get(ruleSetId);
         if (container == null) {
             return createOrUpdateKieContainer(ruleSetId);
@@ -39,9 +44,6 @@ public class DroolsExecutionService {
     }
 
 
-    @Getter
-    private long rulesVersion = -1;
-
     public Section executeRules(KieContainer kieContainer, Section section) {
 
         KieSession kieSession = kieContainer.newKieSession();
@@ -54,17 +56,24 @@ public class DroolsExecutionService {
 
     }
 
+
     public KieContainer updateRules(String ruleSetId) {
 
         long version = rulesClient.getVersion(ruleSetId);
-        if (version > rulesVersion) {
-            rulesVersion = version;
+        Long rulesVersion = rulesVersionPerRuleSetId.get(ruleSetId);
+        if (rulesVersion == null) {
+            rulesVersion = -1L;
+        }
+
+        if (version > rulesVersion.longValue()) {
+            rulesVersionPerRuleSetId.put(ruleSetId, version);
             return createOrUpdateKieContainer(ruleSetId);
         }
         return getKieContainer(ruleSetId);
 
     }
 
+
     private KieContainer createOrUpdateKieContainer(String ruleSetId) {
 
         try {
@@ -92,7 +101,9 @@ public class DroolsExecutionService {
 
     }
 
+
     private KieModule getKieModule(String ruleSetId, String rules, KieServices kieServices) {
+
         KieFileSystem kieFileSystem = kieServices.newKieFileSystem();
         InputStream input = new ByteArrayInputStream(rules.getBytes(StandardCharsets.UTF_8));
         kieFileSystem.write("src/main/resources/drools/rules" + ruleSetId + ".drl", kieServices.getResources()
@@ -104,10 +115,22 @@ public class DroolsExecutionService {
 
 
     public void testRules(String rules) {
+
         KieServices kieServices = KieServices.Factory.get();
         KieModule kieModule = getKieModule("test-rules", rules, kieServices);
         var container = kieServices.newKieContainer(kieModule.getReleaseId());
         container.newKieSession();
         container.dispose();
     }
+
+
+    public long getRulesVersion(String ruleSetId) {
+
+        Long rulesVersion = rulesVersionPerRuleSetId.get(ruleSetId);
+        if (rulesVersion == null) {
+            return -1;
+        }
+        return rulesVersion.longValue();
+    }
+
 }

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

@@ -17,10 +17,15 @@ import org.springframework.stereotype.Service;
 
 import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
 import com.iqser.red.service.redaction.v1.model.ManualRedactions;
+import com.iqser.red.service.redaction.v1.model.Point;
 import com.iqser.red.service.redaction.v1.model.Rectangle;
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
+import com.iqser.red.service.redaction.v1.server.classification.model.Footer;
+import com.iqser.red.service.redaction.v1.server.classification.model.Header;
 import com.iqser.red.service.redaction.v1.server.classification.model.Paragraph;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+import com.iqser.red.service.redaction.v1.server.classification.model.UnclassifiedText;
+import com.iqser.red.service.redaction.v1.model.SectionArea;
 import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
 import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryModel;
@@ -29,6 +34,7 @@ import com.iqser.red.service.redaction.v1.server.redaction.model.EntityPositionS
 import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Section;
 import com.iqser.red.service.redaction.v1.server.redaction.model.SectionSearchableTextPair;
+import com.iqser.red.service.redaction.v1.model.SectionText;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.EntitySearchUtils;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
@@ -50,7 +56,7 @@ public class EntityRedactionService {
 
         dictionaryService.updateDictionary(ruleSetId);
         KieContainer container = droolsExecutionService.updateRules(ruleSetId);
-        long rulesVersion = droolsExecutionService.getRulesVersion();
+        long rulesVersion = droolsExecutionService.getRulesVersion(ruleSetId);
 
         Dictionary dictionary = dictionaryService.getDeepCopyDictionary(ruleSetId);
 
@@ -67,10 +73,7 @@ public class EntityRedactionService {
             });
 
             Set<Entity> foundByLocal = findEntities(classifiedDoc, container, manualRedactions, dictionary, true, hintsPerSectionNumber);
-            // HashSet keeps the older value, but we want the new only.
-            documentEntities.removeAll(foundByLocal);
-            documentEntities.addAll(foundByLocal);
-
+            EntitySearchUtils.addEntitiesWithHigherRank(documentEntities, foundByLocal, dictionary);
             EntitySearchUtils.removeEntitiesContainedInLarger(documentEntities);
         }
 
@@ -86,7 +89,7 @@ public class EntityRedactionService {
                         .computeIfAbsent(entry.getKey(), (x) -> new ArrayList<>())
                         .add(new Entity(entity.getWord(), entity.getType(), entity.isRedaction(), entity.getRedactionReason(), entry
                                 .getValue(), entity.getHeadline(), entity.getMatchedRule(), entity.getSectionNumber(), entity
-                                .getLegalBasis(), entity.isDictionaryEntry(), entity.getTextBefore(), entity.getTextAfter()));
+                                .getLegalBasis(), entity.isDictionaryEntry(), entity.getTextBefore(), entity.getTextAfter(), entity.getStart(), entity.getEnd()));
             }
         }
 
@@ -110,13 +113,29 @@ public class EntityRedactionService {
             List<Table> tables = paragraph.getTables();
             for (Table table : tables) {
                 if (table.getColCount() == 2) {
-                    sectionSearchableTextPairs.addAll(processTableAsOneText(table, manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+                    sectionSearchableTextPairs.addAll(processTableAsOneText(classifiedDoc, table, manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
                 } else {
-                    sectionSearchableTextPairs.addAll(processTablePerRow(table, manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+                    sectionSearchableTextPairs.addAll(processTablePerRow(classifiedDoc, table, manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
                 }
                 sectionNumber.incrementAndGet();
             }
-            sectionSearchableTextPairs.add(processText(paragraph, manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+            sectionSearchableTextPairs.add(processText(classifiedDoc, paragraph.getSearchableText(), paragraph.getTextBlocks(), paragraph
+                    .getHeadline(), manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+            sectionNumber.incrementAndGet();
+        }
+
+        for (Header header : classifiedDoc.getHeaders()) {
+            sectionSearchableTextPairs.add(processText(classifiedDoc, header.getSearchableText(), header.getTextBlocks(), "Header", manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+            sectionNumber.incrementAndGet();
+        }
+
+        for (Footer footer : classifiedDoc.getFooters()) {
+            sectionSearchableTextPairs.add(processText(classifiedDoc, footer.getSearchableText(), footer.getTextBlocks(), "Footer", manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+            sectionNumber.incrementAndGet();
+        }
+
+        for (UnclassifiedText unclassifiedText : classifiedDoc.getUnclassifiedTexts()) {
+            sectionSearchableTextPairs.add(processText(classifiedDoc, unclassifiedText.getSearchableText(), unclassifiedText.getTextBlocks(), "", manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
             sectionNumber.incrementAndGet();
         }
 
@@ -153,32 +172,44 @@ public class EntityRedactionService {
     }
 
 
-    private List<SectionSearchableTextPair> processTablePerRow(Table table, ManualRedactions manualRedactions,
+    private List<SectionSearchableTextPair> processTablePerRow(Document classifiedDoc, Table table, ManualRedactions manualRedactions,
                                                                AtomicInteger sectionNumber, Dictionary dictionary,
                                                                boolean local,
                                                                Map<Integer, Set<Entity>> hintsPerSectionNumber) {
 
         List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();
+
         for (List<Cell> row : table.getRows()) {
             SearchableText searchableRow = new SearchableText();
             Map<String, CellValue> tabularData = new HashMap<>();
             int start = 0;
             List<Integer> cellStarts = new ArrayList<>();
+            SectionText sectionText = new SectionText();
             for (Cell cell : row) {
+
                 if (CollectionUtils.isEmpty(cell.getTextBlocks())) {
                     continue;
                 }
+
+                SectionArea sectionArea = new SectionArea(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
+                        .getWidth(), (float) cell.getHeight(), cell.getTextBlocks().get(0).getSequences().get(0).getPage());
+                sectionText.getSectionAreas().add(sectionArea);
+
                 addSectionToManualRedactions(cell.getTextBlocks(), manualRedactions, table.getHeadline(), sectionNumber.intValue());
                 int cellStart = start;
-                cell.getHeaderCells().forEach(headerCell -> {
-                    StringBuilder headerBuilder = new StringBuilder();
-                    headerCell.getTextBlocks().forEach(textBlock -> headerBuilder.append(textBlock.getText()));
-                    String headerName = headerBuilder.toString()
-                            .replaceAll("\n", "")
-                            .replaceAll(" ", "")
-                            .replaceAll("-", "");
-                    tabularData.put(headerName, new CellValue(cell.getTextBlocks(), cellStart));
-                });
+
+                if (!cell.isHeaderCell()) {
+                    cell.getHeaderCells().forEach(headerCell -> {
+                        StringBuilder headerBuilder = new StringBuilder();
+                        headerCell.getTextBlocks().forEach(textBlock -> headerBuilder.append(textBlock.getText()));
+                        String headerName = headerBuilder.toString()
+                                .replaceAll("\n", "")
+                                .replaceAll(" ", "")
+                                .replaceAll("-", "");
+                        sectionArea.setHeader(headerName);
+                        tabularData.put(headerName, new CellValue(cell.getTextBlocks(), cellStart));
+                    });
+                }
 
                 for (TextBlock textBlock : cell.getTextBlocks()) {
                     // TODO avoid cell overlap merging.
@@ -186,6 +217,7 @@ public class EntityRedactionService {
                 }
                 cellStarts.add(cellStart);
                 start = start + cell.toString().trim().length() + 1;
+
             }
             Set<Entity> rowEntities = findEntities(searchableRow, table.getHeadline(), sectionNumber.intValue(), dictionary, local);
             surroundingWordsService.addSurroundingText(rowEntities, searchableRow, dictionary, cellStarts);
@@ -205,24 +237,41 @@ public class EntityRedactionService {
                     .dictionary(dictionary)
                     .build(), searchableRow));
 
+            if(!local) {
+                sectionText.setText(searchableRow.toString());
+                sectionText.setHeadline(table.getHeadline());
+                sectionText.setSectionNumber(sectionNumber.intValue());
+                sectionText.setTable(true);
+                classifiedDoc.getSectionText().add(sectionText);
+            }
+
             sectionNumber.incrementAndGet();
         }
+
         return sectionSearchableTextPairs;
     }
 
 
-    private List<SectionSearchableTextPair> processTableAsOneText(Table table, ManualRedactions manualRedactions,
+    private List<SectionSearchableTextPair> processTableAsOneText(Document classifiedDoc, Table table, ManualRedactions manualRedactions,
                                                                   AtomicInteger sectionNumber, Dictionary dictionary,
                                                                   boolean local,
                                                                   Map<Integer, Set<Entity>> hintsPerSectionNumber) {
 
         List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();
         SearchableText entireTableText = new SearchableText();
+        SectionText sectionText = new SectionText();
         for (List<Cell> row : table.getRows()) {
             for (Cell cell : row) {
                 if (CollectionUtils.isEmpty(cell.getTextBlocks())) {
                     continue;
                 }
+
+                if(!local) {
+                    SectionArea sectionArea = new SectionArea(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
+                            .getWidth(), (float) cell.getHeight(), cell.getTextBlocks().get(0).getSequences().get(0).getPage());
+                    sectionText.getSectionAreas().add(sectionArea);
+                }
+
                 for (TextBlock textBlock : cell.getTextBlocks()) {
                     entireTableText.addAll(textBlock.getSequences());
                 }
@@ -230,6 +279,7 @@ public class EntityRedactionService {
             }
         }
 
+
         Set<Entity> rowEntities = findEntities(entireTableText, table.getHeadline(), sectionNumber.intValue(), dictionary, local);
         surroundingWordsService.addSurroundingText(rowEntities, entireTableText, dictionary);
 
@@ -247,18 +297,40 @@ public class EntityRedactionService {
                 .dictionary(dictionary)
                 .build(), entireTableText));
 
+        if(!local) {
+            sectionText.setText(entireTableText.toString());
+            sectionText.setHeadline(table.getHeadline());
+            sectionText.setSectionNumber(sectionNumber.intValue());
+            sectionText.setTable(true);
+            classifiedDoc.getSectionText().add(sectionText);
+        }
+
         return sectionSearchableTextPairs;
     }
 
 
-    private SectionSearchableTextPair processText(Paragraph paragraph, ManualRedactions manualRedactions,
+    private SectionSearchableTextPair processText(Document classifiedDoc, SearchableText searchableText, List<TextBlock> paragraphTextBlocks,
+                                                  String headline, ManualRedactions manualRedactions,
                                                   AtomicInteger sectionNumber, Dictionary dictionary, boolean local,
                                                   Map<Integer, Set<Entity>> hintsPerSectionNumber) {
 
-        SearchableText searchableText = paragraph.getSearchableText();
-        addSectionToManualRedactions(paragraph.getTextBlocks(), manualRedactions, paragraph.getHeadline(), sectionNumber
-                .intValue());
-        Set<Entity> entities = findEntities(searchableText, paragraph.getHeadline(), sectionNumber.intValue(), dictionary, local);
+        if(!local) {
+            SectionText sectionText = new SectionText();
+            for (TextBlock paragraphTextBlock : paragraphTextBlocks) {
+                SectionArea sectionArea = new SectionArea(new Point(paragraphTextBlock.getMinX(), paragraphTextBlock.getMinY()), paragraphTextBlock
+                        .getWidth(), paragraphTextBlock.getHeight(), paragraphTextBlock.getPage());
+                sectionText.getSectionAreas().add(sectionArea);
+            }
+
+            sectionText.setText(searchableText.toString());
+            sectionText.setHeadline(headline);
+            sectionText.setSectionNumber(sectionNumber.intValue());
+            sectionText.setTable(false);
+            classifiedDoc.getSectionText().add(sectionText);
+        }
+
+        addSectionToManualRedactions(paragraphTextBlocks, manualRedactions, headline, sectionNumber.intValue());
+        Set<Entity> entities = findEntities(searchableText, headline, sectionNumber.intValue(), dictionary, local);
         surroundingWordsService.addSurroundingText(entities, searchableText, dictionary);
 
         return new SectionSearchableTextPair(Section.builder()
@@ -269,7 +341,7 @@ public class EntityRedactionService {
                         .collect(Collectors.toSet()) : entities)
                 .text(searchableText.getAsStringWithLinebreaks())
                 .searchText(searchableText.toString())
-                .headline(paragraph.getHeadline())
+                .headline(headline)
                 .sectionNumber(sectionNumber.intValue())
                 .searchableText(searchableText)
                 .dictionary(dictionary)
@@ -277,7 +349,7 @@ public class EntityRedactionService {
     }
 
 
-    private Set<Entity> findEntities(SearchableText searchableText, String headline, int sectionNumber,
+    public Set<Entity> findEntities(SearchableText searchableText, String headline, int sectionNumber,
                                      Dictionary dictionary, boolean local) {
 
         Set<Entity> found = new HashSet<>();

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

@@ -0,0 +1,281 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import java.awt.geom.Rectangle2D;
+import java.io.ByteArrayInputStream;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
+
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.pdmodel.common.PDRectangle;
+import org.kie.api.runtime.KieContainer;
+import org.springframework.stereotype.Service;
+import org.springframework.web.bind.annotation.RequestBody;
+
+import com.iqser.red.service.redaction.v1.model.Comment;
+import com.iqser.red.service.redaction.v1.model.IdRemoval;
+import com.iqser.red.service.redaction.v1.model.ManualForceRedact;
+import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
+import com.iqser.red.service.redaction.v1.model.ManualRedactions;
+import com.iqser.red.service.redaction.v1.model.ReanalyzeResult;
+import com.iqser.red.service.redaction.v1.model.Rectangle;
+import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import com.iqser.red.service.redaction.v1.model.RenalyzeRequest;
+import com.iqser.red.service.redaction.v1.model.SectionArea;
+import com.iqser.red.service.redaction.v1.model.SectionText;
+import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
+import com.iqser.red.service.redaction.v1.server.parsing.PDFAreaTextStripper;
+import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
+import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
+import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryIncrement;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
+import com.iqser.red.service.redaction.v1.server.redaction.model.EntityPositionSequence;
+import com.iqser.red.service.redaction.v1.server.redaction.model.ReanalysisSection;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Section;
+import com.iqser.red.service.redaction.v1.server.redaction.model.SectionSearchableTextPair;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.EntitySearchUtils;
+import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
+
+import lombok.RequiredArgsConstructor;
+
+@Service
+@RequiredArgsConstructor
+public class ReanalyzeService {
+
+    private final DictionaryService dictionaryService;
+    private final DroolsExecutionService droolsExecutionService;
+    private final SurroundingWordsService surroundingWordsService;
+    private final EntityRedactionService entityRedactionService;
+    private final RedactionLogCreatorService redactionLogCreatorService;
+
+
+    public ReanalyzeResult reanalyze(@RequestBody RenalyzeRequest renalyzeRequest) {
+
+        DictionaryIncrement dictionaryIncrement = dictionaryService.getDictionaryIncrements(renalyzeRequest.getRuleSetId(), renalyzeRequest
+                .getRedactionLog()
+                .getDictionaryVersion());
+
+        Set<String> manualForceAndRemoveIds = getForceAndRemoveIds(renalyzeRequest.getManualRedactions());
+        Map<String, List<Comment>> comments = null;
+        Set<ManualRedactionEntry> manualAdds = null;
+
+        if (renalyzeRequest.getManualRedactions() != null) {
+            // TODO comments will be removed from redactionLog, so we ignore this first.
+            comments = renalyzeRequest.getManualRedactions().getComments();
+            manualAdds = renalyzeRequest.getManualRedactions().getEntriesToAdd();
+        }
+
+        Set<Integer> sectionsToReanaylse = new HashSet<>();
+        for (RedactionLogEntry entry : renalyzeRequest.getRedactionLog().getRedactionLogEntry()) {
+            if (entry.isManual() || manualForceAndRemoveIds.contains(entry.getId())) {
+                sectionsToReanaylse.add(entry.getSectionNumber());
+            }
+        }
+
+        for (SectionText sectionText : renalyzeRequest.getText().getSectionTexts()) {
+            Set<Entity> entities = EntitySearchUtils.find(sectionText.getText(), dictionaryIncrement.getValues(), "find", sectionText
+                    .getHeadline(), sectionText.getSectionNumber(), false);
+            if (!entities.isEmpty()) {
+                sectionsToReanaylse.add(sectionText.getSectionNumber());
+            }
+
+            if (manualAdds != null) {
+                for (SectionArea sectionArea : sectionText.getSectionAreas()) {
+                    for (ManualRedactionEntry manualAdd : manualAdds) {
+                        for (Rectangle manualPosition : manualAdd.getPositions()) {
+                            if (sectionArea.contains(manualPosition)) {
+                                manualAdd.setSection(sectionText.getHeadline());
+                                manualAdd.setSectionNumber(sectionText.getSectionNumber());
+                            }
+                        }
+                    }
+                }
+            }
+        }
+
+        if (sectionsToReanaylse.isEmpty() && (manualAdds == null || manualAdds.isEmpty())) {
+            renalyzeRequest.getRedactionLog().setDictionaryVersion(dictionaryIncrement.getDictionaryVersion());
+            return ReanalyzeResult.builder().redactionLog(renalyzeRequest.getRedactionLog()).build();
+        }
+
+        try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(renalyzeRequest.getDocument()))) {
+
+            List<ReanalysisSection> reanalysisSections = new ArrayList<>();
+            for (SectionText sectionText : renalyzeRequest.getText().getSectionTexts()) {
+
+                if (!sectionsToReanaylse.contains(sectionText.getSectionNumber())) {
+                    continue;
+                }
+
+                ReanalysisSection reanalysisSection = new ReanalysisSection();
+                reanalysisSection.setHeadline(sectionText.getHeadline());
+                reanalysisSection.setSectionNumber(sectionText.getSectionNumber());
+                List<TextBlock> textBlocks = new ArrayList<>();
+
+                Map<Integer, List<SectionArea>> sectionAreasPerPage = new HashMap<>();
+                for (SectionArea sectionArea : sectionText.getSectionAreas()) {
+                    sectionAreasPerPage.computeIfAbsent(sectionArea.getPage(), (x) -> new ArrayList<>())
+                            .add(sectionArea);
+                }
+
+                Map<String, CellValue> tabularData = new HashMap<>();
+                List<Integer> cellStarts = new ArrayList<>();
+                for (Integer page : sectionAreasPerPage.keySet()) {
+                    List<SectionArea> areasOnPage = sectionAreasPerPage.get(page);
+
+                    PDPage pdPage = pdDocument.getPage(page - 1);
+                    PDRectangle cropBox = pdPage.getCropBox();
+                    PDFAreaTextStripper textStripper = new PDFAreaTextStripper();
+                    textStripper.setPageNumber(page);
+
+                    int cellStart = 0;
+                    for (SectionArea sectionArea : areasOnPage) {
+
+                        Rectangle2D rect = null;
+                        if (pdPage.getRotation() == 90) {
+                            rect = new Rectangle2D.Float(sectionArea.getTopLeft().getY(), sectionArea.getTopLeft()
+                                    .getX(), sectionArea.getHeight(), sectionArea.getWidth() + 0.001f);
+                        } else {
+                            rect = new Rectangle2D.Float(sectionArea.getTopLeft().getX(), -sectionArea.getTopLeft()
+                                    .getY() + cropBox.getUpperRightY() - sectionArea.getHeight(), sectionArea.getWidth(), sectionArea
+                                    .getHeight() + 0.001f);
+                        }
+
+                        textStripper.addRegion(String.valueOf(1), rect);
+                        textStripper.extractRegions(pdPage);
+                        textStripper.getTextForRegion(String.valueOf(1));
+                        List<TextPositionSequence> positions = textStripper.getTextPositionSequences();
+
+                        TextBlock textBlock = new TextBlock(sectionArea.getTopLeft().getX(), sectionArea.getTopLeft()
+                                .getX() + sectionArea.getWidth(), sectionArea.getTopLeft()
+                                .getY(), sectionArea.getTopLeft().getY() + sectionArea.getHeight(), positions, 0);
+
+                        if (sectionText.isTable()) {
+                            Cell cell = new Cell();
+                            cell.addTextBlock(textBlock);
+                            tabularData.put(sectionArea.getHeader(), new CellValue(cell.getTextBlocks(), cellStart));
+                            cellStarts.add(cellStart);
+                            cellStart = cellStart + cell.toString().trim().length() + 1;
+                        }
+
+                        textBlocks.add(textBlock);
+                        textStripper.clearPositions();
+                    }
+
+                }
+                reanalysisSection.setTextBlocks(textBlocks);
+                reanalysisSection.setTabularData(tabularData);
+                reanalysisSections.add(reanalysisSection);
+                if (sectionText.isTable()) {
+                    reanalysisSection.setCellStarts(cellStarts);
+                }
+            }
+
+            //--
+
+            KieContainer kieContainer = droolsExecutionService.updateRules(renalyzeRequest.getRuleSetId());
+
+            Dictionary dictionary = dictionaryService.getDeepCopyDictionary(renalyzeRequest.getRuleSetId());
+
+            List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();
+            for (ReanalysisSection reanalysisSection : reanalysisSections) {
+
+                Set<Entity> entities = entityRedactionService.findEntities(reanalysisSection.getSearchableText(), reanalysisSection
+                        .getHeadline(), reanalysisSection.getSectionNumber(), dictionary, false);
+                if (reanalysisSection.getCellStarts() != null) {
+                    surroundingWordsService.addSurroundingText(entities, reanalysisSection.getSearchableText(), dictionary, reanalysisSection
+                            .getCellStarts());
+                } else {
+                    surroundingWordsService.addSurroundingText(entities, reanalysisSection.getSearchableText(), dictionary);
+                }
+
+                sectionSearchableTextPairs.add(new SectionSearchableTextPair(Section.builder()
+                        .isLocal(false)
+                        .dictionaryTypes(dictionary.getTypes())
+                        .entities(entities)
+                        .text(reanalysisSection.getSearchableText().getAsStringWithLinebreaks())
+                        .searchText(reanalysisSection.getSearchableText().toString())
+                        .headline(reanalysisSection.getHeadline())
+                        .sectionNumber(reanalysisSection.getSectionNumber())
+                        .tabularData(reanalysisSection.getTabularData())
+                        .searchableText(reanalysisSection.getSearchableText())
+                        .dictionary(dictionary)
+                        .build(), reanalysisSection.getSearchableText()));
+            }
+
+            Set<Entity> entities = new HashSet<>();
+            sectionSearchableTextPairs.forEach(sectionSearchableTextPair -> {
+                Section analysedRowSection = droolsExecutionService.executeRules(kieContainer, sectionSearchableTextPair
+                        .getSection());
+                entities.addAll(analysedRowSection.getEntities());
+            });
+
+            Map<Integer, List<Entity>> entitiesPerPage = new HashMap<>();
+            for (Entity entity : entities) {
+                Map<Integer, List<EntityPositionSequence>> sequenceOnPage = new HashMap<>();
+                for (EntityPositionSequence entityPositionSequence : entity.getPositionSequences()) {
+                    sequenceOnPage.computeIfAbsent(entityPositionSequence.getPageNumber(), (x) -> new ArrayList<>())
+                            .add(entityPositionSequence);
+                }
+
+                for (Map.Entry<Integer, List<EntityPositionSequence>> entry : sequenceOnPage.entrySet()) {
+                    entitiesPerPage.computeIfAbsent(entry.getKey(), (x) -> new ArrayList<>())
+                            .add(new Entity(entity.getWord(), entity.getType(), entity.isRedaction(), entity.getRedactionReason(), entry
+                                    .getValue(), entity.getHeadline(), entity.getMatchedRule(), entity.getSectionNumber(), entity
+                                    .getLegalBasis(), entity.isDictionaryEntry(), entity.getTextBefore(), entity.getTextAfter(), entity
+                                    .getStart(), entity.getEnd()));
+                }
+            }
+
+            List<RedactionLogEntry> newRedactionLogEntries = new ArrayList<>();
+            for (int page = 1; page <= pdDocument.getNumberOfPages(); page++) {
+                if (entitiesPerPage.get(page) != null) {
+                    newRedactionLogEntries.addAll(redactionLogCreatorService.addEntries(entitiesPerPage, renalyzeRequest
+                            .getManualRedactions(), page, renalyzeRequest.getRuleSetId()));
+                }
+                newRedactionLogEntries.addAll(redactionLogCreatorService.addManualAddEntries(manualAdds, comments, page, renalyzeRequest
+                        .getRuleSetId()));
+            }
+
+            Iterator<RedactionLogEntry> itty = renalyzeRequest.getRedactionLog().getRedactionLogEntry().iterator();
+            while (itty.hasNext()) {
+                RedactionLogEntry entry = itty.next();
+                if (sectionsToReanaylse.contains(entry.getSectionNumber()) || entry.getSectionNumber() == 0) {
+                    itty.remove();
+                }
+            }
+
+            renalyzeRequest.getRedactionLog().getRedactionLogEntry().addAll(newRedactionLogEntries);
+            renalyzeRequest.getRedactionLog().setDictionaryVersion(dictionaryIncrement.getDictionaryVersion());
+
+            return ReanalyzeResult.builder().redactionLog(renalyzeRequest.getRedactionLog()).build();
+
+        } catch (Exception e) {
+            throw new RedactionException(e);
+        }
+
+    }
+
+
+    private Set<String> getForceAndRemoveIds(ManualRedactions manualRedactions) {
+
+        if (manualRedactions == null) {
+            return new HashSet<>();
+        }
+
+        return Stream.concat(manualRedactions.getIdsToRemove()
+                .stream()
+                .map(IdRemoval::getId), manualRedactions.getForceRedacts().stream().map(ManualForceRedact::getId))
+                .collect(Collectors.toSet());
+    }
+
+}

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

@@ -1,8 +1,21 @@
-package com.iqser.red.service.redaction.v1.server.visualization.service;
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import java.awt.geom.Rectangle2D;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import org.apache.commons.collections4.CollectionUtils;
+import org.apache.pdfbox.text.TextPosition;
+import org.springframework.stereotype.Service;
 
 import com.iqser.red.service.redaction.v1.model.CellRectangle;
 import com.iqser.red.service.redaction.v1.model.Comment;
 import com.iqser.red.service.redaction.v1.model.IdRemoval;
+import com.iqser.red.service.redaction.v1.model.ManualForceRedact;
 import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
 import com.iqser.red.service.redaction.v1.model.ManualRedactionType;
 import com.iqser.red.service.redaction.v1.model.ManualRedactions;
@@ -17,59 +30,62 @@ import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
 import com.iqser.red.service.redaction.v1.server.redaction.model.EntityPositionSequence;
-import com.iqser.red.service.redaction.v1.server.redaction.service.DictionaryService;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.IdBuilder;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
-import lombok.RequiredArgsConstructor;
-import org.apache.commons.collections4.CollectionUtils;
-import org.apache.pdfbox.pdmodel.PDDocument;
-import org.apache.pdfbox.pdmodel.PDPage;
-import org.apache.pdfbox.pdmodel.PDPageContentStream;
-import org.apache.pdfbox.pdmodel.common.PDRectangle;
-import org.apache.pdfbox.pdmodel.font.PDType1Font;
-import org.apache.pdfbox.pdmodel.graphics.color.PDColor;
-import org.apache.pdfbox.pdmodel.graphics.color.PDDeviceRGB;
-import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotation;
-import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotationText;
-import org.apache.pdfbox.pdmodel.interactive.annotation.PDAnnotationTextMarkup;
-import org.apache.pdfbox.text.TextPosition;
-import org.springframework.stereotype.Service;
 
-import java.awt.Color;
-import java.io.IOException;
-import java.util.ArrayList;
-import java.util.GregorianCalendar;
-import java.util.HashSet;
-import java.util.List;
-import java.util.Set;
-import java.util.stream.Collectors;
+import lombok.RequiredArgsConstructor;
 
 @Service
 @RequiredArgsConstructor
-public class AnnotationHighlightService {
+public class RedactionLogCreatorService {
+
+    private static final String IMAGE = "image";
 
     private final DictionaryService dictionaryService;
 
 
-    public void highlight(PDDocument document, Document classifiedDoc, boolean flatRedaction,
-                          ManualRedactions manualRedactions, String ruleSetId) throws IOException {
+    public void createRedactionLog(Document classifiedDoc, int numberOfPages, ManualRedactions manualRedactions,
+                                   String ruleSetId) {
 
         Set<Integer> manualRedactionPages = getManualRedactionPages(manualRedactions);
 
-        for (int page = 1; page <= document.getNumberOfPages(); page++) {
+        for (int page = 1; page <= numberOfPages; page++) {
 
-            PDPage pdPage = document.getPage(page - 1);
-
-            drawSectionFrames(document, classifiedDoc, flatRedaction, pdPage, page);
+            addSectionGrid(classifiedDoc, page);
 
             if (classifiedDoc.getEntities().get(page) != null) {
-                addAnnotations(pdPage, classifiedDoc, flatRedaction, manualRedactions, page, ruleSetId);
+                classifiedDoc.getRedactionLogEntities().addAll(addEntries(classifiedDoc.getEntities(), manualRedactions, page, ruleSetId));
             }
 
             if (manualRedactionPages.contains(page)) {
-                addManualAnnotations(pdPage, classifiedDoc, manualRedactions, page, ruleSetId);
+                classifiedDoc.getRedactionLogEntities().addAll(addManualAddEntries(manualRedactions.getEntriesToAdd(), manualRedactions.getComments(), page, ruleSetId));
             }
+
+            if (!classifiedDoc.getPages().get(page - 1).getImageBounds().isEmpty()) {
+                addImageEntries(classifiedDoc, page, ruleSetId);
+            }
+        }
+    }
+
+
+    private void addImageEntries(Document classifiedDoc, int pageNumber, String ruleSetId) {
+
+        for (Rectangle2D imageBounds : classifiedDoc.getPages().get(pageNumber - 1).getImageBounds()) {
+            RedactionLogEntry redactionLogEntry = RedactionLogEntry.builder()
+                    .id(IdBuilder.buildId(imageBounds, pageNumber))
+                    .color(getColor(IMAGE, ruleSetId))
+                    .type(IMAGE)
+                    .redacted(false)
+                    .isHint(true)
+                    .manual(false)
+                    .isDictionaryEntry(false)
+                    .isRecommendation(false)
+                    .positions(List.of(new Rectangle(new Point((float) imageBounds.getX(), (float) imageBounds.getY()), (float) imageBounds
+                            .getWidth(), (float) imageBounds.getHeight(), pageNumber)))
+                    .build();
+            classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
         }
     }
 
@@ -91,22 +107,16 @@ public class AnnotationHighlightService {
     }
 
 
-    private void addAnnotations(PDPage pdPage, Document classifiedDoc, boolean flatRedaction,
-                                ManualRedactions manualRedactions, int page, String ruleSetId) throws IOException {
+    public List<RedactionLogEntry> addEntries(Map<Integer, List<Entity>> entities, ManualRedactions manualRedactions, int page, String ruleSetId) {
 
-        List<PDAnnotation> annotations = pdPage.getAnnotations();
+        List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
 
         // Duplicates can exist due table extraction colums over multiple rows.
         Set<String> processedIds = new HashSet<>();
 
         entityLoop:
-        for (Entity entity : classifiedDoc.getEntities().get(page)) {
+        for (Entity entity : entities.get(page)) {
 
-            if (flatRedaction && !isRedactionType(entity, ruleSetId)) {
-                continue;
-            }
-
-            boolean requestedToRemove = false;
             List<Comment> comments = null;
 
             for (EntityPositionSequence entityPositionSequence : entity.getPositionSequences()) {
@@ -130,10 +140,11 @@ public class AnnotationHighlightService {
                                 redactionLogEntry.setRedacted(false);
                                 redactionLogEntry.setStatus(Status.APPROVED);
                                 manualOverrideReason = entity.getRedactionReason() + ", removed by manual override";
+                                redactionLogEntry.setColor(getColor(entity, ruleSetId, false));
                             } else if (manualRemoval.getStatus().equals(Status.REQUESTED)) {
-                                requestedToRemove = true;
                                 manualOverrideReason = entity.getRedactionReason() + ", requested to remove";
                                 redactionLogEntry.setStatus(Status.REQUESTED);
+                                redactionLogEntry.setColor(getColor(entity, ruleSetId, true));
                             } else {
                                 redactionLogEntry.setStatus(Status.DECLINED);
                             }
@@ -144,7 +155,34 @@ public class AnnotationHighlightService {
                             redactionLogEntry.setManualRedactionType(ManualRedactionType.REMOVE);
                         }
                     }
+                }
 
+                if (manualRedactions != null && !manualRedactions.getForceRedacts().isEmpty()) {
+                    for (ManualForceRedact manualForceRedact : manualRedactions.getForceRedacts()) {
+                        if (manualForceRedact.getId().equals(entityPositionSequence.getId())) {
+                            String manualOverrideReason = null;
+                            if (manualForceRedact.getStatus().equals(Status.APPROVED)) {
+                                entity.setRedaction(true);
+                                redactionLogEntry.setRedacted(true);
+                                redactionLogEntry.setStatus(Status.APPROVED);
+                                redactionLogEntry.setColor(getColor(entity, ruleSetId, false));
+                                manualOverrideReason = entity.getRedactionReason() + ", forced by manual override";
+                                redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
+                            } else if (manualForceRedact.getStatus().equals(Status.REQUESTED)) {
+                                manualOverrideReason = entity.getRedactionReason() + ", requested to force redact";
+                                redactionLogEntry.setStatus(Status.REQUESTED);
+                                redactionLogEntry.setColor(getColor(entity, ruleSetId, true));
+                                redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
+                            } else {
+                                redactionLogEntry.setStatus(Status.DECLINED);
+                            }
+
+                            entity.setRedactionReason(manualOverrideReason != null ? manualOverrideReason : entity.getRedactionReason());
+                            redactionLogEntry.setReason(manualOverrideReason);
+                            redactionLogEntry.setManual(true);
+                            redactionLogEntry.setManualRedactionType(ManualRedactionType.FORCE_REDACT);
+                        }
+                    }
                 }
 
                 if (CollectionUtils.isNotEmpty(entityPositionSequence.getSequences())) {
@@ -157,21 +195,21 @@ public class AnnotationHighlightService {
                         comments = manualRedactions.getComments().get(entityPositionSequence.getId());
                     }
 
+                    redactionLogEntry.setComments(comments);
                     redactionLogEntry.getPositions().addAll(rectanglesPerLine);
 
-                    annotations.addAll(createAnnotation(rectanglesPerLine, entityPositionSequence.getId(), createAnnotationContent(entity), getColor(entity, ruleSetId, requestedToRemove), comments, !isHint(entity, ruleSetId)));
                 }
 
                 redactionLogEntry.setId(entityPositionSequence.getId());
 
                 // FIXME ids should never be null. Figure out why this happens.
                 if (redactionLogEntry.getId() != null) {
-                    classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
+                    redactionLogEntities.add(redactionLogEntry);
                 }
             }
-
         }
 
+        return redactionLogEntities;
     }
 
 
@@ -191,7 +229,7 @@ public class AnnotationHighlightService {
                     startIndex = i;
                 }
             }
-            if (startIndex != textPositions.size() - 1) {
+            if (startIndex != textPositions.size()) {
                 rectangles.add(new TextPositionSequence(textPositions.subList(startIndex, textPositions.size()), page).getRectangle());
             }
         }
@@ -200,16 +238,16 @@ public class AnnotationHighlightService {
     }
 
 
-    private void addManualAnnotations(PDPage pdPage, Document classifiedDoc, ManualRedactions manualRedactions,
-                                      int page, String ruleSetId) throws IOException {
+    public List<RedactionLogEntry> addManualAddEntries(Set<ManualRedactionEntry> manualAdds,  Map<String, List<Comment>> comments, int page,
+                                  String ruleSetId) {
 
-        if (manualRedactions == null) {
-            return;
+        List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
+
+        if (manualAdds == null) {
+            return redactionLogEntities;
         }
 
-        List<PDAnnotation> annotations = pdPage.getAnnotations();
-
-        for (ManualRedactionEntry manualRedactionEntry : manualRedactions.getEntriesToAdd()) {
+        for (ManualRedactionEntry manualRedactionEntry : manualAdds) {
 
             String id = manualRedactionEntry.getId();
 
@@ -223,12 +261,13 @@ public class AnnotationHighlightService {
                 }
             }
 
+            redactionLogEntry.setComments(comments.get(id));
             if (!rectanglesOnPage.isEmpty() && !approvedAndShouldBeInDictionary(manualRedactionEntry)) {
-                annotations.addAll(createAnnotation(rectanglesOnPage, id, createAnnotationContent(manualRedactionEntry), getColorForManualAdd(manualRedactionEntry
-                        .getType(), ruleSetId, manualRedactionEntry.getStatus()), manualRedactions.getComments().get(id), true));
-                classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
+                redactionLogEntities.add(redactionLogEntry);
             }
         }
+
+        return redactionLogEntities;
     }
 
 
@@ -238,11 +277,12 @@ public class AnnotationHighlightService {
     }
 
 
-    private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry, String id, String ruleSetId) {
+    private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry, String id,
+                                                      String ruleSetId) {
 
         return RedactionLogEntry.builder()
                 .id(id)
-                .color(getColor(manualRedactionEntry.getType(), ruleSetId))
+                .color(getColorForManualAdd(manualRedactionEntry.getType(), ruleSetId, manualRedactionEntry.getStatus()))
                 .reason(manualRedactionEntry.getReason())
                 .legalBasis(manualRedactionEntry.getLegalBasis())
                 .value(manualRedactionEntry.getValue())
@@ -276,126 +316,12 @@ public class AnnotationHighlightService {
                 .isDictionaryEntry(entity.isDictionaryEntry())
                 .textAfter(entity.getTextAfter())
                 .textBefore(entity.getTextBefore())
+                .startOffset(entity.getStart())
+                .endOffset(entity.getEnd())
                 .build();
     }
 
 
-    private List<PDAnnotation> createAnnotation(List<Rectangle> rectangles, String id, String content, float[] color,
-                                                List<Comment> comments, boolean popup) {
-
-        List<PDAnnotation> annotations = new ArrayList<>();
-
-        PDAnnotationTextMarkup annotation = new PDAnnotationTextMarkup(PDAnnotationTextMarkup.SUB_TYPE_HIGHLIGHT);
-        annotation.constructAppearances();
-        PDRectangle pdRectangle = toPDRectangle(rectangles);
-        annotation.setRectangle(pdRectangle);
-        annotation.setQuadPoints(toQuadPoints(rectangles));
-        if (popup) {
-            annotation.setContents(content);
-        }
-        annotation.setTitlePopup(id);
-        annotation.setAnnotationName(id);
-        annotation.setColor(new PDColor(color, PDDeviceRGB.INSTANCE));
-        annotations.add(annotation);
-
-        if (comments != null) {
-            for (Comment comment : comments) {
-                PDAnnotationText txtAnnot = new PDAnnotationText();
-                txtAnnot.setAnnotationName(comment.getId());
-                txtAnnot.setInReplyTo(annotation); // Reference to highlight annotation
-                txtAnnot.setName(PDAnnotationText.NAME_COMMENT);
-                txtAnnot.setCreationDate(GregorianCalendar.from(comment.getDate().toZonedDateTime()));
-                txtAnnot.setTitlePopup(comment.getUser());
-                txtAnnot.setContents(comment.getText());
-                txtAnnot.setRectangle(pdRectangle);
-                annotations.add(txtAnnot);
-            }
-        }
-
-        return annotations;
-    }
-
-
-    private String createAnnotationContent(Entity entity) {
-
-        return "\nRule " + entity.getMatchedRule() + " matched\n\n" + entity.getRedactionReason() + "\n\nLegal basis:" + entity
-                .getLegalBasis() + "\n\nIn section: \"" + entity.getHeadline() + "\"";
-    }
-
-
-    private String createAnnotationContent(ManualRedactionEntry entry) {
-
-        return "\nManual Redaction\n\nIn Section : \"" + entry.getSection() + "\"";
-    }
-
-
-    private PDRectangle toPDRectangle(List<Rectangle> rectangles) {
-
-        float lowerLeftX = Float.MAX_VALUE;
-        float upperRightX = 0;
-        float lowerLeftY = 0;
-        float upperRightY = Float.MAX_VALUE;
-
-        for (Rectangle rectangle : rectangles) {
-            if (rectangle.getTopLeft().getX() < lowerLeftX) {
-                lowerLeftX = rectangle.getTopLeft().getX();
-            }
-            if (rectangle.getTopLeft().getX() + rectangle.getWidth() > upperRightX) {
-                upperRightX = rectangle.getTopLeft().getX() + rectangle.getWidth();
-            }
-            if (rectangle.getTopLeft().getY() + rectangle.getHeight() > lowerLeftY) {
-                lowerLeftY = rectangle.getTopLeft().getY() + rectangle.getHeight();
-            }
-            if (rectangle.getTopLeft().getY() < upperRightY) {
-                upperRightY = rectangle.getTopLeft().getY();
-            }
-        }
-
-        PDRectangle annotationPosition = new PDRectangle();
-        annotationPosition.setLowerLeftX(lowerLeftX);
-        annotationPosition.setLowerLeftY(lowerLeftY);
-        annotationPosition.setUpperRightX(upperRightX);
-        annotationPosition.setUpperRightY(upperRightY);
-        return annotationPosition;
-    }
-
-
-    private float[] toQuadPoints(List<Rectangle> rectangles) {
-
-        float[] quadPoints = new float[rectangles.size() * 8];
-        int i = 0;
-
-        for (Rectangle rectangle : rectangles) {
-            float[] quadPoint = toQuadPoint(rectangle);
-            for (int j = 0; j <= 7; j++) {
-                quadPoints[i + j] = quadPoint[j];
-            }
-            i += 8;
-        }
-        return quadPoints;
-    }
-
-
-    private float[] toQuadPoint(Rectangle rectangle) {
-
-        // quadPoints is array of x,y coordinates in Z-like order (top-left, top-right, bottom-left,bottom-right)
-        // of the area to be highlighted
-        return new float[]{rectangle.getTopLeft().getX(), rectangle.getTopLeft().getY(), rectangle.getTopLeft()
-                .getX() + rectangle.getWidth(), rectangle.getTopLeft().getY(), rectangle.getTopLeft().getX(), rectangle.getTopLeft()
-                .getY() + rectangle.getHeight(), rectangle.getTopLeft()
-                .getX() + rectangle.getWidth(), rectangle.getTopLeft().getY() + rectangle.getHeight()};
-    }
-
-
-    private boolean isRedactionType(Entity entity, String ruleSetId) {
-
-        if (!entity.isRedaction()) {
-            return false;
-        }
-        return !isHint(entity, ruleSetId);
-    }
-
-
     private float[] getColor(Entity entity, String ruleSetId, boolean requestedToRemove) {
 
         if (requestedToRemove) {
@@ -420,27 +346,25 @@ public class AnnotationHighlightService {
 
 
     private float[] getColor(String type, String ruleSetId) {
+
         return dictionaryService.getColor(type, ruleSetId);
     }
 
 
     private boolean isHint(Entity entity, String ruleSetId) {
+
         return dictionaryService.isHint(entity.getType(), ruleSetId);
     }
 
+
     private boolean isRecommendation(Entity entity, String ruleSetId) {
+
         return dictionaryService.isRecommendation(entity.getType(), ruleSetId);
     }
 
 
-    private void drawSectionFrames(PDDocument document, Document classifiedDoc, boolean flatRedaction, PDPage pdPage,
-                                   int page) throws IOException {
+    private void addSectionGrid(Document classifiedDoc, int page) {
 
-        if (flatRedaction) {
-            return;
-        }
-
-        PDPageContentStream contentStream = new PDPageContentStream(document, pdPage, PDPageContentStream.AppendMode.APPEND, true);
         for (Paragraph paragraph : classifiedDoc.getParagraphs()) {
 
             for (int i = 0; i <= paragraph.getPageBlocks().size() - 1; i++) {
@@ -450,10 +374,8 @@ public class AnnotationHighlightService {
                 if (textBlock.getPage() != page) {
                     continue;
                 }
-                if (textBlock instanceof TextBlock) {
 
-                    textBlock.setClassification((i + 1) + "/" + paragraph.getPageBlocks().size());
-                    visualizeTextBlock((TextBlock) textBlock, contentStream);
+                if (textBlock instanceof TextBlock) {
 
                     classifiedDoc.getSectionGrid()
                             .getRectanglesPerPage()
@@ -463,8 +385,15 @@ public class AnnotationHighlightService {
 
                 } else if (textBlock instanceof Table) {
 
-                    textBlock.setClassification((i + 1) + "/" + paragraph.getPageBlocks().size());
-                    List<CellRectangle> cellRectangles = visualizeTable((Table) textBlock, contentStream);
+                    List<CellRectangle> cellRectangles = new ArrayList<>();
+                    for (List<Cell> row : ((Table) textBlock).getRows()) {
+                        for (Cell cell : row) {
+                            if (cell != null) {
+                                cellRectangles.add(new CellRectangle(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
+                                        .getWidth(), (float) cell.getHeight()));
+                            }
+                        }
+                    }
 
                     classifiedDoc.getSectionGrid()
                             .getRectanglesPerPage()
@@ -475,56 +404,6 @@ public class AnnotationHighlightService {
                 }
             }
         }
-        contentStream.close();
-    }
-
-
-    private void visualizeTextBlock(TextBlock textBlock, PDPageContentStream contentStream) throws IOException {
-
-        contentStream.setStrokingColor(Color.LIGHT_GRAY);
-        contentStream.setLineWidth(0.5f);
-        contentStream.addRect(textBlock.getMinX(), textBlock.getMinY(), textBlock.getWidth(), textBlock.getHeight());
-        contentStream.stroke();
-
-        if (textBlock.getClassification() != null) {
-            contentStream.beginText();
-            contentStream.setNonStrokingColor(Color.DARK_GRAY);
-            contentStream.setFont(PDType1Font.TIMES_ROMAN, 8f);
-            contentStream.newLineAtOffset(textBlock.getMinX(), textBlock.getMaxY());
-            contentStream.showText(textBlock.getClassification());
-            contentStream.endText();
-        }
-    }
-
-
-    private List<CellRectangle> visualizeTable(Table table, PDPageContentStream contentStream) throws IOException {
-
-        List<CellRectangle> cellRectangles = new ArrayList<>();
-        for (List<Cell> row : table.getRows()) {
-            for (Cell cell : row) {
-
-                if (cell != null) {
-                    contentStream.setLineWidth(0.5f);
-                    contentStream.setStrokingColor(Color.CYAN);
-                    contentStream.addRect((float) cell.getX(), (float) cell.getY(), (float) cell.getWidth(), (float) cell
-                            .getHeight());
-                    contentStream.stroke();
-                    cellRectangles.add(new CellRectangle(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
-                            .getWidth(), (float) cell.getHeight()));
-                }
-            }
-        }
-
-        if (table.getClassification() != null) {
-            contentStream.beginText();
-            contentStream.setNonStrokingColor(Color.DARK_GRAY);
-            contentStream.setFont(PDType1Font.TIMES_ROMAN, 8f);
-            contentStream.newLineAtOffset(table.getMinX(), table.getMinY());
-            contentStream.showText(table.getClassification());
-            contentStream.endText();
-        }
-
-        return cellRectangles;
     }
 
 }

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

@@ -104,4 +104,27 @@ public class EntitySearchUtils {
         }
         entities.removeAll(wordsToRemove);
     }
+
+
+    public void addEntitiesWithHigherRank(Set<Entity> entities, Set<Entity> found, Dictionary dictionary) {
+        found.forEach(f -> addEntitiesWithHigherRank(entities, f, dictionary));
+    }
+
+
+    public void addEntitiesWithHigherRank(Set<Entity> entities, Entity found, Dictionary dictionary) {
+
+        if(entities.contains(found)){
+            Entity existing = entities.stream().filter(entity -> entity.equals(found)).findFirst().get();
+            if (dictionary.getDictionaryRank(existing.getType()) <= dictionary.getDictionaryRank(found.getType())){
+                entities.remove(found);
+            }
+        }
+        entities.add(found);
+    }
+
+    public void addEntitiesIgnoreRank(Set<Entity> entities, Set<Entity> found){
+        // HashSet keeps old value but we want the new.
+        entities.removeAll(found);
+        entities.addAll(found);
+    }
 }

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

@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.redaction.utils;
 
+import java.awt.geom.Rectangle2D;
 import java.nio.charset.StandardCharsets;
 import java.util.List;
 
@@ -23,4 +24,16 @@ public class IdBuilder {
 
         return hashFunction.hashString(sb.toString(), StandardCharsets.UTF_8).toString();
     }
+
+
+    public String buildId(Rectangle2D rectangle2D, int page){
+
+        StringBuilder sb = new StringBuilder();
+        sb.append("x").append(rectangle2D.getX()).append("y").append(rectangle2D.getY()).append("h").append(rectangle2D.getHeight()).append("w").append(rectangle2D.getWidth()).append("p").append(page);
+
+        return hashFunction.hashString(sb.toString(), StandardCharsets.UTF_8).toString();
+    }
+
+
+
 }

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

@@ -48,6 +48,7 @@ public class PdfSegmentationService {
             stripper.setPageNumber(pageNumber);
             stripper.setStartPage(pageNumber);
             stripper.setEndPage(pageNumber);
+            stripper.setPdpage(pdPage);
             stripper.getText(pdDocument);
 
             PDRectangle pdr = pdPage.getMediaBox();
@@ -56,18 +57,17 @@ public class PdfSegmentationService {
             int rotation = pdPage.getRotation();
             boolean isRotated = rotation != 0 && rotation != 360;
 
-
             ParsedElements parsedElements = ParsedElements.builder()
                     .rulings(stripper.getRulings())
                     .sequences(stripper.getTextPositionSequences())
-                    .maxCharWidth(stripper.getMaxCharWidths())
-                    .maxCharHeight(stripper.getMaxCharWidths())
+                    .imageBounds(stripper.getImageBounds())
+                    .minCharWidth(stripper.getMinCharWidth())
+                    .maxCharWidth(stripper.getMaxCharWidth())
                     .landscape(isLandscape)
                     .rotated(isRotated)
                     .build();
 
-            CleanRulings cleanRulings = rulingCleaningService.getCleanRulings(parsedElements.getRulings(), parsedElements
-                    .getMaxCharWidth(), parsedElements.getMaxCharHeight());
+            CleanRulings cleanRulings = rulingCleaningService.getCleanRulings(parsedElements.getRulings(), stripper.getMinCharWidth(), stripper.getMaxCharHeight());
 
             Page page = blockificationService.blockify(parsedElements.getSequences(), cleanRulings.getHorizontal(), cleanRulings
                     .getVertical());
@@ -81,8 +81,10 @@ public class PdfSegmentationService {
 
             page.setPageNumber(pageNumber);
             increaseDocumentStatistics(page, document);
+            page.setImageBounds(parsedElements.getImageBounds());
             pages.add(page);
         }
+
         document.setPages(pages);
 
         classificationService.classifyDocument(document);
@@ -90,11 +92,9 @@ public class PdfSegmentationService {
         sectionsBuilderService.buildSections(document);
 
         return document;
-
     }
 
 
-
     private void increaseDocumentStatistics(Page page, Document document) {
 
         if (!page.isLandscape()) {

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

@@ -10,9 +10,12 @@ import org.apache.commons.collections4.CollectionUtils;
 import org.springframework.stereotype.Service;
 
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
+import com.iqser.red.service.redaction.v1.server.classification.model.Footer;
+import com.iqser.red.service.redaction.v1.server.classification.model.Header;
 import com.iqser.red.service.redaction.v1.server.classification.model.Page;
 import com.iqser.red.service.redaction.v1.server.classification.model.Paragraph;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+import com.iqser.red.service.redaction.v1.server.classification.model.UnclassifiedText;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
@@ -24,25 +27,46 @@ public class SectionsBuilderService {
 
         List<AbstractTextContainer> chunkWords = new ArrayList<>();
         List<Paragraph> chunkBlockList = new ArrayList<>();
+        List<Header> headers = new ArrayList<>();
+        List<Footer> footers = new ArrayList<>();
+        List<UnclassifiedText> unclassifiedTexts = new ArrayList<>();
 
         AbstractTextContainer prev = null;
 
         String lastHeadline = "";
         Table previousTable = null;
         for (Page page : document.getPages()) {
+            List<TextBlock> header = new ArrayList<>();
+            List<TextBlock> footer = new ArrayList<>();
+            List<TextBlock> unclassifiedText = new ArrayList<>();
             for (AbstractTextContainer current : page.getTextBlocks()) {
 
-                if (current.getClassification() == null || current.getClassification()
-                        .equals("Header") || current.getClassification().equals("Footer")) {
+                if (current.getClassification() == null) {
                     continue;
                 }
 
                 current.setPage(page.getPageNumber());
 
-                if (prev != null && current.getClassification().startsWith("H ") && !prev.getClassification().startsWith("H ") || !document.isHeadlines()) {
+                if (current.getClassification().equals("Header")) {
+                    header.add((TextBlock) current);
+                    continue;
+                }
+
+                if (current.getClassification().equals("Footer")) {
+                    footer.add((TextBlock) current);
+                    continue;
+                }
+
+                if (current.getClassification().equals("Other")) {
+                    unclassifiedText.add((TextBlock) current);
+                    continue;
+                }
+
+                if (prev != null && current.getClassification().startsWith("H ") && !prev.getClassification()
+                        .startsWith("H ") || !document.isHeadlines()) {
                     Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
                     chunkBlock.setHeadline(lastHeadline);
-                    if(document.isHeadlines()) {
+                    if (document.isHeadlines()) {
                         lastHeadline = current.getText();
                     }
                     chunkBlockList.add(chunkBlock);
@@ -60,6 +84,16 @@ public class SectionsBuilderService {
                 chunkWords.add(current);
                 prev = current;
             }
+
+            if(!header.isEmpty()) {
+                headers.add(new Header(header));
+            }
+            if(!footer.isEmpty()) {
+                footers.add(new Footer(footer));
+            }
+            if(!unclassifiedText.isEmpty()) {
+                unclassifiedTexts.add(new UnclassifiedText(unclassifiedText));
+            }
         }
 
         Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
@@ -67,6 +101,9 @@ public class SectionsBuilderService {
         chunkBlockList.add(chunkBlock);
 
         document.setParagraphs(chunkBlockList);
+        document.setHeaders(headers);
+        document.setFooters(footers);
+        document.setUnclassifiedTexts(unclassifiedTexts);
     }
 
 
@@ -175,9 +212,9 @@ public class SectionsBuilderService {
 
     private boolean hasInvalidHeaderInformation(Table table) {
 
-        return table.getRows().stream()
-                .flatMap(row -> row.stream()
-                        .filter(cell -> CollectionUtils.isNotEmpty(cell.getHeaderCells())))
+        return table.getRows()
+                .stream()
+                .flatMap(row -> row.stream().filter(cell -> CollectionUtils.isNotEmpty(cell.getHeaderCells())))
                 .findAny()
                 .isEmpty();
 
@@ -188,6 +225,9 @@ public class SectionsBuilderService {
 
         for (int i = table.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
             List<Cell> row = table.getRows().get(i);
+            if (row.size() == 1) {
+                continue;
+            }
             boolean allNonHeader = true;
             for (Cell cell : row) {
                 if (cell.isHeaderCell()) {

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/tableextraction/model/Cell.java

@@ -11,10 +11,12 @@ import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizati
 
 import lombok.Data;
 import lombok.EqualsAndHashCode;
+import lombok.NoArgsConstructor;
 
 @SuppressWarnings("serial")
 @Data
 @EqualsAndHashCode(callSuper = true)
+@NoArgsConstructor
 public class Cell extends Rectangle {
 
     private List<TextBlock> textBlocks = new ArrayList<>();

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/tableextraction/model/Table.java

@@ -59,6 +59,16 @@ public class Table extends AbstractTextContainer {
 
         if (rows == null) {
             rows = computeRows();
+
+            // Ignore rows that does not contain any cells and values.
+            List<List<Cell>> rowsToRemove = new ArrayList<>();
+            for (List<Cell> row: rows){
+                if (row.size() == 1 && row.get(0).getTextBlocks().isEmpty()){
+                    rowsToRemove.add(row);
+                }
+            }
+            rows.removeAll(rowsToRemove);
+
             computeHeaders();
         }
 
@@ -100,6 +110,10 @@ public class Table extends AbstractTextContainer {
         // we move from left to right and top to bottom
         for (int rowIndex = 0; rowIndex < rows.size(); rowIndex++) {
             List<Cell> rowCells = rows.get(rowIndex);
+            if(rowCells.size() == 1){
+                continue;
+            }
+
             for (int colIndex = 0; colIndex < rowCells.size(); colIndex++) {
                 Cell cell = rowCells.get(colIndex);
                 List<Cell> cellsToTheLeft = rowCells.subList(0, colIndex);

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

@@ -18,9 +18,10 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;
 @Service
 public class RulingCleaningService {
 
-    public CleanRulings getCleanRulings(List<Ruling> rulings, float maxCharWidth, float maxCharHeight){
+    public CleanRulings getCleanRulings(List<Ruling> rulings, float minCharWidth, float maxCharHeight) {
+
         if (!rulings.isEmpty()) {
-            snapPoints(rulings, maxCharWidth , maxCharHeight);
+            snapPoints(rulings, minCharWidth, maxCharHeight);
         }
 
         List<Ruling> vrs = new ArrayList<>();
@@ -39,13 +40,10 @@ public class RulingCleaningService {
         }
         List<Ruling> horizontalRulingLines = collapseOrientedRulings(hrs);
 
-        return CleanRulings
-                .builder()
-                .vertical(verticalRulingLines)
-                .horizontal(horizontalRulingLines)
-                .build();
+        return CleanRulings.builder().vertical(verticalRulingLines).horizontal(horizontalRulingLines).build();
     }
 
+
     public void snapPoints(List<? extends Line2D.Float> rulings, float xThreshold, float yThreshold) {
 
         // collect points and keep a Line -> p1,p2 map
@@ -122,12 +120,14 @@ public class RulingCleaningService {
 
 
     private List<Ruling> collapseOrientedRulings(List<Ruling> lines) {
+
         int COLINEAR_OR_PARALLEL_PIXEL_EXPAND_AMOUNT = 1;
         return collapseOrientedRulings(lines, COLINEAR_OR_PARALLEL_PIXEL_EXPAND_AMOUNT);
     }
 
 
     private List<Ruling> collapseOrientedRulings(List<Ruling> lines, int expandAmount) {
+
         ArrayList<Ruling> rv = new ArrayList<>();
         lines.sort((a, b) -> {
             final float diff = a.getPosition() - b.getPosition();
@@ -141,25 +141,24 @@ public class RulingCleaningService {
                 final float lastStart = last.getStart();
                 final float lastEnd = last.getEnd();
 
-                final boolean lastFlipped = lastStart            > lastEnd;
+                final boolean lastFlipped = lastStart > lastEnd;
                 final boolean nextFlipped = next_line.getStart() > next_line.getEnd();
 
                 boolean differentDirections = nextFlipped != lastFlipped;
-                float nextS = differentDirections ? next_line.getEnd()   : next_line.getStart();
+                float nextS = differentDirections ? next_line.getEnd() : next_line.getStart();
                 float nextE = differentDirections ? next_line.getStart() : next_line.getEnd();
 
                 final float newStart = lastFlipped ? Math.max(nextS, lastStart) : Math.min(nextS, lastStart);
-                final float newEnd   = lastFlipped ? Math.min(nextE, lastEnd)   : Math.max(nextE, lastEnd);
+                final float newEnd = lastFlipped ? Math.min(nextE, lastEnd) : Math.max(nextE, lastEnd);
                 last.setStartEnd(newStart, newEnd);
                 assert !last.oblique();
-            }
-            else if (next_line.length() == 0) {
+            } else if (next_line.length() == 0) {
                 continue;
-            }
-            else {
+            } else {
                 rv.add(next_line);
             }
         }
         return rv;
     }
+
 }

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

@@ -1,27 +1,43 @@
 package com.iqser.red.service.redaction.v1.server;
 
-import com.iqser.red.service.configuration.v1.api.model.Colors;
-import com.iqser.red.service.configuration.v1.api.model.DictionaryResponse;
-import com.iqser.red.service.configuration.v1.api.model.RulesResponse;
-import com.iqser.red.service.configuration.v1.api.model.TypeResponse;
-import com.iqser.red.service.configuration.v1.api.model.TypeResult;
-import com.iqser.red.service.redaction.v1.model.Comment;
-import com.iqser.red.service.redaction.v1.model.IdRemoval;
-import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
-import com.iqser.red.service.redaction.v1.model.ManualRedactions;
-import com.iqser.red.service.redaction.v1.model.Point;
-import com.iqser.red.service.redaction.v1.model.Rectangle;
-import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
-import com.iqser.red.service.redaction.v1.model.RedactionRequest;
-import com.iqser.red.service.redaction.v1.model.RedactionResult;
-import com.iqser.red.service.redaction.v1.model.Status;
-import com.iqser.red.service.redaction.v1.server.client.DictionaryClient;
-import com.iqser.red.service.redaction.v1.server.client.RulesClient;
-import com.iqser.red.service.redaction.v1.server.controller.RedactionController;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
+import static org.assertj.core.api.Assertions.assertThat;
+import static org.mockito.Mockito.when;
+import static org.springframework.boot.test.context.SpringBootTest.WebEnvironment.RANDOM_PORT;
+
+import java.awt.Color;
+import java.awt.geom.AffineTransform;
+import java.awt.geom.Rectangle2D;
+import java.io.BufferedReader;
+import java.io.ByteArrayInputStream;
+import java.io.ByteArrayOutputStream;
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.FileOutputStream;
+import java.io.IOException;
+import java.io.InputStream;
+import java.io.InputStreamReader;
+import java.net.URL;
+import java.nio.charset.StandardCharsets;
+import java.time.OffsetDateTime;
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.UUID;
+import java.util.stream.Collectors;
+
 import org.apache.commons.io.IOUtils;
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.pdmodel.PDPageContentStream;
+import org.apache.pdfbox.pdmodel.common.PDRectangle;
+import org.apache.pdfbox.pdmodel.graphics.image.PDImageXObject;
+import org.apache.pdfbox.text.PDFTextStripperByArea;
+import org.apache.pdfbox.util.Matrix;
 import org.junit.Before;
+import org.junit.Ignore;
 import org.junit.Test;
 import org.junit.runner.RunWith;
 import org.kie.api.KieServices;
@@ -37,28 +53,44 @@ import org.springframework.context.annotation.Bean;
 import org.springframework.core.io.ClassPathResource;
 import org.springframework.test.context.junit4.SpringRunner;
 
-import java.io.BufferedReader;
-import java.io.ByteArrayInputStream;
-import java.io.File;
-import java.io.FileInputStream;
-import java.io.FileOutputStream;
-import java.io.IOException;
-import java.io.InputStream;
-import java.io.InputStreamReader;
-import java.net.URL;
-import java.nio.charset.StandardCharsets;
-import java.time.OffsetDateTime;
-import java.util.ArrayList;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-import java.util.Set;
-import java.util.UUID;
-import java.util.stream.Collectors;
-
-import static org.assertj.core.api.Assertions.assertThat;
-import static org.mockito.Mockito.when;
-import static org.springframework.boot.test.context.SpringBootTest.WebEnvironment.RANDOM_PORT;
+import com.fasterxml.jackson.databind.ObjectMapper;
+import com.iqser.red.service.configuration.v1.api.model.Colors;
+import com.iqser.red.service.configuration.v1.api.model.DictionaryResponse;
+import com.iqser.red.service.configuration.v1.api.model.RulesResponse;
+import com.iqser.red.service.configuration.v1.api.model.TypeResponse;
+import com.iqser.red.service.configuration.v1.api.model.TypeResult;
+import com.iqser.red.service.redaction.v1.model.AnalyzeRequest;
+import com.iqser.red.service.redaction.v1.model.AnalyzeResult;
+import com.iqser.red.service.redaction.v1.model.AnnotateRequest;
+import com.iqser.red.service.redaction.v1.model.AnnotateResponse;
+import com.iqser.red.service.redaction.v1.model.Comment;
+import com.iqser.red.service.redaction.v1.model.IdRemoval;
+import com.iqser.red.service.redaction.v1.model.ManualForceRedact;
+import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
+import com.iqser.red.service.redaction.v1.model.ManualRedactions;
+import com.iqser.red.service.redaction.v1.model.Point;
+import com.iqser.red.service.redaction.v1.model.ReanalyzeResult;
+import com.iqser.red.service.redaction.v1.model.Rectangle;
+import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import com.iqser.red.service.redaction.v1.model.RedactionRequest;
+import com.iqser.red.service.redaction.v1.model.RedactionResult;
+import com.iqser.red.service.redaction.v1.model.RenalyzeRequest;
+import com.iqser.red.service.redaction.v1.model.SectionArea;
+import com.iqser.red.service.redaction.v1.model.SectionText;
+import com.iqser.red.service.redaction.v1.model.Status;
+import com.iqser.red.service.redaction.v1.server.classification.model.Paragraph;
+import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+import com.iqser.red.service.redaction.v1.server.client.DictionaryClient;
+import com.iqser.red.service.redaction.v1.server.client.RulesClient;
+import com.iqser.red.service.redaction.v1.server.controller.RedactionController;
+import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
+import com.iqser.red.service.redaction.v1.server.parsing.PDFAreaTextStripper;
+import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
+import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
+import com.iqser.red.service.redaction.v1.server.redaction.model.ReanalysisSection;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
+import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
 
 @RunWith(SpringRunner.class)
 @SpringBootTest(webEnvironment = RANDOM_PORT)
@@ -75,6 +107,8 @@ public class RedactionIntegrationTest {
     private static final String MUST_REDACT = "must_redact";
     private static final String PUBLISHED_INFORMATION = "published_information";
     private static final String TEST_METHOD = "test_method";
+    private static final String PURITY = "purity";
+    private static final String IMAGE = "image";
 
     private static final String RECOMMENDATION_AUTHOR = "recommendation_CBI_author";
     private static final String RECOMMENDATION_ADDRESS = "recommendation_CBI_address";
@@ -86,6 +120,9 @@ public class RedactionIntegrationTest {
     @Autowired
     private RedactionController redactionController;
 
+    @Autowired
+    private ObjectMapper objectMapper;
+
     @MockBean
     private RulesClient rulesClient;
 
@@ -97,6 +134,7 @@ public class RedactionIntegrationTest {
     private final Map<String, Boolean> hintTypeMap = new HashMap<>();
     private final Map<String, Boolean> caseInSensitiveMap = new HashMap<>();
     private final Map<String, Boolean> recommendationTypeMap = new HashMap<>();
+    private final Map<String, Integer> rankTypeMap = new HashMap<>();
     private final Colors colors = new Colors();
 
     private final static String TEST_RULESET_ID = "123";
@@ -132,7 +170,9 @@ public class RedactionIntegrationTest {
         loadDictionaryForTest();
         loadTypeForTest();
         when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(0L);
-        when(dictionaryClient.getAllTypes(TEST_RULESET_ID)).thenReturn(TypeResponse.builder().types(getTypeResponse()).build());
+        when(dictionaryClient.getAllTypes(TEST_RULESET_ID)).thenReturn(TypeResponse.builder()
+                .types(getTypeResponse())
+                .build());
         when(dictionaryClient.getDictionaryForType(VERTEBRATE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(VERTEBRATE));
         when(dictionaryClient.getDictionaryForType(ADDRESS, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(ADDRESS));
         when(dictionaryClient.getDictionaryForType(AUTHOR, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(AUTHOR));
@@ -147,6 +187,8 @@ public class RedactionIntegrationTest {
         when(dictionaryClient.getDictionaryForType(RECOMMENDATION_AUTHOR, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(RECOMMENDATION_AUTHOR));
         when(dictionaryClient.getDictionaryForType(RECOMMENDATION_ADDRESS, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(RECOMMENDATION_ADDRESS));
         when(dictionaryClient.getDictionaryForType(FALSE_POSITIVE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(FALSE_POSITIVE));
+        when(dictionaryClient.getDictionaryForType(PURITY, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(PURITY));
+        when(dictionaryClient.getDictionaryForType(IMAGE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(IMAGE));
         when(dictionaryClient.getColors(TEST_RULESET_ID)).thenReturn(colors);
     }
 
@@ -223,6 +265,16 @@ public class RedactionIntegrationTest {
                         .stream()
                         .map(this::cleanDictionaryEntry)
                         .collect(Collectors.toSet()));
+        dictionary.computeIfAbsent(PURITY, v -> new ArrayList<>())
+                .addAll(ResourceLoader.load("dictionaries/purity.txt")
+                        .stream()
+                        .map(this::cleanDictionaryEntry)
+                        .collect(Collectors.toSet()));
+        dictionary.computeIfAbsent(IMAGE, v -> new ArrayList<>())
+                .addAll(ResourceLoader.load("dictionaries/image.txt")
+                        .stream()
+                        .map(this::cleanDictionaryEntry)
+                        .collect(Collectors.toSet()));
     }
 
 
@@ -248,6 +300,8 @@ public class RedactionIntegrationTest {
         typeColorMap.put(RECOMMENDATION_AUTHOR, "#8df06c");
         typeColorMap.put(RECOMMENDATION_ADDRESS, "#8df06c");
         typeColorMap.put(FALSE_POSITIVE, "#ffffff");
+        typeColorMap.put(PURITY, "#ffe187");
+        typeColorMap.put(IMAGE, "#fcc5fb");
 
         hintTypeMap.put(VERTEBRATE, true);
         hintTypeMap.put(ADDRESS, false);
@@ -263,6 +317,8 @@ public class RedactionIntegrationTest {
         hintTypeMap.put(RECOMMENDATION_AUTHOR, false);
         hintTypeMap.put(RECOMMENDATION_ADDRESS, false);
         hintTypeMap.put(FALSE_POSITIVE, true);
+        hintTypeMap.put(PURITY, false);
+        hintTypeMap.put(IMAGE, true);
 
         caseInSensitiveMap.put(VERTEBRATE, true);
         caseInSensitiveMap.put(ADDRESS, false);
@@ -278,6 +334,8 @@ public class RedactionIntegrationTest {
         caseInSensitiveMap.put(RECOMMENDATION_AUTHOR, false);
         caseInSensitiveMap.put(RECOMMENDATION_ADDRESS, false);
         caseInSensitiveMap.put(FALSE_POSITIVE, false);
+        caseInSensitiveMap.put(PURITY, false);
+        caseInSensitiveMap.put(IMAGE, true);
 
         recommendationTypeMap.put(VERTEBRATE, false);
         recommendationTypeMap.put(ADDRESS, false);
@@ -293,6 +351,25 @@ public class RedactionIntegrationTest {
         recommendationTypeMap.put(RECOMMENDATION_AUTHOR, true);
         recommendationTypeMap.put(RECOMMENDATION_ADDRESS, true);
         recommendationTypeMap.put(FALSE_POSITIVE, false);
+        recommendationTypeMap.put(PURITY, false);
+        recommendationTypeMap.put(IMAGE, false);
+
+        rankTypeMap.put(FALSE_POSITIVE, 160);
+        rankTypeMap.put(PURITY, 155);
+        rankTypeMap.put(PII, 150);
+        rankTypeMap.put(ADDRESS, 140);
+        rankTypeMap.put(AUTHOR, 130);
+        rankTypeMap.put(SPONSOR, 120);
+        rankTypeMap.put(VERTEBRATE, 110);
+        rankTypeMap.put(MUST_REDACT, 100);
+        rankTypeMap.put(REDACTION_INDICATOR, 90);
+        rankTypeMap.put(NO_REDACTION_INDICATOR, 80);
+        rankTypeMap.put(PUBLISHED_INFORMATION, 70);
+        rankTypeMap.put(TEST_METHOD, 60);
+        rankTypeMap.put(HINT_ONLY, 50);
+        rankTypeMap.put(RECOMMENDATION_AUTHOR, 40);
+        rankTypeMap.put(RECOMMENDATION_ADDRESS, 30);
+        rankTypeMap.put(IMAGE, 30);
 
         colors.setDefaultColor("#acfc00");
         colors.setNotRedacted("#cccccc");
@@ -312,6 +389,7 @@ public class RedactionIntegrationTest {
                         .isHint(hintTypeMap.get(typeColor.getKey()))
                         .isCaseInsensitive(caseInSensitiveMap.get(typeColor.getKey()))
                         .isRecommendation(recommendationTypeMap.get(typeColor.getKey()))
+                        .rank(rankTypeMap.get(typeColor.getKey()))
                         .build())
 
                 .collect(Collectors.toList());
@@ -326,6 +404,7 @@ public class RedactionIntegrationTest {
                 .isHint(hintTypeMap.get(type))
                 .isCaseInsensitive(caseInSensitiveMap.get(type))
                 .isRecommendation(recommendationTypeMap.get(type))
+                .rank(rankTypeMap.get(type))
                 .build();
     }
 
@@ -343,12 +422,12 @@ public class RedactionIntegrationTest {
             input.addAll(getPathsRecursively(file));
         }
         for (File path : input) {
-            RedactionRequest request = RedactionRequest.builder()
+            AnalyzeRequest request = AnalyzeRequest.builder()
                     .ruleSetId(TEST_RULESET_ID)
                     .document(IOUtils.toByteArray(new FileInputStream(path)))
                     .build();
             System.out.println("Redacting file : " + path.getName());
-            RedactionResult result = redactionController.redact(request);
+            AnalyzeResult result = redactionController.analyze(request);
 
             Map<String, List<RedactionLogEntry>> duplicates = new HashMap<>();
             result.getRedactionLog().getRedactionLogEntry().forEach(entry -> {
@@ -358,6 +437,23 @@ public class RedactionIntegrationTest {
             duplicates.entrySet().forEach(entry -> {
                 assertThat(entry.getValue().size()).isEqualTo(1);
             });
+
+
+            dictionary.get(AUTHOR).add("Drinking water");
+            when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(1L);
+
+            long rstart = System.currentTimeMillis();
+            ReanalyzeResult reanalyzeResult = redactionController.reanalyze(RenalyzeRequest.builder()
+                    .redactionLog(result.getRedactionLog())
+                    .document(IOUtils.toByteArray(new FileInputStream(path)))
+                    .manualRedactions(null)
+                    .text(result.getText())
+                    .ruleSetId(TEST_RULESET_ID)
+                    .build());
+
+            long rend = System.currentTimeMillis();
+            System.out.println("reanalysis analysis duration: " + (rend - rstart));
+
         }
 
         long end = System.currentTimeMillis();
@@ -390,32 +486,194 @@ public class RedactionIntegrationTest {
 
         System.out.println("redactionTest");
         long start = System.currentTimeMillis();
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Applicant Producer Table.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Metolachlor/S-Metolachlor_RAR_01_Volume_1_2018-09-06.pdf");
 
-        RedactionRequest request = RedactionRequest.builder()
+        AnalyzeRequest request = AnalyzeRequest.builder()
                 .ruleSetId(TEST_RULESET_ID)
                 .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
                 .build();
-        request.setFlatRedaction(false);
 
-        RedactionResult result = redactionController.redact(request);
+        AnalyzeResult result = redactionController.analyze(request);
 
-//        result.getRedactionLog().getRedactionLogEntry().forEach(entry -> {
-//            if(!entry.isHint()){
-//                System.out.println(entry.getPositions().get(0).getPage() +":"+ entry.getTextBefore() +"--->"+ entry.getValue() + "--->" + entry.getTextAfter());
-//            }
-//        });
+        long end = System.currentTimeMillis();
 
-        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Redacted.pdf")) {
-            fileOutputStream.write(result.getDocument());
+        System.out.println("first analysis duration: " + (end - start));
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Test.json")) {
+            fileOutputStream.write(objectMapper.writeValueAsBytes(result.getText()));
         }
+
+        int correctFound = 0;
+        loop:
+        for (RedactionLogEntry redactionLogEntry : result.getRedactionLog().getRedactionLogEntry()) {
+            for (SectionText sectionText : result.getText().getSectionTexts()) {
+                if (redactionLogEntry.getType().equals("image")) {
+                    correctFound++;
+                    continue loop;
+                }
+                if (redactionLogEntry.getSectionNumber() == sectionText.getSectionNumber()) {
+                    String value = sectionText.getText()
+                            .substring(redactionLogEntry.getStartOffset(), redactionLogEntry.getEndOffset());
+                    if (redactionLogEntry.getValue().equalsIgnoreCase(value)) {
+                        correctFound++;
+                    } else {
+                        throw new RuntimeException("WTF");
+                    }
+                }
+            }
+        }
+        assertThat(correctFound).isEqualTo(result.getRedactionLog().getRedactionLogEntry().size());
+
+
+        dictionary.get(AUTHOR).add("Properties");
+        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(1L);
+
+        start = System.currentTimeMillis();
+        ReanalyzeResult reanalyzeResult = redactionController.reanalyze(RenalyzeRequest.builder()
+                .redactionLog(result.getRedactionLog())
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .manualRedactions(null)
+                .text(result.getText())
+                .ruleSetId(TEST_RULESET_ID)
+                .build());
+
+        end = System.currentTimeMillis();
+        System.out.println("reanalysis analysis duration: " + (end - start));
+
+        AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .redactionLog(reanalyzeResult.getRedactionLog())
+                .sectionGrid(result.getSectionGrid())
+                .build());
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Annotated.pdf")) {
+            fileOutputStream.write(annotateResponse.getDocument());
+        }
+
+    }
+
+
+
+    @Test
+    @Ignore
+    public void fillRecanTest() throws IOException {
+
+        System.out.println("redactionTest");
+        long start = System.currentTimeMillis();
+        ClassPathResource pdfFileResource = new ClassPathResource("files/S5.pdf");
+
+        AnalyzeRequest request = AnalyzeRequest.builder()
+                .ruleSetId(TEST_RULESET_ID)
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .build();
+
+        AnalyzeResult result = redactionController.analyze(request);
+
+        AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .redactionLog(result.getRedactionLog())
+                .sectionGrid(result.getSectionGrid())
+                .build());
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Annotated.pdf")) {
+            fileOutputStream.write(annotateResponse.getDocument());
+        }
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Test.json")) {
+            fileOutputStream.write(objectMapper.writeValueAsBytes(result.getText()));
+        }
+
+        int correctFound = 0;
+        loop:
+        for (RedactionLogEntry redactionLogEntry : result.getRedactionLog().getRedactionLogEntry()) {
+            for (SectionText sectionText : result.getText().getSectionTexts()) {
+                if (redactionLogEntry.getType().equals("image")) {
+                    correctFound++;
+                    continue loop;
+                }
+                if (redactionLogEntry.getSectionNumber() == sectionText.getSectionNumber()) {
+                    String value = sectionText.getText()
+                            .substring(redactionLogEntry.getStartOffset(), redactionLogEntry.getEndOffset());
+                    if (redactionLogEntry.getValue().equalsIgnoreCase(value)) {
+                        correctFound++;
+                    } else {
+                        throw new RuntimeException("WTF");
+                    }
+                }
+            }
+        }
+        assertThat(correctFound).isEqualTo(result.getRedactionLog().getRedactionLogEntry().size());
+
+        System.out.println("correctFound " + correctFound);
+
         long end = System.currentTimeMillis();
 
         System.out.println("duration: " + (end - start));
         System.out.println("numberOfPages: " + result.getNumberOfPages());
+
+        SectionArea sectionArea = result.getText().getSectionTexts().get(3).getSectionAreas().get(5);
+
+        try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(IOUtils.toByteArray(pdfFileResource.getInputStream())))) {
+
+            PDPage docPage = pdDocument.getPage(0);
+
+            PDFTextStripperByArea textStripper = new PDFTextStripperByArea();
+
+            PDRectangle cropBox = docPage.getCropBox();
+            PDRectangle mediaBox = docPage.getMediaBox();
+
+
+//            if (textPositions.get(0).getRotation() == 90) {
+//                posXEnd = textPositions.get(0).getYDirAdj() + 2;
+//                posYInit = getY1();
+//                posYEnd = textPositions.get(textPositions.size() - 1).getXDirAdj() - height + 4;
+//            } else {
+//                posXEnd = textPositions.get(textPositions.size() - 1)
+//                        .getXDirAdj() + textPositions.get(textPositions.size() - 1).getWidth() + 1;
+//                posYInit = textPositions.get(0).getPageHeight() - textPositions.get(0).getYDirAdj() - 2;
+//                posYEnd = textPositions.get(0).getPageHeight() - textPositions.get(textPositions.size() - 1)
+//                        .getYDirAdj() + 2;
+//            }
+
+
+            Rectangle2D rect = new Rectangle2D.Float(sectionArea.getTopLeft()
+                    .getY(), sectionArea.getTopLeft()
+                    .getX() , sectionArea.getHeight(), sectionArea
+                    .getWidth() + 0.001f);
+
+            textStripper.addRegion("region", rect);
+
+
+
+            textStripper.extractRegions(docPage);
+
+            String textForRegion = textStripper.getTextForRegion("region");
+
+            System.out.println(textForRegion);
+
+            // fill a rectangle
+            PDPageContentStream contents = new PDPageContentStream (pdDocument, docPage, PDPageContentStream.AppendMode.APPEND, false, false);
+            contents.setNonStrokingColor (Color.RED);
+            contents.addRect (sectionArea.getTopLeft().getX(), sectionArea.getTopLeft().getY(), sectionArea.getWidth(), sectionArea.getHeight());
+            contents.fill ();
+            contents.close ();
+            try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream()) {
+                pdDocument.save(byteArrayOutputStream);
+                try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Annotated2.pdf")) {
+                    fileOutputStream.write(byteArrayOutputStream.toByteArray());
+                }
+            }
+
+        } catch (Exception e) {
+            throw new RedactionException(e);
+        }
+
     }
 
 
+
+
+
     @Test
     public void testTableRedaction() throws IOException {
 
@@ -423,15 +681,21 @@ public class RedactionIntegrationTest {
         long start = System.currentTimeMillis();
         ClassPathResource pdfFileResource = new ClassPathResource("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");
 
-        RedactionRequest request = RedactionRequest.builder()
+        AnalyzeRequest request = AnalyzeRequest.builder()
                 .ruleSetId(TEST_RULESET_ID)
                 .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
                 .build();
 
-        RedactionResult result = redactionController.redact(request);
+        AnalyzeResult result = redactionController.analyze(request);
 
-        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Redacted.pdf")) {
-            fileOutputStream.write(result.getDocument());
+        AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .redactionLog(result.getRedactionLog())
+                .sectionGrid(result.getSectionGrid())
+                .build());
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Annotated.pdf")) {
+            fileOutputStream.write(annotateResponse.getDocument());
         }
         long end = System.currentTimeMillis();
 
@@ -443,6 +707,8 @@ public class RedactionIntegrationTest {
     @Test
     public void testManualRedaction() throws IOException {
 
+//        675eba69b0c2917de55462c817adaa05
+
         System.out.println("testManualRedaction");
         long start = System.currentTimeMillis();
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Single Table.pdf");
@@ -457,8 +723,13 @@ public class RedactionIntegrationTest {
                 .text("This is a comment test")
                 .build();
         manualRedactions.setIdsToRemove(Set.of(IdRemoval.builder()
-                .id("0836727c3508a0b2ea271da69c04cc2f")
-                .status(Status.REQUESTED)
+                .id("5b940b2cb401ed9f5be6fc24f6e77bcf")
+                .status(Status.DECLINED)
+                .build()));
+        manualRedactions.setForceRedacts(Set.of(ManualForceRedact.builder()
+                .id("675eba69b0c2917de55462c817adaa05")
+                .legalBasis("Something")
+                .status(Status.APPROVED)
                 .build()));
 
         manualRedactions.getComments().put("e5be0f1d941bbb92a068e198648d06c4", List.of(comment));
@@ -473,18 +744,40 @@ public class RedactionIntegrationTest {
         manualRedactionEntry.setReason("Manual Redaction");
         manualRedactionEntry.setPositions(List.of(new Rectangle(new Point(375.61096f, 241.282f), 7.648041f, 43.72262f, 1), new Rectangle(new Point(384.83517f, 241.282f), 7.648041f, 17.043358f, 1)));
 
-        manualRedactions.getEntriesToAdd().add(manualRedactionEntry);
+//        manualRedactions.getEntriesToAdd().add(manualRedactionEntry);
 
-        RedactionRequest request = RedactionRequest.builder()
+        AnalyzeRequest request = AnalyzeRequest.builder()
                 .ruleSetId(TEST_RULESET_ID)
                 .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
                 .manualRedactions(manualRedactions)
                 .build();
 
-        RedactionResult result = redactionController.redact(request);
+        AnalyzeResult result = redactionController.analyze(request);
 
-        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Redacted.pdf")) {
-            fileOutputStream.write(result.getDocument());
+        manualRedactions.getEntriesToAdd().add(manualRedactionEntry);
+        manualRedactions.setIdsToRemove(Set.of(IdRemoval.builder()
+                .id("5b940b2cb401ed9f5be6fc24f6e77bcf")
+                .status(Status.APPROVED)
+                .build()));
+
+
+        ReanalyzeResult reanalyzeResult = redactionController.reanalyze(RenalyzeRequest.builder()
+                .redactionLog(result.getRedactionLog())
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .manualRedactions(manualRedactions)
+                .text(result.getText())
+                .ruleSetId(TEST_RULESET_ID)
+                .build());
+
+
+        AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .redactionLog(reanalyzeResult.getRedactionLog())
+                .sectionGrid(result.getSectionGrid())
+                .build());
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Annotated.pdf")) {
+            fileOutputStream.write(annotateResponse.getDocument());
         }
         long end = System.currentTimeMillis();
 
@@ -533,7 +826,7 @@ public class RedactionIntegrationTest {
     public void htmlTablesTest() throws IOException {
 
         System.out.println("htmlTablesTest");
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Fludioxonil/52 Fludioxonil_RAR_07_Volume_3CA_B-5_2018-02-21.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");
 
         RedactionRequest request = RedactionRequest.builder()
                 .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
@@ -570,13 +863,12 @@ public class RedactionIntegrationTest {
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Phantom Cells.pdf");
 
-        RedactionRequest request = RedactionRequest.builder()
+        AnalyzeRequest request = AnalyzeRequest.builder()
                 .ruleSetId(TEST_RULESET_ID)
                 .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
                 .build();
-        request.setFlatRedaction(false);
 
-        RedactionResult result = redactionController.redact(request);
+        AnalyzeResult result = redactionController.analyze(request);
 
         result.getRedactionLog().getRedactionLogEntry().forEach(entry -> {
             if (!entry.isHint()) {
@@ -592,16 +884,21 @@ public class RedactionIntegrationTest {
         long start = System.currentTimeMillis();
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/sponsor_companies.pdf");
 
-        RedactionRequest request = RedactionRequest.builder()
+        AnalyzeRequest request = AnalyzeRequest.builder()
                 .ruleSetId(TEST_RULESET_ID)
-                .flatRedaction(false)
                 .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
                 .build();
 
-        RedactionResult result = redactionController.redact(request);
+        AnalyzeResult result = redactionController.analyze(request);
 
-        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Redacted.pdf")) {
-            fileOutputStream.write(result.getDocument());
+        AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .redactionLog(result.getRedactionLog())
+                .sectionGrid(result.getSectionGrid())
+                .build());
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Annotated.pdf")) {
+            fileOutputStream.write(annotateResponse.getDocument());
         }
         long end = System.currentTimeMillis();
 

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

@@ -16,6 +16,7 @@ import com.iqser.red.service.redaction.v1.server.segmentation.PdfSegmentationSer
 import org.apache.commons.io.IOUtils;
 import org.apache.pdfbox.pdmodel.PDDocument;
 import org.junit.Before;
+import org.junit.Ignore;
 import org.junit.Test;
 import org.junit.runner.RunWith;
 import org.kie.api.KieServices;
@@ -412,6 +413,7 @@ public class EntityRedactionServiceTest {
 
 
     @Test
+    @Ignore
     public void testNGuideline() throws IOException {
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Empty Tabular Data.pdf");

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

@@ -1,15 +1,36 @@
 package com.iqser.red.service.redaction.v1.server.redaction.utils;
 
+import org.apache.commons.lang3.StringUtils;
 import org.junit.Test;
 
 import java.util.ArrayList;
 import java.util.List;
+import java.util.Set;
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 
+import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
+
 public class RegExPatternTest {
 
 
+    @Test
+    public void testExpand(){
+        String pattern = "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)";
+        String text = ", G.R., Simoneaux,";
+        Pattern compiledPattern = Pattern.compile(pattern, 0);
+        Matcher matcher = compiledPattern.matcher(text);
+
+        while (matcher.find()) {
+            String match = matcher.group(1);
+            if (StringUtils.isNotBlank(match)) {
+                System.out.println(match);
+            }
+        }
+    }
+
+
+
     @Test
     public void testEmailRegEx(){
         String text = "Address: Schwarzwaldalle " +

redaction-service-v1/redaction-service-server-v1/src/test/resources/dictionaries/false_positive.txt

@@ -1,10 +1,238 @@
-Long-term
+(Parent
+ADME - Bioanalyses, Vergèze,
+Aeration: Gentle
+Andrews mean
+Andrews pKi
+BASF study code
+Billington –
+Bond elute
+Brown Liquid
+Brown Orange
+Brown Solid
+Brown disco
+Brown discoloured
+Brown fat
 Brown liquid
 Brown solid
+Buehler (1980),
+Buehler (1980).
+Buehler 9
+Buehler Test.
+Buehler assay
+Buehler test
+Buehler)
+Buehler-like :
+Calandra lark
+Cary, NC,
+Chambersas for
+Class 4
+Class III’
+Class III’)
+Class method,
+Class2 (effectively
+ClassMethod) Vischim
+Clay =
+Cochran-Armitage Test.
+Cochran-armitage test.
+Dunnett 2
+Dunnett's "t''
+Dunnett's Test
+Dunnett's multiple
+Dunnett's t-test
+Dunnett's test
+Dunnett) **
+Dunnett) Dose
+Dunnett) Food
+Dunnett) Lower
+Dunnett, 1964)
+Dunnett’s Multiple
+Dunnett’s T
+Dunnett’s Test)
+Dunnett’s adjustment
+Dunnett’s multiple
+Dunnett’s post
+Dunnett’s t-test)
+Fine vacuolation
+Fine white
+Finewhite powder
+Fischer 344
+Fischer 344)
+Fischer F344
+Fischer rat
+Fischer) Contaminants:
+Fischer-344 acceptability
+Fischer-344/
+Fischer-344/DuCrj acceptability
+Fischer344 rats.
+Fischer344) :
+Fischerrat embryo
+Fischer’s exact
+Fisher 344
+Fisher Exact Test
+Fisher Exact test
+Fisher exact test
+Fisher's Exact
+Fisher-344 rat
+Fisher344 (F344/
+Fisher344 rat
+Fisher344 rats
+Fisher344 rats.
+Fisher344 rats/
+Fisher344) group
+Fisher’s exact.
+Freeman-Tukey transform
+Gentle aeration
+Grade 2:
+Grade 3)
+Grade 3:
+Green
+Green Alga
+Green as
+Green-, blue-,
+GreenAlga, Selenastrum
+GreenAlga.. Springborn
+Gross External
+Gross Necropsy
+Gross and
+Gross necropsies
+Gross necropsy
+Gross pathology
+Gross pathology:
+Hall Farm
+Hall Farm)
+HallFarm (396
+Ham's F12
+Ham’s F10
 Hand-held
+HarlanWinkelmann, Borchem,
+Hill International,
+Hill Top
+Hill criteria
+Hill, Bracknell,
+HillTop Chamber,
+Hillmodel was
+Hill’s model
+Japanese White
+Jerneplaque assay
+Kent. Blood
+Klimisch score
+Litchfield and Wilcoxon
+Long term
+Long-Term Toxicity
+Long-term
+Longbw/
+Major Macroscopic findings
+Major biliary
+Major defects
+Major deviations
+Major fetal
+Major metabolic
+Major metabolites
+Major route
+Major treatment-related
+Mallard (Anas
+Mallard with
 Manual-Hand held
-Manual-Hand held
-Weight:
+Marsh Frog
+Marsh harrier
+Masonjars (500
+Meier formula)
+Meier survival
+Miles anti-Canine
+Miles anti-dog
+Netherlands, 2011
+Netherlands, published:
+Netherlands. Report
+Penn State
+Porewater:
+Rattray study
+Read across
+Reed Bunting
+Reichert-Jung Cryo
+RifconGmbH. Oxon
+Ritz and
+S.P.-U/
+San Diego,
+Shaw's Farm,
 Sprague
+Strain JapaneseWhite
+Strong morphological
+Weight -
+Weight 2744
+Weight Gain
+Weight On
+Weight Range
+Weight and
 Weight and length
-Aeration: Gentle
+Weight at
+Weight gain
+Weight loss
+Weight of
+Weight only
+Weight range
+Weight, age
+Weight, feed
+Weight-of-evidence approach
+Weight:
+Weight; Adult
+Weight; Whole
+WeightGain 0-1
+WeightRange:
+Weightat dosing:
+Weightrange 200-250g.
+White (Tif:
+White 10
+White 3
+White Age
+White Powder
+White Strain
+White blood
+White cell
+White crystalline
+White guideline
+White guinea-pigs
+White hygroscopic
+White powder
+White solid
+White stork
+White swan
+White to off-white
+White wagtail
+White, solid
+WhitePowder Lot/
+Whiteguideline :
+Whitehygroscopic solid
+Whitepowder Lot/
+Whitepowder Stability
+Whitesolid Batch
+Whiteswan)
+Whiteto off-white
+Wildlife International specifications
+Wildlife Internationalspecifications Water:
+Williams E
+Williams medium
+Williams' medium
+WilliamsE medium
+Williams’ test
+Wilson's method.
+Wilson. All foetal
+Wilson.All foetal
+Wood pigeons
+York.
+Zeneca dossier);
+Zenecadossier); see
+green algae
+CTL
+No details reported
+Not given
+Not indicated
+Not reported
+Not reported.
+Not stated
+Syngenta
+Syngenta AG
+N/A
+No details reported
+Not available
+Test facility
+TBD

redaction-service-v1/redaction-service-server-v1/src/test/resources/drools/allAuthors.drl

@@ -0,0 +1,261 @@
+package drools
+
+import com.iqser.red.service.redaction.v1.server.redaction.model.Section
+
+global Section section
+
+
+// --------------------------------------- CBI rules -------------------------------------------------------------------
+
+rule "0: Expand CBI Authors with firstname initials"
+    when
+        Section(matchesType("CBI_author") || matchesType("recommendation_CBI_author"))
+    then
+        section.expandByRegEx("CBI_author", "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)", false, 1);
+        section.expandByRegEx("recommendation_CBI_author", "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)", false, 1);
+    end
+
+
+rule "1: Redact CBI Authors"
+    when
+        Section(matchesType("CBI_author") || matchesType("recommendation_CBI_author"))
+    then
+        section.redact("CBI_author", 1, "Author found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "2: Redact CBI Address"
+    when
+        Section(matchesType("CBI_address") || matchesType("recommendation_CBI_address"))
+    then
+        section.redact("CBI_address", 2, "Address found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "3: Do not redact genitive CBI_author"
+    when
+        Section(matchesType("CBI_author") || matchesType("recommendation_CBI_author"))
+    then
+        section.expandToHintAnnotationByRegEx("CBI_author", "['’’'ʼˈ´`‘′ʻ’']s", false, 0, "test_method");
+        section.expandToHintAnnotationByRegEx("recommendation_CBI_author", "['’’'ʼˈ´`‘′ʻ’']s", false, 0, "test_method");
+    end
+
+
+rule "4: Redact Author(s) cells in Tables with Author(s) header"
+    when
+        Section(hasTableHeader("Author(s)") && !hasTableHeader("Vertebrate study Y/N"))
+    then
+        section.redactCell("Author(s)", 4, "CBI_author", false, "Author found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "5: Redact Author cells in Tables with Author header"
+    when
+        Section(hasTableHeader("Author") && !hasTableHeader("Vertebrate study Y/N"))
+    then
+        section.redactCell("Author", 5, "CBI_author", false, "Author found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "6: Redact and recommand Authors in Tables with Vertebrate study Y/N header"
+    when
+        Section(rowEquals("Vertebrate study Y/N", "Y") || rowEquals("Vertebrate study Y/N", "Yes") || rowEquals("Vertebrate study Y/N", "N") || rowEquals("Vertebrate study Y/N", "No"))
+    then
+        section.redactCell("Author(s)", 6, "CBI_author", true, "Author found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "7: Redact if CTL/* or BL/*  was found"
+    when
+        Section(searchText.contains("CTL/") || searchText.contains("BL/"))
+    then
+        section.addRedaction("CTL", "must_redact", 7, "Laboratory for vertebrate studies found", "Article 39(1)(2) of Regulation (EC) No 178/2002" );
+        section.addRedaction("BL", "must_redact", 7, "Laboratory for vertebrate studies found", "Article 39(1)(2) of Regulation (EC) No 178/2002" );
+    end
+
+
+rule "8: Redact and add recommendation for et al. author"
+    when
+        Section(searchText.contains("et al"))
+    then
+		section.redactAndRecommendByRegEx("\\b([A-ZÄÖÜ][^\\s\\.,]+( [A-ZÄÖÜ]{1,2}\\.?)?( ?[A-ZÄÖÜ]\\.?)?) et al\\.?", false, 1, "CBI_author", 8, "Author found", "Reg (EC) No 1107/2009 Art. 63 (2g)");
+    end
+
+
+rule "9: Add recommendation for Addresses in Test Organism sections"
+    when
+        Section(searchText.contains("Species:") && searchText.contains("Source:"))
+    then
+		section.recommendLineAfter("Source:", "CBI_address");
+    end
+
+
+rule "10: Add recommendation for Addresses in Test Animals sections"
+    when
+        Section(searchText.contains("Species") && searchText.contains("Source"))
+    then
+		section.recommendLineAfter("Source", "CBI_address");
+    end
+
+
+rule "11: Do not redact Names and Addresses if Published Information found"
+    when
+        Section(matchesType("published_information"))
+    then
+        section.redactNot("CBI_author", 11, "Published Information found");
+        section.redactNot("CBI_address", 11, "Published Information found");
+    end
+
+
+// --------------------------------------- PII rules -------------------------------------------------------------------
+
+
+rule "12: Redacted PII Personal Identification Information"
+    when
+        Section(matchesType("PII"))
+    then
+        section.redact("PII", 12, "PII (Personal Identification Information) found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "13: Redact Emails by RegEx"
+    when
+        Section(searchText.contains("@"))
+    then
+        section.redactByRegEx("\\b[A-Z0-9._%+-]+@[A-Z0-9.-]+\\.[A-Z]{2,4}\\b", true, 0, "PII", 13, "PII (Personal Identification Information) found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "14: Redact contact information"
+    when
+       Section(text.contains("Contact point:")
+       || text.contains("Phone:")
+       || text.contains("Fax:")
+       || text.contains("Tel.:")
+       || text.contains("Tel:")
+       || text.contains("E-mail:")
+       || text.contains("Email:")
+       || text.contains("e-mail:")
+       || text.contains("E-mail address:")
+       || text.contains("Alternative contact:")
+       || text.contains("Telephone number:")
+       || text.contains("Telephone No:")
+       || text.contains("Fax number:")
+       || text.contains("Telephone:")
+       || text.contains("Phone No.")
+       || text.contains("European contact:"))
+    then
+          section.redactLineAfter("Contact point:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Phone:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Fax:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Tel.:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Tel:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("E-mail:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Email:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("e-mail:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("E-mail address:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Contact:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Alternative contact:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Telephone number:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Telephone No:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Fax number:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Telephone:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("Phone No.", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactBetween("No:", "Fax", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactBetween("Contact:", "Tel.:", "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+              section.redactLineAfter("European contact:",  "PII", 14, true, "Contact information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+     end
+
+
+rule "15: Redact contact information if applicant is found"
+    when
+        Section(headlineContainsWord("applicant") || text.contains("Applicant") || headlineContainsWord("Primary contact") || headlineContainsWord("Alternative contact") || text.contains("Telephone number:"))
+    then
+        section.redactLineAfter("Contact point:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Phone:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Fax:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Tel.:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Tel:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("E-mail:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Email:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("e-mail:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("E-mail address:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Contact:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Alternative contact:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Telephone number:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Telephone No:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Fax number:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Telephone:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Phone No.", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactBetween("No:", "Fax", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactBetween("Contact:", "Tel.:", "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("European contact:",  "PII", 15, true, "Applicant information was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "16: Redact contact information if Producer is found"
+    when
+        Section(text.toLowerCase().contains("producer of the plant protection") || text.toLowerCase().contains("producer of the active substance") || text.contains("Manufacturer of the active substance") || text.contains("Manufacturer:") || text.contains("Producer or producers of the active substance"))
+    then
+        section.redactLineAfter("Contact:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Telephone:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Phone:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Fax:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("E-mail:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Contact:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Fax number:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Telephone number:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Tel:", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactLineAfter("Phone No.", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+        section.redactBetween("No:", "Fax", "PII", 16, true, "Producer was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "17: Redact AUTHOR(S)"
+    when
+        Section(searchText.contains("AUTHOR(S):"))
+    then
+        section.redactLinesBetween("AUTHOR(S):", "COMPLETION DATE:", "PII", 17, true, "AUTHOR(S) was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "18: Redact PERFORMING LABORATORY"
+    when
+        Section(searchText.contains("PERFORMING LABORATORY:"))
+    then
+        section.redactBetween("PERFORMING LABORATORY:", "LABORATORY PROJECT ID:", "PII", 18, true, "PERFORMING LABORATORY was found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "19: Redact On behalf of Sequani Ltd.:"
+    when
+        Section(searchText.contains("On behalf of Sequani Ltd.: Name Title"))
+    then
+        section.redactBetween("On behalf of Sequani Ltd.: Name Title", "On behalf of", "PII", 19, false , "PII (Personal Identification Information) found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+rule "20: Redact On behalf of Syngenta Ltd.:"
+    when
+        Section(searchText.contains("On behalf of Syngenta Ltd.: Name Title"))
+    then
+        section.redactBetween("On behalf of Syngenta Ltd.: Name Title", "Study dates", "PII", 20, false , "PII (Personal Identification Information) found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end
+
+
+// --------------------------------------- other rules -------------------------------------------------------------------
+
+rule "21: Purity Hint"
+    when
+        Section(searchText.toLowerCase().contains("purity"))
+    then
+	    section.addHintAnnotationByRegEx("(purity ?( of|\\(.{1,20}\\))?( ?:)?) .{0,5}[\\d\\.]+( .{0,4}\\.)? ?%", true, 1, "hint_only");
+    end
+
+
+rule "22: Redact Must Redact"
+    when
+        Section(matchesType("must_redact"))
+    then
+        section.redact("must_redact", 22, "Must Redact found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+    end

redaction-service-v1/redaction-service-server-v1/src/test/resources/drools/rules.drl

@@ -7,6 +7,15 @@ global Section section
 
 // --------------------------------------- CBI rules -------------------------------------------------------------------
 
+rule "0: Expand CBI Authors with firstname initials"
+    when
+        Section(matchesType("CBI_author") || matchesType("recommendation_CBI_author"))
+    then
+        section.expandByRegEx("CBI_author", "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)", false, 1);
+        section.expandByRegEx("recommendation_CBI_author", "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)", false, 1);
+    end
+
+
 rule "1: Redacted because Section contains Vertebrate"
     when
         Section(matchesType("vertebrate"))
@@ -34,21 +43,21 @@ rule "3: Do not redact Names and Addresses if no redaction Indicator is containe
     end
 
 
-rule "4: Do not redact Names and Addresses if no redaction Indicator is contained"
-    when
-        Section(matchesType("vertebrate"), matchesType("published_information"))
-    then
-        section.redactNot("CBI_author", 4, "Vertebrate and Published Information found");
-        section.redactNot("CBI_address", 4, "Vertebrate and Published Information found");
-    end
-
-
-rule "5: Redact Names and Addresses if no_redaction_indicator and redaction_indicator is contained"
+rule "4: Redact Names and Addresses if no_redaction_indicator and redaction_indicator is contained"
     when
         Section(matchesType("vertebrate"), matchesType("no_redaction_indicator"), matchesType("redaction_indicator"))
     then
-        section.redact("CBI_author", 5, "Vertebrate and Redaction Indicator found", "Reg (EC) No 1107/2009 Art. 63 (2g)");
-        section.redact("CBI_address", 5, "Vertebrate and Redaction Indicator found", "Reg (EC) No 1107/2009 Art. 63 (2g)");
+        section.redact("CBI_author", 4, "Vertebrate and Redaction Indicator found", "Reg (EC) No 1107/2009 Art. 63 (2g)");
+        section.redact("CBI_address", 4, "Vertebrate and Redaction Indicator found", "Reg (EC) No 1107/2009 Art. 63 (2g)");
+    end
+
+
+rule "5: Do not redact Names and Addresses if no redaction Indicator is contained"
+    when
+        Section(matchesType("vertebrate"), matchesType("published_information"))
+    then
+        section.redactNot("CBI_author", 5, "Vertebrate and Published Information found");
+        section.redactNot("CBI_address", 5, "Vertebrate and Published Information found");
     end
 
 
@@ -134,33 +143,49 @@ rule "11: Redact if CTL/* or BL/*  was found"
     end
 
 
-rule "12: Add recommendation for et al. author"
+rule "12: Redact and add recommendation for et al. author"
     when
-        Section(searchText.contains("et al."))
+        Section(searchText.contains("et al"))
     then
-		section.addRecommendationByRegEx("\\b([A-ZÄÖÜ][^\\s\\.,]+( [A-ZÄÖÜ]\\.?)?( [A-ZÄÖÜ]\\.?)?) et al\\.?", false, 1, "CBI_author");
+		section.redactAndRecommendByRegEx("\\b([A-ZÄÖÜ][^\\s\\.,]+( [A-ZÄÖÜ]{1,2}\\.?)?( ?[A-ZÄÖÜ]\\.?)?) et al\\.?", false, 1, "CBI_author", 12, "Author found", "Reg (EC) No 1107/2009 Art. 63 (2g)");
+    end
+
+
+rule "13: Add recommendation for Addresses in Test Organism sections"
+    when
+        Section(searchText.contains("Species:") && searchText.contains("Source:"))
+    then
+		section.recommendLineAfter("Source:", "CBI_address");
+    end
+
+
+rule "14: Add recommendation for Addresses in Test Animals sections"
+    when
+        Section(searchText.contains("Species") && searchText.contains("Source"))
+    then
+		section.recommendLineAfter("Source", "CBI_address");
     end
 
 // --------------------------------------- PII rules -------------------------------------------------------------------
 
 
-rule "13: Redacted PII Personal Identification Information"
+rule "14: Redacted PII Personal Identification Information"
     when
         Section(matchesType("PII"))
     then
-        section.redact("PII", 13, "PII (Personal Identification Information) found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redact("PII", 14, "PII (Personal Identification Information) found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
     end
 
 
-rule "14: Redact Emails by RegEx"
+rule "15: Redact Emails by RegEx"
     when
         Section(searchText.contains("@"))
     then
-        section.redactByRegEx("\\b[A-Z0-9._%+-]+@[A-Z0-9.-]+\\.[A-Z]{2,4}\\b", true, 0, "PII", 14, "PII (Personal Identification Information) found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactByRegEx("\\b[A-Z0-9._%+-]+@[A-Z0-9.-]+\\.[A-Z]{2,4}\\b", true, 0, "PII", 15, "PII (Personal Identification Information) found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
     end
 
 
-rule "15: Redact contact information"
+rule "16: Redact contact information"
     when
        Section(text.contains("Contact point:")
        || text.contains("Phone:")
@@ -178,96 +203,105 @@ rule "15: Redact contact information"
        || text.contains("Telephone:")
        || text.contains("European contact:"))
     then
-          section.redactLineAfter("Contact point:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Phone:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Fax:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Tel.:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Tel:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("E-mail:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Email:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("e-mail:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("E-mail address:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Contact:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Alternative contact:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Telephone number:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Telephone No:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Fax number:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("Telephone:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactBetween("No:", "Fax", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactBetween("Contact:", "Tel.:", "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-              section.redactLineAfter("European contact:",  "PII", 15, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+          section.redactLineAfter("Contact point:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Phone:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Fax:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Tel.:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Tel:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("E-mail:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Email:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("e-mail:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("E-mail address:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Contact:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Alternative contact:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Telephone number:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Telephone No:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Fax number:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("Telephone:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactBetween("No:", "Fax", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactBetween("Contact:", "Tel.:", "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+              section.redactLineAfter("European contact:",  "PII", 16, true, "Contact information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
      end
 
 
-rule "16: Redact contact information if applicant is found"
+rule "17: Redact contact information if applicant is found"
     when
         Section(headlineContainsWord("applicant") || text.contains("Applicant") || headlineContainsWord("Primary contact") || headlineContainsWord("Alternative contact") || text.contains("Telephone number:"))
     then
-        section.redactLineAfter("Contact point:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Phone:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Fax:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Tel.:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Tel:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("E-mail:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Email:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("e-mail:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("E-mail address:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Contact:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Alternative contact:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Telephone number:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Telephone No:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Fax number:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Telephone:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactBetween("No:", "Fax", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactBetween("Contact:", "Tel.:", "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("European contact:",  "PII", 16, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Contact point:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Phone:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Fax:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Tel.:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Tel:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("E-mail:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Email:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("e-mail:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("E-mail address:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Contact:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Alternative contact:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Telephone number:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Telephone No:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Fax number:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Telephone:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactBetween("No:", "Fax", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactBetween("Contact:", "Tel.:", "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("European contact:",  "PII", 17, true, "Applicant information was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
     end
 
 
-rule "17: Redact contact information if Producer is found"
+rule "18: Redact contact information if Producer is found"
     when
         Section(text.toLowerCase().contains("producer of the plant protection") || text.toLowerCase().contains("producer of the active substance") || text.contains("Manufacturer of the active substance") || text.contains("Manufacturer:") || text.contains("Producer or producers of the active substance"))
     then
-        section.redactLineAfter("Contact:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Telephone:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Phone:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Fax:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("E-mail:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Contact:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Fax number:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Telephone number:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactLineAfter("Tel:", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
-        section.redactBetween("No:", "Fax", "PII", 17, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Contact:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Telephone:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Phone:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Fax:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("E-mail:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Contact:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Fax number:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Telephone number:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLineAfter("Tel:", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactBetween("No:", "Fax", "PII", 18, true, "Producer was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
     end
 
 
-rule "18: Redact AUTHOR(S)"
+rule "19: Redact AUTHOR(S)"
     when
         Section(searchText.contains("AUTHOR(S):"))
     then
-        section.redactLinesBetween("AUTHOR(S):", "COMPLETION DATE:", "PII", 18, true, "AUTHOR(S) was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactLinesBetween("AUTHOR(S):", "COMPLETION DATE:", "PII", 19, true, "AUTHOR(S) was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
     end
 
 
-rule "19: Redact PERFORMING LABORATORY"
+rule "20: Redact PERFORMING LABORATORY"
     when
         Section(searchText.contains("PERFORMING LABORATORY:"))
     then
-        section.redactBetween("PERFORMING LABORATORY:", "LABORATORY PROJECT ID:", "PII", 19, true, "PERFORMING LABORATORY was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactBetween("PERFORMING LABORATORY:", "LABORATORY PROJECT ID:", "PII", 20, true, "PERFORMING LABORATORY was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
     end
 
 
-rule "20: Redact On behalf of Sequani Ltd.:"
+rule "21: Redact On behalf of Sequani Ltd.:"
     when
         Section(searchText.contains("On behalf of Sequani Ltd.: Name Title"))
     then
-        section.redactBetween("On behalf of Sequani Ltd.: Name Title", "On behalf of", "PII", 20, false , "PII (Personal Identification Information) found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactBetween("On behalf of Sequani Ltd.: Name Title", "On behalf of", "PII", 21, false , "PII (Personal Identification Information) found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
     end
 
 
-rule "21: Redact On behalf of Syngenta Ltd.:"
+rule "22: Redact On behalf of Syngenta Ltd.:"
     when
         Section(searchText.contains("On behalf of Syngenta Ltd.: Name Title"))
     then
-        section.redactBetween("On behalf of Syngenta Ltd.: Name Title", "Study dates", "PII", 21, false , "PII (Personal Identification Information) found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+        section.redactBetween("On behalf of Syngenta Ltd.: Name Title", "Study dates", "PII", 22, false , "PII (Personal Identification Information) found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
+    end
+
+// --------------------------------------- other rules -------------------------------------------------------------------
+
+rule "25: Redact Purity"
+    when
+        Section(searchText.contains("purity"))
+    then
+	    section.redactByRegEx("purity ?:? (([\\d\\.]+)( .{0,4}\\.)? ?%)", true, 1, "purity", 17, "Purity found", "Reg (EC) No 1107/2009 Art. 63 (2a)");
     end

redaction-service-v1/redaction-service-server-v1/src/test/resources/log4j2-test.xml

@@ -10,7 +10,7 @@
         <Root level="warn">
             <AppenderRef ref="CONSOLE" />
         </Root>
-        <Logger name="com.iqser.gin4" level="info" />
+        <Logger name="com.iqser" level="info" />
     </Loggers>
 
 </Configuration>