Pull request #48 : RED-372: Suffix cross page annotations with -1, -2, etc., to avoid duplicate annotation ids

Merge in RED/redaction-service from RED-372 to master * commit 'ecdfd29803e09ac57cc74b19408db54d9a8247a7': RED-372: Suffix cross page annotations with -1, -2, etc., to avoid duplicate annotation ids
RED-372: Suffix cross page annotations with -1, -2, etc., to avoid duplicate annotation ids
2020-10-01 13:01:24 +02:00 · 2020-10-01 12:48:44 +02:00 · 2020-10-01 12:47:58 +02:00 · 2020-10-01 11:55:38 +02:00 · 2020-10-01 11:44:41 +02:00 · 2020-10-01 11:15:28 +02:00
59 changed files with 17019 additions and 532 deletions
@@ -5,7 +5,7 @@
  <parent>
    <groupId>com.atlassian.bamboo</groupId>
    <artifactId>bamboo-specs-parent</artifactId>
-    <version>7.0.4</version>
+    <version>7.1.2</version>
    <relativePath/>
  </parent>

@@ -34,4 +34,4 @@

  <!-- run 'mvn test' to perform offline validation of the plan -->
  <!-- run 'mvn -Ppublish-specs' to upload the plan to your Bamboo server -->
-</project>
+</project>
@@ -1,4 +1,4 @@
-FROM gin5/platform-base:5.2.0
+FROM red/base-image:1.0.0

 ARG PLATFORM_JAR

@@ -6,4 +6,4 @@ ENV PLATFORM_JAR ${PLATFORM_JAR}

 ENV USES_ELASTICSEARCH false

-COPY ["${PLATFORM_JAR}", "/"]
+COPY ["${PLATFORM_JAR}", "/"]
@@ -5,7 +5,7 @@
    <parent>
        <artifactId>platform-dependency</artifactId>
        <groupId>com.iqser.red</groupId>
-        <version>1.0.1</version>
+        <version>1.0.2</version>
    </parent>
    <modelVersion>4.0.0</modelVersion>

@@ -22,7 +22,7 @@
    </modules>

    <properties>
-        <pdfbox.version>2.0.16</pdfbox.version>
+        <pdfbox.version>2.0.21</pdfbox.version>
    </properties>


@@ -32,21 +32,21 @@
            <dependency>
                <groupId>com.iqser.red</groupId>
                <artifactId>platform-commons-dependency</artifactId>
-                <version>1.0.0</version>
+                <version>1.1.0</version>
                <scope>import</scope>
                <type>pom</type>
            </dependency>

-                <dependency>
-                    <groupId>org.apache.pdfbox</groupId>
-                    <artifactId>pdfbox</artifactId>
-                    <version>${pdfbox.version}</version>
-                </dependency>
-                <dependency>
-                    <groupId>org.apache.pdfbox</groupId>
-                    <artifactId>pdfbox-tools</artifactId>
-                    <version>${pdfbox.version}</version>
-                </dependency>
+            <dependency>
+                <groupId>org.apache.pdfbox</groupId>
+                <artifactId>pdfbox</artifactId>
+                <version>${pdfbox.version}</version>
+            </dependency>
+            <dependency>
+                <groupId>org.apache.pdfbox</groupId>
+                <artifactId>pdfbox-tools</artifactId>
+                <version>${pdfbox.version}</version>
+            </dependency>

        </dependencies>

@@ -11,6 +11,19 @@ import lombok.NoArgsConstructor;
@NoArgsConstructor
 public class RedactionLog {

+    public RedactionLog(List<RedactionLogEntry> redactionLogEntry, long dictionaryVersion, long rulesVersion) {
+
+        this.redactionLogEntry = redactionLogEntry;
+        this.dictionaryVersion = dictionaryVersion;
+        this.rulesVersion = rulesVersion;
+    }
+
    private List<RedactionLogEntry> redactionLogEntry;

+    private long dictionaryVersion = -1;
+
+    private long rulesVersion = -1;
+
+    private String filename;
+
 }
@@ -11,25 +11,6 @@

    <artifactId>redaction-service-server-v1</artifactId>

-    <properties>
-        <pdfbox.version>2.0.20</pdfbox.version>
-    </properties>
-
-    <dependencyManagement>
-        <dependencies>
-            <dependency>
-                <groupId>org.apache.pdfbox</groupId>
-                <artifactId>pdfbox</artifactId>
-                <version>${pdfbox.version}</version>
-            </dependency>
-            <dependency>
-                <groupId>org.apache.pdfbox</groupId>
-                <artifactId>pdfbox-tools</artifactId>
-                <version>${pdfbox.version}</version>
-            </dependency>
-        </dependencies>
-    </dependencyManagement>
-
    <dependencies>
        <dependency>
            <groupId>com.iqser.red.service</groupId>
@@ -63,15 +44,15 @@
        </dependency>
        <!-- commons -->
        <dependency>
-            <groupId>com.iqser.gin4.commons</groupId>
+            <groupId>com.iqser.red.commons</groupId>
            <artifactId>spring-commons</artifactId>
        </dependency>
        <dependency>
-            <groupId>com.iqser.gin4.commons</groupId>
+            <groupId>com.iqser.red.commons</groupId>
            <artifactId>logging-commons</artifactId>
        </dependency>
        <dependency>
-            <groupId>com.iqser.gin4.commons</groupId>
+            <groupId>com.iqser.red.commons</groupId>
            <artifactId>metric-commons</artifactId>
        </dependency>
        <!-- other external -->
@@ -104,7 +85,7 @@
            <scope>test</scope>
        </dependency>
        <dependency>
-            <groupId>com.iqser.gin4.commons</groupId>
+            <groupId>com.iqser.red.commons</groupId>
            <artifactId>test-commons</artifactId>
            <scope>test</scope>
        </dependency>
@@ -20,7 +20,7 @@ import org.springframework.cloud.openfeign.EnableFeignClients;
 import org.springframework.context.annotation.Bean;
 import org.springframework.context.annotation.Import;

-import com.iqser.gin4.commons.spring.DefaultWebMvcConfiguration;
+import com.iqser.red.commons.spring.DefaultWebMvcConfiguration;
 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;
@@ -84,7 +84,7 @@ public class ClassificationService {
        }
        else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter().getMostPopular() && textBlock.getMostPopularWordStyle().equals("italic") && !document.getFontStyleCounter().getMostPopular().equals("italic") && PositionUtils.getApproxLineCount(textBlock) < 2.9) {
            textBlock.setClassification("TextBlock Italic");
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getSequences().size() > 3){
+        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock)){
            textBlock.setClassification("TextBlock Unknown");
        }
    }
@@ -2,13 +2,13 @@ package com.iqser.red.service.redaction.v1.server.controller;

 import java.time.OffsetDateTime;

+import com.iqser.red.commons.spring.ErrorMessage;
 import org.springframework.http.HttpStatus;
 import org.springframework.web.bind.annotation.ExceptionHandler;
 import org.springframework.web.bind.annotation.ResponseBody;
 import org.springframework.web.bind.annotation.ResponseStatus;
 import org.springframework.web.bind.annotation.RestControllerAdvice;

-import com.iqser.gin4.commons.api.errorhandling.ErrorMessage;
 import com.iqser.red.service.redaction.v1.server.exception.RulesValidationException;

 import lombok.extern.slf4j.Slf4j;
@@ -3,18 +3,21 @@ package com.iqser.red.service.redaction.v1.server.controller;
 import java.io.ByteArrayInputStream;
 import java.io.ByteArrayOutputStream;
 import java.io.IOException;
+import java.util.List;

 import org.apache.pdfbox.pdmodel.PDDocument;
 import org.springframework.web.bind.annotation.RequestBody;
 import org.springframework.web.bind.annotation.RestController;

 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.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.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.segmentation.PdfSegmentationService;
@@ -36,6 +39,7 @@ public class RedactionController implements RedactionResource {
    private final EntityRedactionService entityRedactionService;
    private final PdfFlattenService pdfFlattenService;
    private final DroolsExecutionService droolsExecutionService;
+    private final DictionaryService dictionaryService;

    @Override
    public RedactionResult redact(@RequestBody RedactionRequest redactionRequest) {
@@ -49,10 +53,10 @@ public class RedactionController implements RedactionResource {

            if (redactionRequest.isFlatRedaction()) {
                PDDocument flatDocument = pdfFlattenService.flattenPDF(pdDocument);
-                return convert(flatDocument, classifiedDoc.getPages().size(), new RedactionLog(classifiedDoc.getRedactionLogEntities()));
+                return convert(flatDocument, classifiedDoc.getPages().size(), classifiedDoc.getRedactionLogEntities());
            }

-            return convert(pdDocument, classifiedDoc.getPages().size(), new RedactionLog(classifiedDoc.getRedactionLogEntities()));
+            return convert(pdDocument, classifiedDoc.getPages().size(), classifiedDoc.getRedactionLogEntities());

        } catch (IOException e) {
            throw new RedactionException(e);
@@ -130,14 +134,14 @@ public class RedactionController implements RedactionResource {
        return convert(document, numberOfPages, null);
    }

-    private RedactionResult convert(PDDocument document, int numberOfPages, RedactionLog redactionLog) throws IOException {
+    private RedactionResult convert(PDDocument document, int numberOfPages, List<RedactionLogEntry> redactionLogEntities) throws IOException {

        try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream()) {
            document.save(byteArrayOutputStream);
            return RedactionResult.builder()
                    .document(byteArrayOutputStream.toByteArray())
                    .numberOfPages(numberOfPages)
-                    .redactionLog(redactionLog)
+                    .redactionLog(new RedactionLog(redactionLogEntities,  dictionaryService.getDictionaryVersion(), droolsExecutionService.getRulesVersion()))
                    .build();
        }

@@ -44,10 +44,10 @@ import lombok.extern.slf4j.Slf4j;
 public class PDFLinesTextStripper extends PDFTextStripper {

    @Getter
-    private float minCharWidth = Float.MAX_VALUE;
+    private int maxCharWidths;

    @Getter
-    private float minCharHeight = Float.MAX_VALUE;
+    private int maxCharHeight;

    @Getter
    private final List<TextPositionSequence> textPositionSequences = new ArrayList<>();
@@ -201,8 +201,16 @@ public class PDFLinesTextStripper extends PDFTextStripper {

        int startIndex = 0;
        for (int i = 0; i <= textPositions.size() - 1; i++) {
-            minCharWidth = Math.min(minCharWidth, textPositions.get(i).getWidthDirAdj());
-            minCharHeight = Math.min(minCharHeight, textPositions.get(i).getHeightDir());
+
+            int charHeight = (int) textPositions.get(i).getHeightDir();
+            if(charHeight > maxCharHeight){
+                maxCharHeight = charHeight;
+            }
+
+            int charWidth = (int) textPositions.get(i).getWidthDirAdj();
+            if(charWidth > maxCharWidths){
+                maxCharWidths = charWidth;
+            }

            if (i == 0 && textPositions.get(i).getUnicode().equals(" ")) {
                startIndex++;
@@ -241,8 +249,8 @@ public class PDFLinesTextStripper extends PDFTextStripper {
    @Override
    public String getText(PDDocument doc) throws IOException {

-        minCharWidth = Float.MAX_VALUE;
-        minCharHeight = Float.MAX_VALUE;
+        maxCharWidths = 0;
+        maxCharWidths = 0;
        textPositionSequences.clear();
        rulings.clear();
        graphicsPath.clear();
@@ -17,6 +17,6 @@ public class ParsedElements {
    private boolean landscape;
    private boolean rotated;

-    private float minCharWidth;
-    private float minCharHeight;
+    private float maxCharWidth;
+    private float maxCharHeight;
 }
@@ -9,9 +9,7 @@ import com.iqser.red.service.redaction.v1.model.Point;
 import com.iqser.red.service.redaction.v1.model.Rectangle;

 import lombok.Data;
-import lombok.Getter;
 import lombok.RequiredArgsConstructor;
-import lombok.Setter;

@Data
@RequiredArgsConstructor
@@ -19,10 +17,6 @@ public class TextPositionSequence implements CharSequence {

    private List<TextPosition> textPositions = new ArrayList<>();

-    @Getter
-    @Setter
-    private float[] annotationColor;
-
    private final int page;


@@ -194,22 +188,18 @@ public class TextPositionSequence implements CharSequence {

    public Rectangle getRectangle() {

-        float height = textPositions.get(0).getHeightDir() + 2;
+        float height = getTextHeight();

-        float posXInit;
+        float posXInit = getX1();
        float posXEnd;
        float posYInit;
        float posYEnd;

        if (textPositions.get(0).getRotation() == 90) {
-
            posXEnd = textPositions.get(0).getYDirAdj() + 2;
-            posXInit = textPositions.get(0).getYDirAdj() - height;
-            posYInit = textPositions.get(0).getXDirAdj();
+            posYInit = getY1();
            posYEnd = textPositions.get(textPositions.size() - 1).getXDirAdj() - height + 4;
        } else {
-
-            posXInit = textPositions.get(0).getXDirAdj();
            posXEnd = textPositions.get(textPositions.size() - 1)
                    .getXDirAdj() + textPositions.get(textPositions.size() - 1).getWidth() + 1;
            posYInit = textPositions.get(0).getPageHeight() - textPositions.get(0).getYDirAdj() - 2;
@@ -220,4 +210,5 @@ public class TextPositionSequence implements CharSequence {
        return new Rectangle(new Point(posXInit, posYInit), posXEnd - posXInit, posYEnd - posYInit + height, page);
    }

+
 }
@@ -0,0 +1,43 @@
+package com.iqser.red.service.redaction.v1.server.redaction.model;
+
+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.utils.TextNormalizationUtilities;
+
+import lombok.RequiredArgsConstructor;
+import lombok.Value;
+
+@Value
+@RequiredArgsConstructor
+public class CellValue {
+
+    TextBlock textBlock;
+
+    int rowSpanStart;
+
+    @Override
+    public String toString() {
+
+        StringBuilder sb = new StringBuilder();
+
+
+            TextPositionSequence previous = null;
+            for (TextPositionSequence word : textBlock.getSequences()) {
+
+                if (previous != null) {
+                    if (Math.abs(previous.getY1() - word.getY1()) > word.getTextHeight()) {
+                        sb.append('\n');
+                    } else {
+                        sb.append(' ');
+                    }
+                }
+                sb.append(word.toString());
+                previous = word;
+            }
+
+        return TextNormalizationUtilities.removeHyphenLineBreaks(sb.toString())
+                .replaceAll("\n", " ")
+                .replaceAll(" {2}", " ");
+    }
+
+}
@@ -3,6 +3,8 @@ package com.iqser.red.service.redaction.v1.server.redaction.model;
 import java.util.ArrayList;
 import java.util.List;

+import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
+
 import lombok.Data;
 import lombok.EqualsAndHashCode;

@@ -16,6 +18,7 @@ public class Entity {
    private boolean redaction;
    private String redactionReason;
    private List<EntityPositionSequence> positionSequences = new ArrayList<>();
+    private List<TextPositionSequence> targetSequences;
    private Integer start;
    private Integer end;

@@ -1,6 +1,7 @@
 package com.iqser.red.service.redaction.v1.server.redaction.model;

 import java.util.ArrayList;
+import java.util.Collections;
 import java.util.List;
 import java.util.regex.Pattern;

@@ -25,9 +26,16 @@ public class SearchableText {
    }


-    @SuppressWarnings("checkstyle:ModifiedControlVariable")
    public List<EntityPositionSequence> getSequences(String searchString, boolean caseInsensitive) {

+        return getSequences(searchString, caseInsensitive, null);
+
+    }
+
+    @SuppressWarnings("checkstyle:ModifiedControlVariable")
+    public List<EntityPositionSequence> getSequences(String searchString, boolean caseInsensitive,
+                                                     List<TextPositionSequence> sequencesSubList) {
+
        String normalizedSearchString;
        if (caseInsensitive) {
            normalizedSearchString = searchString.toLowerCase();
@@ -40,37 +48,50 @@ public class SearchableText {

        List<TextPositionSequence> crossSequenceParts = new ArrayList<>();
        List<EntityPositionSequence> finalMatches = new ArrayList<>();
-        for (int i = 0; i < sequences.size(); i++) {
-            TextPositionSequence partMatch = new TextPositionSequence(sequences.get(i).getPage());
-            for (int j = 0; j < sequences.get(i).length(); j++) {

-                if (i > 0 && j == 0 && sequences.get(i).charAt(0, caseInsensitive) == ' ' && sequences.get(i - 1)
-                        .charAt(sequences.get(i - 1).length() - 1, caseInsensitive) == ' ' || j > 0 && sequences.get(i)
-                        .charAt(j, caseInsensitive) == ' ' && sequences.get(i).charAt(j - 1, caseInsensitive) == ' ') {
-                    if (j == sequences.get(i).length() - 1 && counter != 0 && !partMatch.getTextPositions().isEmpty()) {
+        List<TextPositionSequence> searchSpace;
+        if (sequencesSubList != null) {
+            int subListIndex = Collections.indexOfSubList(sequences, sequencesSubList);
+            if (subListIndex != -1) {
+                searchSpace = sequences.subList(subListIndex, subListIndex + sequencesSubList.size());
+            } else {
+                searchSpace = sequences;
+            }
+        } else {
+            searchSpace = sequences;
+        }
+
+        for (int i = 0; i < searchSpace.size(); i++) {
+            TextPositionSequence partMatch = new TextPositionSequence(searchSpace.get(i).getPage());
+            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()) {
                        crossSequenceParts.add(partMatch);
                    }
                    continue;
                }

-                if (j == 0 && sequences.get(i).charAt(j, caseInsensitive) != ' ' && i != 0 && sequences.get(i - 1)
-                        .charAt(sequences.get(i - 1)
+                if (j == 0 && searchSpace.get(i).charAt(j, caseInsensitive) != ' ' && i != 0 && searchSpace.get(i - 1)
+                        .charAt(searchSpace.get(i - 1)
                                .length() - 1, caseInsensitive) != ' ' && searchChars[counter] == ' ') {
                    counter++;
                }

-                if (sequences.get(i)
-                        .charAt(j, caseInsensitive) == searchChars[counter] || counter != 0 && sequences.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(sequences.get(i)
-                            .charAt(j - 1, caseInsensitive)) || j == 0 && i != 0 && isSeparator(sequences.get(i - 1)
-                            .charAt(sequences.get(i - 1)
-                                    .length() - 1, caseInsensitive)) || j == 0 && i != 0 && sequences.get(i - 1)
-                            .charAt(sequences.get(i - 1).length() - 1, caseInsensitive) != ' ' && sequences.get(i)
+                    if (counter != 0 || i == 0 && j == 0 || j != 0 && isSeparator(searchSpace.get(i)
+                            .charAt(j - 1, caseInsensitive)) || j == 0 && i != 0 && isSeparator(searchSpace.get(i - 1)
+                            .charAt(searchSpace.get(i - 1)
+                                    .length() - 1, caseInsensitive)) || j == 0 && i != 0 && searchSpace.get(i - 1)
+                            .charAt(searchSpace.get(i - 1).length() - 1, caseInsensitive) != ' ' && searchSpace.get(i)
                            .charAt(j, caseInsensitive) != ' ') {
-                        partMatch.add(sequences.get(i).textPositionAt(j));
-                        if (!(j == sequences.get(i).length() - 1 && sequences.get(i)
+                        partMatch.add(searchSpace.get(i).textPositionAt(j));
+                        if (!(j == searchSpace.get(i).length() - 1 && searchSpace.get(i)
                                .charAt(j, caseInsensitive) == '-' && searchChars[counter] != '-')) {
                            counter++;
                        }
@@ -79,19 +100,19 @@ public class SearchableText {
                    if (counter == searchString.length()) {
                        crossSequenceParts.add(partMatch);

-                        if (i == sequences.size() - 1 && j == sequences.get(i).length() - 1 || j != sequences.get(i)
-                                .length() - 1 && isSeparator(sequences.get(i)
-                                .charAt(j + 1, caseInsensitive)) || j == sequences.get(i)
-                                .length() - 1 && isSeparator(sequences.get(i + 1)
-                                .charAt(0, caseInsensitive)) || j == sequences.get(i).length() - 1 && sequences.get(i)
-                                .charAt(j, caseInsensitive) != ' ' && sequences.get(i + 1)
+                        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(j, caseInsensitive) != ' ' && searchSpace.get(i + 1)
                                .charAt(0, caseInsensitive) != ' ') {
                            finalMatches.addAll(buildEntityPositionSequence(crossSequenceParts));
                        }

                        counter = 0;
                        crossSequenceParts = new ArrayList<>();
-                        partMatch = new TextPositionSequence(sequences.get(i).getPage());
+                        partMatch = new TextPositionSequence(searchSpace.get(i).getPage());
                    }
                } else {
                    counter = 0;
@@ -99,24 +120,27 @@ public class SearchableText {
                        j--;
                    }
                    crossSequenceParts = new ArrayList<>();
-                    partMatch = new TextPositionSequence(sequences.get(i).getPage());
+                    partMatch = new TextPositionSequence(searchSpace.get(i).getPage());
                }

-                if (j == sequences.get(i).length() - 1 && counter != 0) {
+                if (j == searchSpace.get(i).length() - 1 && counter != 0) {
                    crossSequenceParts.add(partMatch);
                }
            }
        }

        return finalMatches;
+
    }


    private List<EntityPositionSequence> buildEntityPositionSequence(List<TextPositionSequence> crossSequenceParts) {

-        String id = IdBuilder.buildId(crossSequenceParts);
+        String plainId = IdBuilder.buildId(crossSequenceParts);
+        String id = plainId;
        List<EntityPositionSequence> result = new ArrayList<>();
        int currentPage = -1;
+        int idDiffentPageSuffix = 1;
        EntityPositionSequence entityPositionSequence = new EntityPositionSequence(id);
        for (TextPositionSequence textPositionSequence : crossSequenceParts) {
            if (currentPage == -1) {
@@ -126,9 +150,13 @@ public class SearchableText {
            } else if (currentPage == textPositionSequence.getPage()) {
                entityPositionSequence.getSequences().add(textPositionSequence);
            } else {
+                id = plainId + "-" + idDiffentPageSuffix;
+                idDiffentPageSuffix++;
                result.add(entityPositionSequence);
                entityPositionSequence = new EntityPositionSequence(id);
                entityPositionSequence.setPageNumber(textPositionSequence.getPage());
+                entityPositionSequence.getSequences().add(textPositionSequence);
+                currentPage = textPositionSequence.getPage();
            }
        }
        result.add(entityPositionSequence);
@@ -7,7 +7,6 @@ import java.util.Map;
 import java.util.Set;
 import java.util.regex.Pattern;

-import org.apache.commons.collections4.CollectionUtils;
 import org.apache.commons.lang3.StringUtils;

 import lombok.Builder;
@@ -31,10 +30,21 @@ public class Section {

    private int sectionNumber;

-    private Map<String, String> tabularData;
+    private Map<String, CellValue> tabularData;


-    public boolean contains(String type) {
+    public boolean rowEquals(String headerName, String value) {
+
+        String cleanHeaderName = headerName.replaceAll("\n", "").replaceAll(" ", "").replaceAll("-", "");
+
+        return tabularData != null && tabularData.containsKey(cleanHeaderName) && tabularData.get(cleanHeaderName)
+                .getTextBlock()
+                .getText()
+                .equals(value);
+    }
+
+
+    public boolean matchesType(String type) {

        return entities.stream().anyMatch(entity -> entity.getType().equals(type));
    }
@@ -70,6 +80,18 @@ public class Section {
    }


+    public void redactIfPrecededBy(String prefix, String type, int ruleNumber, String reason) {
+
+        entities.forEach(entity -> {
+            if (entity.getType().equals(type) && searchText.indexOf(prefix + entity.getWord()) != 1) {
+                entity.setRedaction(true);
+                entity.setMatchedRule(ruleNumber);
+                entity.setRedactionReason(reason);
+            }
+        });
+    }
+
+
    public void redactLineAfter(String start, String asType, int ruleNumber, String reason) {

        String[] values = StringUtils.substringsBetween(text, start, "\n");
@@ -131,7 +153,8 @@ public class Section {

            if (startIndex > -1 && (startIndex == 0 || Character.isWhitespace(searchText.charAt(startIndex - 1)) || isSeparator(searchText
                    .charAt(startIndex - 1))) && (stopIndex == searchText.length() || isSeparator(searchText.charAt(stopIndex)))) {
-                found.add(new Entity(searchText.substring(startIndex, stopIndex), asType, startIndex, stopIndex, headline, sectionNumber));
+                found.add(new Entity(searchText.substring(startIndex, stopIndex), asType, startIndex, stopIndex,
+                        headline, sectionNumber));
            }
        } while (startIndex > -1);

@@ -161,24 +184,49 @@ public class Section {
    }


-    public void highlightCell(String cellHeader, int ruleNumber) {
+    public void highlightCell(String cellHeader, int ruleNumber, String type) {

-        String value = tabularData.get(cellHeader);
+        annotateCell(cellHeader, ruleNumber, type, false, null);
+    }
+
+
+    public void redactCell(String cellHeader, int ruleNumber, String type, String reason) {
+
+        annotateCell(cellHeader, ruleNumber, type, true, reason);
+    }
+
+
+    public void redactNotCell(String cellHeader, int ruleNumber, String type, String reason) {
+
+        annotateCell(cellHeader, ruleNumber, type, false, reason);
+    }
+
+
+    private void annotateCell(String cellHeader, int ruleNumber, String type, boolean redact, String reason) {
+
+        String cleanHeaderName = cellHeader.replaceAll("\n", "").replaceAll(" ", "").replaceAll("-", "");
+
+        CellValue value = tabularData.get(cleanHeaderName);
        if (value == null) {
            log.warn("Could not find any data for {}.", cellHeader);
        } else {
-            Set<Entity> found = findEntities(value, "must_redact");
-            if (CollectionUtils.isEmpty(found)) {
-                log.warn("Could not identify value {} in row.", value);
-            } else {
-                Entity entity = found.iterator().next();
-                entity.setRedaction(false);
-                entity.setMatchedRule(ruleNumber);
-                entity.setRedactionReason(cellHeader);
-                entities.add(entity);
-            }
-        }
+            Entity entity = new Entity(value.toString(), type, value.getRowSpanStart(),
+                    value.getRowSpanStart() + value.toString()
+                    .length(), headline, sectionNumber);
+            entity.setRedaction(redact);
+            entity.setMatchedRule(ruleNumber);
+            entity.setRedactionReason(reason);
+            entity.setTargetSequences(value.getTextBlock()
+                    .getSequences()); // Make sure no other cells with same content are highlighted

+            // HashSet keeps the older value, but we want the new only.
+            if (entities.contains(entity)) {
+                entities.remove(entity);
+            }
+            entities.add(entity);
+
+            entities = removeEntitiesContainedInLarger(entities);
+        }
    }

-}
+}
@@ -27,6 +27,7 @@ public class DictionaryService {

    private final DictionaryClient dictionaryClient;

+    @Getter
    private long dictionaryVersion = -1;

    @Getter
@@ -18,6 +18,7 @@ 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;

@Service
@@ -29,6 +30,7 @@ public class DroolsExecutionService {
    @Autowired
    private KieContainer kieContainer;

+    @Getter
    private long rulesVersion = -1;

    public Section executeRules(Section section) {
@@ -18,6 +18,7 @@ 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.Paragraph;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
 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.SearchableText;
@@ -26,9 +27,7 @@ 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 lombok.extern.slf4j.Slf4j;

-@Slf4j
@Service
@RequiredArgsConstructor
 public class EntityRedactionService {
@@ -51,24 +50,33 @@ public class EntityRedactionService {
            List<Table> tables = paragraph.getTables();

            for (Table table : tables) {
-                List<String> metadata = table.getHeaders();
+                boolean singleCellTable = table.getRowCount() == 1 && table.getColCount() == 1;
                for (List<Cell> row : table.getRows()) {
                    SearchableText searchableRow = new SearchableText();
-                    List<String> cellValues = new ArrayList<>();
+                    Map<String, CellValue> tabularData = new HashMap<>();
+                    int start = 0;
                    for (Cell cell : row) {
-                        if (cell == null || CollectionUtils.isEmpty(cell.getTextBlocks())) {
-                            cellValues.add(null);
+                        if (!singleCellTable && cell.isHeaderCell() || CollectionUtils.isEmpty(cell.getTextBlocks())) {
                            continue;
                        }
-                        cellValues.add(cell.getTextBlocks().get(0).getText());
                        addSectionToManualRedactions(cell.getTextBlocks(), manualRedactions, table.getHeadline(), sectionNumber);
+                        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().get(0), cellStart));
+                        });
+                        start = start + cell.toString().length();
                        for (TextBlock textBlock : cell.getTextBlocks()) {
                            searchableRow.addAll(textBlock.getSequences());
                        }
                    }
                    Set<Entity> rowEntities = findEntities(searchableRow, table.getHeadline(), sectionNumber);

-                    Map<String, String> tabularData = toMap(metadata, cellValues);
                    Section analysedRowSection = droolsExecutionService.executeRules(Section.builder()
                            .entities(rowEntities)
                            .text(searchableRow.getAsStringWithLinebreaks())
@@ -116,35 +124,15 @@ public class EntityRedactionService {
    }


-    private Map<String, String> toMap(List<String> keys, List<String> values) {
-
-        if (keys.size() != values.size()) {
-            log.warn("Cannot merge lists of unequal size, returning empty map.");
-            return new HashMap<>();
-        }
-        Map<String, String> result = new HashMap<>();
-        for (int i = 0; i < keys.size(); i++) {
-            String value = values.get(i);
-            if (value == null) {
-                continue;
-            }
-            result.put(keys.get(i), value);
-        }
-
-        return result;
-
-    }
-
-
    private Set<Entity> clearAndFindPositions(Set<Entity> entities, SearchableText text) {

        removeEntitiesContainedInLarger(entities);

        for (Entity entity : entities) {
            if (dictionaryService.getCaseInsensitiveTypes().contains(entity.getType())) {
-                entity.setPositionSequences(text.getSequences(entity.getWord(), true));
+                entity.setPositionSequences(text.getSequences(entity.getWord(), true, entity.getTargetSequences()));
            } else {
-                entity.setPositionSequences(text.getSequences(entity.getWord(), false));
+                entity.setPositionSequences(text.getSequences(entity.getWord(), false, entity.getTargetSequences()));
            }
        }

@@ -155,7 +143,7 @@ public class EntityRedactionService {
    private Set<Entity> findEntities(SearchableText searchableText, String headline, int sectionNumber) {

        Set<Entity> found = new HashSet<>();
-        if (StringUtils.isEmpty(searchableText.toString()) && StringUtils.isEmpty(headline)) {
+        if (StringUtils.isEmpty(searchableText.toString())) {
            return found;
        }

@@ -21,7 +21,6 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractT
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.CleanRulings;
 import com.iqser.red.service.redaction.v1.server.tableextraction.service.RulingCleaningService;
 import com.iqser.red.service.redaction.v1.server.tableextraction.service.TableExtractionService;
-import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;

 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
@@ -37,6 +36,7 @@ public class PdfSegmentationService {
    private final ClassificationService classificationService;
    private final SectionsBuilderService sectionsBuilderService;

+
    public Document parseDocument(PDDocument pdDocument) throws IOException {

        Document document = new Document();
@@ -56,19 +56,21 @@ public class PdfSegmentationService {
            int rotation = pdPage.getRotation();
            boolean isRotated = rotation != 0 && rotation != 360;

-            ParsedElements parsedElements = ParsedElements
-                    .builder()
+
+            ParsedElements parsedElements = ParsedElements.builder()
                    .rulings(stripper.getRulings())
                    .sequences(stripper.getTextPositionSequences())
-                    .minCharWidth(Utils.round(stripper.getMinCharWidth(), 2))
-                    .minCharHeight(Utils.round(stripper.getMinCharHeight(), 2))
+                    .maxCharWidth(stripper.getMaxCharWidths())
+                    .maxCharHeight(stripper.getMaxCharWidths())
                    .landscape(isLandscape)
                    .rotated(isRotated)
                    .build();

-            CleanRulings cleanRulings = rulingCleaningService.getCleanRulings(parsedElements.getRulings(), parsedElements.getMinCharWidth(), parsedElements.getMinCharHeight());
+            CleanRulings cleanRulings = rulingCleaningService.getCleanRulings(parsedElements.getRulings(), parsedElements
+                    .getMaxCharWidth(), parsedElements.getMaxCharHeight());

-            Page page = blockificationService.blockify(parsedElements.getSequences(), cleanRulings.getHorizontal(), cleanRulings.getVertical());
+            Page page = blockificationService.blockify(parsedElements.getSequences(), cleanRulings.getHorizontal(), cleanRulings
+                    .getVertical());
            page.setRotation(rotation);

            tableExtractionService.extractTables(cleanRulings, page);
@@ -91,7 +93,10 @@ public class PdfSegmentationService {

    }

+
+
    private void increaseDocumentStatistics(Page page, Document document) {
+
        if (!page.isLandscape()) {
            document.getFontSizeCounter().addAll(page.getFontSizeCounter().getCountPerValue());
        }
@@ -100,6 +105,7 @@ public class PdfSegmentationService {
        document.getFontStyleCounter().addAll(page.getFontStyleCounter().getCountPerValue());
    }

+
    private void buildPageStatistics(Page page) {

        // Collect all statistics for the page, except from blocks inside tables, as tables will always be added to BodyTextFrame.
@@ -1,11 +1,12 @@
 package com.iqser.red.service.redaction.v1.server.segmentation;

 import java.util.ArrayList;
+import java.util.Collections;
 import java.util.Iterator;
 import java.util.List;
+import java.util.stream.Collectors;

 import org.apache.commons.collections4.CollectionUtils;
-import org.apache.commons.lang3.StringUtils;
 import org.springframework.stereotype.Service;

 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
@@ -13,6 +14,7 @@ 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.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;

@Service
@@ -38,22 +40,29 @@ public class SectionsBuilderService {
                current.setPage(page.getPageNumber());

                if (prev != null && current.getClassification().startsWith("H ") || !document.isHeadlines()) {
-                    Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline, previousTable);
+                    Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
                    chunkBlock.setHeadline(lastHeadline);
-                    lastHeadline = current.getText();
-                    if (CollectionUtils.isNotEmpty(chunkBlock.getTables())) {
-                        previousTable = chunkBlock.getTables().get(0);
+                    if(document.isHeadlines()) {
+                        lastHeadline = current.getText();
                    }
                    chunkBlockList.add(chunkBlock);
                    chunkWords = new ArrayList<>();
+                    if (CollectionUtils.isNotEmpty(chunkBlock.getTables())) {
+                        previousTable = chunkBlock.getTables().get(chunkBlock.getTables().size() - 1);
+                    }
+                }
+                if (current instanceof Table) {
+                    Table table = (Table) current;
+                    // Distribute header information for subsequent tables
+                    mergeTableMetadata(table, previousTable);
+                    previousTable = table;
                }
-
                chunkWords.add(current);
                prev = current;
            }
        }

-        Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline, previousTable);
+        Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
        chunkBlock.setHeadline(lastHeadline);
        chunkBlockList.add(chunkBlock);

@@ -61,7 +70,38 @@ public class SectionsBuilderService {
    }


-    private Paragraph buildTextBlock(List<AbstractTextContainer> wordBlockList, String lastHeadline, Table previousTable) {
+    private void mergeTableMetadata(Table currentTable, Table previousTable) {
+
+        // Distribute header information for subsequent tables
+        if (previousTable != null && hasInvalidHeaderInformation(currentTable) && hasValidHeaderInformation(previousTable)) {
+            List<Cell> previousTableNonHeaderRow = getRowWithNonHeaderCells(previousTable);
+            List<Cell> tableNonHeaderRow = getRowWithNonHeaderCells(currentTable);
+            // Allow merging of tables if header row is separated from first logical non-header row
+            if (previousTableNonHeaderRow.isEmpty() && previousTable.getRowCount() == 1 && previousTable.getRows()
+                    .get(0)
+                    .size() == tableNonHeaderRow.size()) {
+                previousTableNonHeaderRow = previousTable.getRows().get(0).stream().map(cell -> {
+                    Cell fakeCell = new Cell(cell.getPoints()[0], cell.getPoints()[2]);
+                    fakeCell.setHeaderCells(Collections.singletonList(cell));
+                    return fakeCell;
+                }).collect(Collectors.toList());
+            }
+            if (previousTableNonHeaderRow.size() == tableNonHeaderRow.size()) {
+                for (int i = currentTable.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
+                    List<Cell> row = currentTable.getRows().get(i);
+                    if (row.size() == tableNonHeaderRow.size() && row.stream()
+                            .allMatch(cell -> cell.getHeaderCells().isEmpty())) {
+                        for (int j = 0; j < row.size(); j++) {
+                            row.get(j).setHeaderCells(previousTableNonHeaderRow.get(j).getHeaderCells());
+                        }
+                    }
+                }
+            }
+        }
+    }
+
+
+    private Paragraph buildTextBlock(List<AbstractTextContainer> wordBlockList, String lastHeadline) {

        Paragraph paragraph = new Paragraph();
        TextBlock textBlock = null;
@@ -84,12 +124,6 @@ public class SectionsBuilderService {
                } else {
                    table.setHeadline("Table in: " + lastHeadline);
                }
-                // Distribute header information for subsequent tables
-                if (previousTable != null && hasInvalidHeaderInformation(table) && hasValidHeaderInformation(previousTable) &&
-                        (previousTable.isVerticalHeader() && previousTable.getRowCount() == table.getRowCount() ||
-                                previousTable.getColCount() == table.getColCount())) {
-                    table.setHeaders(previousTable.getHeaders());
-                }

                if (textBlock != null && !alreadyAdded) {
                    paragraph.getPageBlocks().add(textBlock);
@@ -141,14 +175,32 @@ public class SectionsBuilderService {

    private boolean hasInvalidHeaderInformation(Table table) {

-        if (CollectionUtils.isEmpty(table.getHeaders())) {
-            return true;
-        }
-        if (table.getHeaders().stream().anyMatch(StringUtils::isEmpty)) {
-            return true;
+        return table.getRows().stream()
+                .flatMap(row -> row.stream()
+                        .filter(cell -> CollectionUtils.isNotEmpty(cell.getHeaderCells())))
+                .findAny()
+                .isEmpty();
+
+    }
+
+
+    private List<Cell> getRowWithNonHeaderCells(Table table) {
+
+        for (int i = table.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
+            List<Cell> row = table.getRows().get(i);
+            boolean allNonHeader = true;
+            for (Cell cell : row) {
+                if (cell.isHeaderCell()) {
+                    allNonHeader = false;
+                    break;
+                }
+            }
+            if (allNonHeader) {
+                return row;
+            }
        }

-        return false;
+        return Collections.emptyList();

    }

@@ -5,6 +5,8 @@ import java.util.ArrayList;
 import java.util.List;

 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.utils.TextNormalizationUtilities;

 import lombok.Data;
 import lombok.EqualsAndHashCode;
@@ -16,6 +18,10 @@ public class Cell extends Rectangle {

    private List<TextBlock> textBlocks = new ArrayList<>();

+    private List<Cell> headerCells = new ArrayList<>();
+
+    private boolean isHeaderCell;
+

    public Cell(Point2D topLeft, Point2D bottomRight) {

@@ -29,4 +35,34 @@ public class Cell extends Rectangle {
        textBlocks.add(textBlock);
    }

+
+    @Override
+    public String toString() {
+
+        StringBuilder sb = new StringBuilder();
+
+        TextPositionSequence previous = null;
+        for (TextBlock textBlock : textBlocks) {
+
+
+            for (TextPositionSequence word : textBlock.getSequences()) {
+
+                if (previous != null) {
+                    if (Math.abs(previous.getY1() - word.getY1()) > word.getTextHeight()) {
+                        sb.append('\n');
+                    } else {
+                        sb.append(' ');
+                    }
+                }
+                sb.append(word.toString());
+                previous = word;
+            }
+        }
+
+        return TextNormalizationUtilities.removeHyphenLineBreaks(sb.toString())
+                .replaceAll("\n", " ")
+                .replaceAll(" {2}", " ");
+    }
+
 }
+
@@ -0,0 +1,22 @@
+package com.iqser.red.service.redaction.v1.server.tableextraction.model;
+
+import lombok.RequiredArgsConstructor;
+import lombok.Value;
+
+@Value
+@RequiredArgsConstructor
+public class CellPosition implements Comparable<CellPosition> {
+
+    int row;
+
+    int col;
+
+
+    @Override
+    public int compareTo(CellPosition other) {
+
+        int rowDiff = row - other.row;
+        return rowDiff != 0 ? rowDiff : col - other.col;
+    }
+
+}
@@ -8,12 +8,10 @@ import java.util.Iterator;
 import java.util.List;
 import java.util.Map;
 import java.util.TreeMap;
-import java.util.stream.Collectors;

 import org.apache.commons.collections4.CollectionUtils;

 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
 import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;

 import lombok.Getter;
@@ -31,22 +29,18 @@ public class Table extends AbstractTextContainer {
    @Setter
    private String headline;

-    @Getter
-    private int rowCount;
+    private int unrotatedRowCount;

-    @Getter
-    private int colCount;
+    private int unrotatedColCount;
+
+    private int rowCount = -1;
+
+    private int colCount = -1;

    private final int rotation;

    private List<List<Cell>> rows;

-    @Getter
-    @Setter
-    private List<String> headers;
-
-    @Getter
-    private boolean verticalHeader;

    public Table(List<Cell> cells, Rectangle area, int rotation) {

@@ -65,7 +59,7 @@ public class Table extends AbstractTextContainer {

        if (rows == null) {
            rows = computeRows();
-            headers = computeHeaders();
+            computeHeaders();
        }

        return rows;
@@ -73,72 +67,78 @@ public class Table extends AbstractTextContainer {
    }


+    public int getRowCount() {
+
+        if (rowCount == -1) {
+            rowCount = getRows().size();
+        }
+        return rowCount;
+    }
+
+
+    public int getColCount() {
+
+        if (colCount == -1) {
+            colCount = getRows().stream().mapToInt(List::size).max().orElse(0);
+        }
+        return colCount;
+
+    }
+
+
    /**
     * Detect header cells (either first row or first column):
     * Column is marked as header if cell text is bold and row cell text is not bold.
     * Defaults to row.
     */
-    private List<String> computeHeaders() {
+    private void computeHeaders() {

-        boolean allBold = true;
-        if (rows.isEmpty()) {
-            return Collections.emptyList();
+        if (rows == null) {
+            rows = computeRows();
        }
-        List<Cell> rowCells = rows.get(0);
-        for (Cell cell : rowCells) {
-            if (cell == null || CollectionUtils.isEmpty(cell.getTextBlocks()) ||
-                    !cell.getTextBlocks().get(0).getMostPopularWordStyle().equals("bold")) {
-                allBold = false;
-                break;
-            }
-        }
-        if (!allBold) {
-            allBold = true;
-            List<Cell> firstColCells = new ArrayList<>();
-            for (List<Cell> row : rows) {
-                Cell firstInRow = row.get(0);
-                if (firstInRow == null || CollectionUtils.isEmpty(firstInRow.getTextBlocks()) ||
-                        !firstInRow.getTextBlocks().get(0).getMostPopularWordStyle().equals("bold")) {
-                    allBold = false;
-                    break;
-                }
-                firstColCells.add(firstInRow);
-            }
-            if (allBold) {
-                log.info("Headers are in first column");
-                verticalHeader = true;
-                return firstColCells.stream().map(cell -> {
-                    if (CollectionUtils.isNotEmpty(cell.getTextBlocks())) {
-                        return TextNormalizationUtilities.removeHyphenLineBreaks(cell.getTextBlocks().get(0).getText())
-                                .replaceAll("\n", " ")
-                                .replaceAll(" ", " ");
+        // A bold cell is a header cell as long as every cell to the left/top is bold, too
+        // we move from left to right and top to bottom
+        for (int rowIndex = 0; rowIndex < rows.size(); rowIndex++) {
+            List<Cell> rowCells = rows.get(rowIndex);
+            for (int colIndex = 0; colIndex < rowCells.size(); colIndex++) {
+                Cell cell = rowCells.get(colIndex);
+                List<Cell> cellsToTheLeft = rowCells.subList(0, colIndex);
+                Cell lastHeaderCell = null;
+                for (Cell leftCell : cellsToTheLeft) {
+                    if (leftCell.isHeaderCell()) {
+                        lastHeaderCell = leftCell;
                    } else {
-                        return null;
+                        break;
                    }
-                }).collect(Collectors.toList());
-            } else {
-                log.info("Headers are defaulted in first row.");
-                return rowCells.stream().map(cell -> {
-                    if (cell != null && CollectionUtils.isNotEmpty(cell.getTextBlocks())) {
-                        return TextNormalizationUtilities.removeHyphenLineBreaks(cell.getTextBlocks().get(0).getText())
-                                .replaceAll("\n", " ")
-                                .replaceAll(" ", " ");
-                    } else {
-                        return null;
-                    }
-                }).collect(Collectors.toList());
-            }
-        } else {
-            log.info("Headers are in first row.");
-            return rowCells.stream().map(cell -> {
-                if (CollectionUtils.isNotEmpty(cell.getTextBlocks())) {
-                    return TextNormalizationUtilities.removeHyphenLineBreaks(cell.getTextBlocks().get(0).getText())
-                            .replaceAll("\n", " ")
-                            .replaceAll(" ", " ");
-                } else {
-                    return null;
                }
-            }).collect(Collectors.toList());
+                if (lastHeaderCell != null) {
+                    cell.getHeaderCells().add(lastHeaderCell);
+                }
+                List<Cell> cellsToTheTop = new ArrayList<>();
+                for (int i = 0; i < rowIndex; i++) {
+                    try {
+                        cellsToTheTop.add(rows.get(i).get(colIndex));
+                    } catch (IndexOutOfBoundsException e) {
+                        log.debug("No cell {} in row {}, ignoring.", colIndex, rowIndex);
+                    }
+                }
+                for (Cell topCell : cellsToTheTop) {
+                    if (topCell.isHeaderCell()) {
+                        lastHeaderCell = topCell;
+                    } else {
+                        break;
+                    }
+                }
+                if (lastHeaderCell != null) {
+                    cell.getHeaderCells().add(lastHeaderCell);
+                }
+                if (CollectionUtils.isNotEmpty(cell.getTextBlocks()) && cell.getTextBlocks()
+                        .get(0)
+                        .getMostPopularWordStyle()
+                        .equals("bold")) {
+                    cell.setHeaderCell(true);
+                }
+            }
        }

    }
@@ -148,29 +148,35 @@ public class Table extends AbstractTextContainer {

        List<List<Cell>> rows = new ArrayList<>();
        if (rotation == 90) {
-            for (int i = 0; i < colCount; i++) { // rows
+            for (int i = 0; i < unrotatedColCount; i++) { // rows
                List<Cell> lastRow = new ArrayList<>();
-                for (int j = rowCount - 1; j >= 0; j--) { // cols
+                for (int j = unrotatedRowCount - 1; j >= 0; j--) { // cols
                    Cell cell = cells.get(new CellPosition(j, i));
-                    lastRow.add(cell);
+                    if (cell != null) {
+                        lastRow.add(cell);
+                    }
                }
                rows.add(lastRow);
            }
        } else if (rotation == 270) {
-            for (int i = colCount - 1; i >= 0; i--) { // rows
+            for (int i = unrotatedColCount - 1; i >= 0; i--) { // rows
                List<Cell> lastRow = new ArrayList<>();
-                for (int j = 0; j < rowCount; j++) { // cols
+                for (int j = 0; j < unrotatedRowCount; j++) { // cols
                    Cell cell = cells.get(new CellPosition(i, j));
-                    lastRow.add(cell);
+                    if (cell != null) {
+                        lastRow.add(cell);
+                    }
                }
                rows.add(lastRow);
            }
        } else {
-            for (int i = 0; i < rowCount; i++) {
+            for (int i = 0; i < unrotatedRowCount; i++) {
                List<Cell> lastRow = new ArrayList<>();
-                for (int j = 0; j < colCount; j++) {
+                for (int j = 0; j < unrotatedColCount; j++) {
                    Cell cell = cells.get(new CellPosition(i, j)); // JAVA_8 use getOrDefault()
-                    lastRow.add(cell);
+                    if (cell != null) {
+                        lastRow.add(cell);
+                    }
                }
                rows.add(lastRow);
            }
@@ -183,8 +189,8 @@ public class Table extends AbstractTextContainer {

    private void add(Cell chunk, int row, int col) {

-        rowCount = Math.max(rowCount, row + 1);
-        colCount = Math.max(colCount, col + 1);
+        unrotatedRowCount = Math.max(unrotatedRowCount, row + 1);
+        unrotatedColCount = Math.max(unrotatedColCount, col + 1);

        CellPosition cp = new CellPosition(row, col);
        cells.put(cp, chunk);
@@ -220,20 +226,21 @@ public class Table extends AbstractTextContainer {
            while (rowCells.hasNext()) {
                Cell cell = rowCells.next();
                if (i > 0) {
-                    List<List<Cell>> others = rowsOfCells(si.contains(new Rectangle(cell.getBottom(), si.getBounds()
-                            .getLeft(), cell.getLeft() - si.getBounds().getLeft() + 1, si.getBounds().getBottom() - cell
-                            .getBottom())));
+                    Rectangle rectangle = new Rectangle(cell.getBottom(),
+                            si.getBounds().getLeft(),
+                            cell.getLeft() - si.getBounds().getLeft() + 1,
+                            si.getBounds().getBottom() - cell.getBottom());
+                    List<List<Cell>> others = rowsOfCells(si.contains(rectangle));

                    for (List<Cell> r : others) {
                        jumpToColumn = Math.max(jumpToColumn, r.size());
                    }
-                }

-                while (startColumn != jumpToColumn) {
-                    add(previousNonNullCellForColumnIndex.get(startColumn), i, startColumn);
-                    startColumn++;
+                    while (startColumn != jumpToColumn) {
+                        add(previousNonNullCellForColumnIndex.get(startColumn), i, startColumn);
+                        startColumn++;
+                    }
                }
-
                add(cell, i, startColumn);
                previousNonNullCellForColumnIndex.put(startColumn, cell);
                startColumn++;
@@ -243,26 +250,23 @@ public class Table extends AbstractTextContainer {
    }


-    private static List<List<Cell>> rowsOfCells(List<Cell> cells) {
+    private List<List<Cell>> rowsOfCells(List<Cell> cells) {

-        Cell c;
-        float lastTop;
        List<List<Cell>> rv = new ArrayList<>();
-        List<Cell> lastRow;

        if (cells.isEmpty()) {
            return rv;
        }
-
        cells.sort(Comparator.comparingDouble(Rectangle::getLeft));

-        cells.sort(Collections.reverseOrder((arg0, arg1) -> Float.compare(Utils.round(arg0.getBottom(), 2), Utils.round(arg1
+        cells.sort(Collections.reverseOrder((arg0, arg1) -> Float.compare(Utils.round(arg0.getBottom(), 2),
+                Utils.round(arg1
                .getBottom(), 2))));

        Iterator<Cell> iter = cells.iterator();
-        c = iter.next();
-        lastTop = c.getBottom();
-        lastRow = new ArrayList<>();
+        Cell c = iter.next();
+        float lastTop = c.getBottom();
+        List<Cell> lastRow = new ArrayList<>();
        lastRow.add(c);
        rv.add(lastRow);

@@ -349,51 +353,4 @@ public class Table extends AbstractTextContainer {
        return sb.toString();
    }

-
-    static class CellPosition implements Comparable<CellPosition> {
-
-        CellPosition(int row, int col) {
-
-            this.row = row;
-            this.col = col;
-        }
-
-
-        final int row;
-        final int col;
-
-
-        @Override
-        public int hashCode() {
-
-            return row + 101 * col;
-        }
-
-
-        @Override
-        public boolean equals(Object obj) {
-
-            if (this == obj) {
-                return true;
-            }
-            if (obj == null) {
-                return false;
-            }
-            if (getClass() != obj.getClass()) {
-                return false;
-            }
-            CellPosition other = (CellPosition) obj;
-            return row == other.row && col == other.col;
-        }
-
-
-        @Override
-        public int compareTo(CellPosition other) {
-
-            int rowDiff = row - other.row;
-            return rowDiff != 0 ? rowDiff : col - other.col;
-        }
-
-    }
-
 }
@@ -18,9 +18,9 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;
@Service
 public class RulingCleaningService {

-    public CleanRulings getCleanRulings(List<Ruling> rulings, float minCharWidth, float minCharHeight){
+    public CleanRulings getCleanRulings(List<Ruling> rulings, float maxCharWidth, float maxCharHeight){
        if (!rulings.isEmpty()) {
-            snapPoints(rulings, minCharWidth , minCharHeight);
+            snapPoints(rulings, maxCharWidth , maxCharHeight);
        }

        List<Ruling> vrs = new ArrayList<>();
@@ -2,7 +2,6 @@ package com.iqser.red.service.redaction.v1.server.tableextraction.service;

 import java.awt.geom.Point2D;
 import java.util.ArrayList;
-import java.util.Collections;
 import java.util.Comparator;
 import java.util.HashMap;
 import java.util.HashSet;
@@ -25,26 +24,28 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
 import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;

@Service
-@SuppressWarnings("all")
 public class TableExtractionService {

-    public void extractTables(CleanRulings cleanRulings, Page page){
+    public void extractTables(CleanRulings cleanRulings, Page page) {

        List<Cell> cells = findCells(cleanRulings.getHorizontal(), cleanRulings.getVertical());

-        Iterator<AbstractTextContainer> itty = page.getTextBlocks().iterator();
-        while (itty.hasNext()) {
-            TextBlock textBlock = (TextBlock) itty.next();
+        for (AbstractTextContainer abstractTextContainer : page.getTextBlocks()) {
+            TextBlock textBlock = (TextBlock) abstractTextContainer;
            for (Cell cell : cells) {
-                if (cell.intersects(textBlock.getMinX(), textBlock.getMinY(), textBlock.getWidth(), textBlock.getHeight())) {
+                if (cell.intersects(textBlock.getMinX(), textBlock.getMinY(), textBlock.getWidth(),
+                        textBlock.getHeight())) {
                    cell.addTextBlock(textBlock);
                    break;
                }
            }
        }

-        List<Rectangle> spreadsheetAreas = findSpreadsheetsFromCells(cells)
-                .stream()
+        cells = new ArrayList<>(new HashSet<>(cells));
+        Utils.sort(cells, Rectangle.ILL_DEFINED_ORDER);
+
+
+        List<Rectangle> spreadsheetAreas = findSpreadsheetsFromCells(cells).stream()
                .filter(r -> r.getWidth() > 0f && r.getHeight() > 0f)
                .collect(Collectors.toList());

@@ -63,9 +64,9 @@ public class TableExtractionService {
        for (Table table : tables) {
            int position = -1;

-            itty = page.getTextBlocks().iterator();
+            Iterator<AbstractTextContainer> itty = page.getTextBlocks().iterator();
            while (itty.hasNext()) {
-                AbstractTextContainer textBlock = (AbstractTextContainer) itty.next();
+                AbstractTextContainer textBlock = itty.next();
                if (table.contains(textBlock)) {
                    if (position == -1) {
                        position = page.getTextBlocks().indexOf(textBlock);
@@ -79,17 +80,18 @@ public class TableExtractionService {
        }
    }

+
    public List<Cell> findCells(List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines) {
+
        List<Cell> cellsFound = new ArrayList<>();
-        Map<Point2D, Ruling[]> intersectionPoints = Ruling.findIntersections(horizontalRulingLines, verticalRulingLines);
+        Map<Point2D, Ruling[]> intersectionPoints = Ruling.findIntersections(horizontalRulingLines,
+                verticalRulingLines);
        List<Point2D> intersectionPointsList = new ArrayList<>(intersectionPoints.keySet());
-        Collections.sort(intersectionPointsList, POINT_COMPARATOR);
-        boolean doBreak;
+        intersectionPointsList.sort(POINT_COMPARATOR);

        for (int i = 0; i < intersectionPointsList.size(); i++) {
            Point2D topLeft = intersectionPointsList.get(i);
            Ruling[] hv = intersectionPoints.get(topLeft);
-            doBreak = false;

            // CrossingPointsDirectlyBelow( topLeft );
            List<Point2D> xPoints = new ArrayList<>();
@@ -106,10 +108,6 @@ public class TableExtractionService {
            }
            outer:
            for (Point2D xPoint : xPoints) {
-                if (doBreak) {
-                    break;
-                }
-
                // is there a vertical edge b/w topLeft and xPoint?
                if (!hv[1].equals(intersectionPoints.get(xPoint)[1])) {
                    continue;
@@ -120,11 +118,9 @@ public class TableExtractionService {
                        continue;
                    }
                    Point2D btmRight = new Point2D.Float((float) yPoint.getX(), (float) xPoint.getY());
-                    if (intersectionPoints.containsKey(btmRight)
-                            && intersectionPoints.get(btmRight)[0].equals(intersectionPoints.get(xPoint)[0])
-                            && intersectionPoints.get(btmRight)[1].equals(intersectionPoints.get(yPoint)[1])) {
+                    if (intersectionPoints.containsKey(btmRight) && intersectionPoints.get(btmRight)[0].equals(intersectionPoints
+                            .get(xPoint)[0]) && intersectionPoints.get(btmRight)[1].equals(intersectionPoints.get(yPoint)[1])) {
                        cellsFound.add(new Cell(topLeft, btmRight));
-                        doBreak = true;
                        break outer;
                    }
                }
@@ -139,7 +135,7 @@ public class TableExtractionService {
    }


-    public List<Rectangle> findSpreadsheetsFromCells(List<? extends Rectangle> cells) {
+    private List<Rectangle> findSpreadsheetsFromCells(List<? extends Rectangle> cells) {
        // via: http://stackoverflow.com/questions/13746284/merging-multiple-adjacent-rectangles-into-one-polygon
        List<Rectangle> rectangles = new ArrayList<>();
        Set<Point2D> pointSet = new HashSet<>();
@@ -147,10 +143,6 @@ public class TableExtractionService {
        Map<Point2D, Point2D> edgesV = new HashMap<>();
        int i = 0;

-        cells = new ArrayList<>(new HashSet<>(cells));
-
-        Utils.sort(cells, Rectangle.ILL_DEFINED_ORDER);
-
        for (Rectangle cell : cells) {
            for (Point2D pt : cell.getPoints()) {
                if (pointSet.contains(pt)) { // shared vertex, remove it
@@ -163,10 +155,10 @@ public class TableExtractionService {

        // X first sort
        List<Point2D> pointsSortX = new ArrayList<>(pointSet);
-        Collections.sort(pointsSortX, X_FIRST_POINT_COMPARATOR);
+        pointsSortX.sort(X_FIRST_POINT_COMPARATOR);
        // Y first sort
        List<Point2D> pointsSortY = new ArrayList<>(pointSet);
-        Collections.sort(pointsSortY, POINT_COMPARATOR);
+        pointsSortY.sort(POINT_COMPARATOR);

        while (i < pointSet.size()) {
            float currY = (float) pointsSortY.get(i).getY();
@@ -203,13 +195,12 @@ public class TableExtractionService {
                    nextVertex = edgesV.get(curr.point);
                    edgesV.remove(curr.point);
                    lastAddedVertex = new PolygonVertex(nextVertex, Direction.VERTICAL);
-                    polygon.add(lastAddedVertex);
                } else {
                    nextVertex = edgesH.get(curr.point);
                    edgesH.remove(curr.point);
                    lastAddedVertex = new PolygonVertex(nextVertex, Direction.HORIZONTAL);
-                    polygon.add(lastAddedVertex);
                }
+                polygon.add(lastAddedVertex);

                if (lastAddedVertex.equals(polygon.get(0))) {
                    // closed polygon
@@ -227,10 +218,10 @@ public class TableExtractionService {

        // calculate grid-aligned minimum area rectangles for each found polygon
        for (List<PolygonVertex> poly : polygons) {
-            float top = java.lang.Float.MAX_VALUE;
-            float left = java.lang.Float.MAX_VALUE;
-            float bottom = java.lang.Float.MIN_VALUE;
-            float right = java.lang.Float.MIN_VALUE;
+            float top = Float.MAX_VALUE;
+            float left = Float.MAX_VALUE;
+            float bottom = Float.MIN_VALUE;
+            float right = Float.MIN_VALUE;
            for (PolygonVertex pt : poly) {
                top = (float) Math.min(top, pt.point.getY());
                left = (float) Math.min(left, pt.point.getX());
@@ -244,69 +235,66 @@ public class TableExtractionService {
    }


-    private static final Comparator<Point2D> X_FIRST_POINT_COMPARATOR = new Comparator<Point2D>() {
-        @Override
-        public int compare(Point2D arg0, Point2D arg1) {
-            int rv = 0;
-            float arg0X = Utils.round(arg0.getX(), 2);
-            float arg0Y = Utils.round(arg0.getY(), 2);
-            float arg1X = Utils.round(arg1.getX(), 2);
-            float arg1Y = Utils.round(arg1.getY(), 2);
+    private static final Comparator<Point2D> X_FIRST_POINT_COMPARATOR = (arg0, arg1) -> {

-            if (arg0X > arg1X) {
-                rv = 1;
-            } else if (arg0X < arg1X) {
-                rv = -1;
-            } else if (arg0Y > arg1Y) {
-                rv = 1;
-            } else if (arg0Y < arg1Y) {
-                rv = -1;
-            }
-            return rv;
+        int rv = 0;
+        float arg0X = Utils.round(arg0.getX(), 2);
+        float arg0Y = Utils.round(arg0.getY(), 2);
+        float arg1X = Utils.round(arg1.getX(), 2);
+        float arg1Y = Utils.round(arg1.getY(), 2);
+
+        if (arg0X > arg1X) {
+            rv = 1;
+        } else if (arg0X < arg1X) {
+            rv = -1;
+        } else if (arg0Y > arg1Y) {
+            rv = 1;
+        } else if (arg0Y < arg1Y) {
+            rv = -1;
        }
+        return rv;
    };

+    private static final Comparator<Point2D> POINT_COMPARATOR = (arg0, arg1) -> {

-    private static final Comparator<Point2D> POINT_COMPARATOR = new Comparator<Point2D>() {
-        @Override
-        public int compare(Point2D arg0, Point2D arg1) {
-            int rv = 0;
-            float arg0X = Utils.round(arg0.getX(), 2);
-            float arg0Y = Utils.round(arg0.getY(), 2);
-            float arg1X = Utils.round(arg1.getX(), 2);
-            float arg1Y = Utils.round(arg1.getY(), 2);
+        int rv = 0;
+        float arg0X = Utils.round(arg0.getX(), 2);
+        float arg0Y = Utils.round(arg0.getY(), 2);
+        float arg1X = Utils.round(arg1.getX(), 2);
+        float arg1Y = Utils.round(arg1.getY(), 2);

-
-            if (arg0Y > arg1Y) {
-                rv = 1;
-            } else if (arg0Y < arg1Y) {
-                rv = -1;
-            } else if (arg0X > arg1X) {
-                rv = 1;
-            } else if (arg0X < arg1X) {
-                rv = -1;
-            }
-            return rv;
+        if (arg0Y > arg1Y) {
+            rv = 1;
+        } else if (arg0Y < arg1Y) {
+            rv = -1;
+        } else if (arg0X > arg1X) {
+            rv = 1;
+        } else if (arg0X < arg1X) {
+            rv = -1;
        }
+        return rv;
    };

-
    private enum Direction {
-        HORIZONTAL,
-        VERTICAL
+        HORIZONTAL, VERTICAL
    }

    static class PolygonVertex {
+
        Point2D point;
        Direction direction;

-        public PolygonVertex(Point2D point, Direction direction) {
+
+        PolygonVertex(Point2D point, Direction direction) {
+
            this.direction = direction;
            this.point = point;
        }

+
        @Override
        public boolean equals(Object other) {
+
            if (this == other) {
                return true;
            }
@@ -316,15 +304,21 @@ public class TableExtractionService {
            return this.point.equals(((PolygonVertex) other).point);
        }

+
        @Override
        public int hashCode() {
+
            return this.point.hashCode();
        }

+
        @Override
        public String toString() {
-            return String.format("%s[point=%s,direction=%s]", this.getClass().getName(), this.point.toString(), this.direction.toString());
+
+            return String.format("%s[point=%s,direction=%s]", this.getClass()
+                    .getName(), this.point.toString(), this.direction.toString());
        }
+
    }

 }
@@ -2,7 +2,11 @@ package com.iqser.red.service.redaction.v1.server.visualization.service;

 import java.awt.Color;
 import java.io.IOException;
+import java.util.ArrayList;
+import java.util.HashSet;
 import java.util.List;
+import java.util.Set;
+import java.util.stream.Collectors;

 import org.apache.commons.collections4.CollectionUtils;
 import org.apache.pdfbox.pdmodel.PDDocument;
@@ -14,6 +18,7 @@ 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.PDAnnotationTextMarkup;
+import org.apache.pdfbox.text.TextPosition;
 import org.springframework.stereotype.Service;

 import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
@@ -33,9 +38,7 @@ 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 lombok.extern.slf4j.Slf4j;

-@Slf4j
@Service
@RequiredArgsConstructor
 public class AnnotationHighlightService {
@@ -43,7 +46,10 @@ public class AnnotationHighlightService {
    private final DictionaryService dictionaryService;


-    public void highlight(PDDocument document, Document classifiedDoc, boolean flatRedaction, ManualRedactions manualRedactions) throws IOException {
+    public void highlight(PDDocument document, Document classifiedDoc, boolean flatRedaction,
+                          ManualRedactions manualRedactions) throws IOException {
+
+        Set<Integer> manualRedactionPages = getManualRedactionPages(manualRedactions);

        for (int page = 1; page <= document.getNumberOfPages(); page++) {

@@ -51,17 +57,36 @@ public class AnnotationHighlightService {

            drawSectionFrames(document, classifiedDoc, flatRedaction, pdPage, page);

-            if (classifiedDoc.getEntities().get(page) == null) {
-                continue;
+            if (classifiedDoc.getEntities().get(page) != null) {
+                addAnnotations(pdPage, classifiedDoc, flatRedaction, manualRedactions, page);
            }

-            addAnnotations(pdPage, classifiedDoc, flatRedaction, manualRedactions, page);
-            addManualAnnotations(pdPage, classifiedDoc, manualRedactions, page);
+            if (manualRedactionPages.contains(page)) {
+                addManualAnnotations(pdPage, classifiedDoc, manualRedactions, page);
+            }
        }
    }


-    private void addAnnotations(PDPage pdPage, Document classifiedDoc, boolean flatRedaction, ManualRedactions manualRedactions, int page) throws IOException {
+    private Set<Integer> getManualRedactionPages(ManualRedactions manualRedactions) {
+
+        Set<Integer> manualRedactionPages = new HashSet<>();
+
+        if (manualRedactions == null) {
+            return manualRedactionPages;
+        }
+
+        manualRedactions.getEntriesToAdd().forEach(entry -> {
+            entry.getPositions().forEach(pos -> {
+                manualRedactionPages.add(pos.getPage());
+            });
+        });
+        return manualRedactionPages;
+    }
+
+
+    private void addAnnotations(PDPage pdPage, Document classifiedDoc, boolean flatRedaction,
+                                ManualRedactions manualRedactions, int page) throws IOException {

        List<PDAnnotation> annotations = pdPage.getAnnotations();

@@ -77,25 +102,59 @@ public class AnnotationHighlightService {

                if (manualRedactions != null && manualRedactions.getIdsToRemove()
                        .contains(entityPositionSequence.getId())) {
+                    String manualOverrideReason = entity.getRedactionReason() + ", removed by manual override";
                    entity.setRedaction(false);
-                    entity.setRedactionReason(entity.getRedactionReason() + ", removed by manual override");
+                    entity.setRedactionReason(manualOverrideReason);
+                    redactionLogEntry.setRedacted(false);
+                    redactionLogEntry.setReason(manualOverrideReason);
                    redactionLogEntry.setManual(true);
                }

-                for (TextPositionSequence textPositions : entityPositionSequence.getSequences()) {
+                if (CollectionUtils.isNotEmpty(entityPositionSequence.getSequences())) {
+                    List<Rectangle> rectanglesPerline = getRectanglesPerLine(entityPositionSequence.getSequences()
+                            .stream()
+                            .flatMap(seq -> seq.getTextPositions().stream())
+                            .collect(Collectors.toList()), page);

-                    Rectangle rectangle = textPositions.getRectangle();
-                    redactionLogEntry.getPositions().add(rectangle);
-                    annotations.add(createAnnotation(rectangle, entityPositionSequence.getId(), createAnnotationContent(entity), getColor(entity), !flatRedaction && !isHint(entity)));
+                    redactionLogEntry.getPositions().addAll(rectanglesPerline);
+                    annotations.add(createAnnotation(rectanglesPerline, entityPositionSequence.getId(), createAnnotationContent(entity), getColor(entity), !flatRedaction && !isHint(entity)));
                }
+
                redactionLogEntry.setId(entityPositionSequence.getId());
            }
            classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
        }
+
    }


-    private void addManualAnnotations(PDPage pdPage, Document classifiedDoc, ManualRedactions manualRedactions, int page) throws IOException {
+    private List<Rectangle> getRectanglesPerLine(List<TextPosition> textPositions, int page) {
+
+        List<Rectangle> rectangles = new ArrayList<>();
+        if (textPositions.size() == 1) {
+            rectangles.add(new TextPositionSequence(textPositions, page).getRectangle());
+        } else {
+            float y = textPositions.get(0).getYDirAdj();
+            int startIndex = 0;
+            for (int i = 1; i < textPositions.size(); i++) {
+                float yDirAdj = textPositions.get(i).getYDirAdj();
+                if (yDirAdj != y) {
+                    rectangles.add(new TextPositionSequence(textPositions.subList(startIndex, i), page).getRectangle());
+                    y = yDirAdj;
+                    startIndex = i;
+                }
+            }
+            if (startIndex != textPositions.size() - 1) {
+                rectangles.add(new TextPositionSequence(textPositions.subList(startIndex, textPositions.size()), page).getRectangle());
+            }
+        }
+
+        return rectangles;
+    }
+
+
+    private void addManualAnnotations(PDPage pdPage, Document classifiedDoc, ManualRedactions manualRedactions,
+                                      int page) throws IOException {

        if (manualRedactions == null) {
            return;
@@ -107,28 +166,34 @@ public class AnnotationHighlightService {

            String id = IdBuilder.buildId(manualRedactionEntry);

-            RedactionLogEntry redactionLogEntry = createRedactionLogEntry(manualRedactionEntry);
+            RedactionLogEntry redactionLogEntry = createRedactionLogEntry(manualRedactionEntry, id);

+            boolean foundOnPage = false;
            for (Rectangle rectangle : manualRedactionEntry.getPositions()) {

                if (page != rectangle.getPage()) {
                    continue;
                }

-                PDAnnotationTextMarkup highlight = createAnnotation(rectangle, id, createAnnotationContent(manualRedactionEntry), getColor(manualRedactionEntry
+                foundOnPage = true;
+
+                PDAnnotationTextMarkup highlight = createAnnotation(List.of(rectangle), id, createAnnotationContent(manualRedactionEntry), getColor(manualRedactionEntry
                        .getType()), true);
                annotations.add(highlight);

                redactionLogEntry.getPositions().add(rectangle);
            }
-            classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
+            if (foundOnPage) {
+                classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
+            }
        }
    }


-    private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry) {
+    private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry, String id) {

        return RedactionLogEntry.builder()
+                .id(id)
                .color(getColor(manualRedactionEntry.getType()))
                .reason(manualRedactionEntry.getReason())
                .value(manualRedactionEntry.getValue())
@@ -157,12 +222,13 @@ public class AnnotationHighlightService {
    }


-    private PDAnnotationTextMarkup createAnnotation(Rectangle rectangle, String id, String content, float[] color, boolean popup) {
+    private PDAnnotationTextMarkup createAnnotation(List<Rectangle> rectangles, String id, String content,
+                                                    float[] color, boolean popup) {

        PDAnnotationTextMarkup annotation = new PDAnnotationTextMarkup(PDAnnotationTextMarkup.SUB_TYPE_HIGHLIGHT);
        annotation.constructAppearances();
-        annotation.setRectangle(toPDRectangle(rectangle));
-        annotation.setQuadPoints(toQuadPoints(rectangle));
+        annotation.setRectangle(toPDRectangle(rectangles.get(0)));
+        annotation.setQuadPoints(toQuadPoints(rectangles));
        if (popup) {
            annotation.setAnnotationName(id);
            annotation.setTitlePopup(id);
@@ -175,25 +241,14 @@ public class AnnotationHighlightService {

    private String createAnnotationContent(Entity entity) {

-        return new StringBuilder().append("\nRule ")
-                .append(entity.getMatchedRule())
-                .append(" matched")
-                .append("\n\n")
-                .append(entity.getRedactionReason())
-                .append("\n\nIn Section : \"")
-                .append(entity.getHeadline())
-                .append("\"")
-                .toString();
+        return "\nRule " + entity.getMatchedRule() + " matched" + "\n\n" + entity.getRedactionReason() + "\n\nIn " + "Section : \"" + entity
+                .getHeadline() + "\"";
    }


    private String createAnnotationContent(ManualRedactionEntry entry) {

-        return new StringBuilder().append("\nManual Redaction")
-                .append("\n\nIn Section : \"")
-                .append(entry.getSection())
-                .append("\"")
-                .toString();
+        return "\nManual Redaction" + "\n\nIn Section : \"" + entry.getSection() + "\"";
    }


@@ -208,7 +263,23 @@ public class AnnotationHighlightService {
    }


-    private float[] toQuadPoints(Rectangle rectangle) {
+    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
@@ -224,10 +295,7 @@ public class AnnotationHighlightService {
        if (!entity.isRedaction()) {
            return false;
        }
-        if (isHint(entity)) {
-            return false;
-        }
-        return true;
+        return !isHint(entity);
    }


@@ -255,14 +323,12 @@ public class AnnotationHighlightService {
    private boolean isHint(Entity entity) {

        List<String> hintTypes = dictionaryService.getHintTypes();
-        if (CollectionUtils.isNotEmpty(hintTypes) && hintTypes.contains(entity.getType())) {
-            return true;
-        }
-        return false;
+        return CollectionUtils.isNotEmpty(hintTypes) && hintTypes.contains(entity.getType());
    }


-    private void drawSectionFrames(PDDocument document, Document classifiedDoc, boolean flatRedaction, PDPage pdPage, int page) throws IOException {
+    private void drawSectionFrames(PDDocument document, Document classifiedDoc, boolean flatRedaction, PDPage pdPage,
+                                   int page) throws IOException {

        if (flatRedaction) {
            return;
@@ -8,12 +8,6 @@ spring:
  profiles:
    active: kubernetes

-platform.multi-tenancy:
-  enabled: ${multitenancy.enabled:false}
-  tenantFilter:
-    urlPatterns: /redact
-    urlPatternsToIgnore:
-
 management:
  endpoint:
    metrics.enabled: ${monitoring.enabled:false}
@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server;

+import static org.assertj.core.api.Assertions.assertThat;
 import static org.mockito.Mockito.when;
 import static org.springframework.boot.test.context.SpringBootTest.WebEnvironment.DEFINED_PORT;

@@ -62,6 +63,7 @@ public class RedactionIntegrationTest {
    private static final String VERTEBRATES_CODE = "vertebrate";
    private static final String ADDRESS_CODE = "address";
    private static final String NAME_CODE = "name";
+    private static final String SPONSOR = "sponsor";
    private static final String NO_REDACTION_INDICATOR = "no_redaction_indicator";
    private static final String REDACTION_INDICATOR = "redaction_indicator";
    private static final String HINT_ONLY = "hint_only";
@@ -116,6 +118,7 @@ public class RedactionIntegrationTest {
        when(dictionaryClient.getDictionaryForType(VERTEBRATES_CODE)).thenReturn(getDictionaryResponse(VERTEBRATES_CODE));
        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(getDictionaryResponse(ADDRESS_CODE));
        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(getDictionaryResponse(NAME_CODE));
+        when(dictionaryClient.getDictionaryForType(SPONSOR)).thenReturn(getDictionaryResponse(SPONSOR));
        when(dictionaryClient.getDictionaryForType(NO_REDACTION_INDICATOR)).thenReturn(getDictionaryResponse(NO_REDACTION_INDICATOR));
        when(dictionaryClient.getDictionaryForType(REDACTION_INDICATOR)).thenReturn(getDictionaryResponse(REDACTION_INDICATOR));
        when(dictionaryClient.getDictionaryForType(HINT_ONLY)).thenReturn(getDictionaryResponse(HINT_ONLY));
@@ -131,6 +134,11 @@ public class RedactionIntegrationTest {
                        .stream()
                        .map(this::cleanDictionaryEntry)
                        .collect(Collectors.toSet()));
+        dictionary.computeIfAbsent(SPONSOR, v -> new ArrayList<>())
+                .addAll(ResourceLoader.load("dictionaries/sponsor_companies.txt")
+                        .stream()
+                        .map(this::cleanDictionaryEntry)
+                        .collect(Collectors.toSet()));
        dictionary.computeIfAbsent(VERTEBRATES_CODE, v -> new ArrayList<>())
                .addAll(ResourceLoader.load("dictionaries/vertebrates.txt")
                        .stream()
@@ -175,6 +183,7 @@ public class RedactionIntegrationTest {
        typeColorMap.put(VERTEBRATES_CODE, new float[]{0, 1, 0});
        typeColorMap.put(ADDRESS_CODE, new float[]{0, 1, 1});
        typeColorMap.put(NAME_CODE, new float[]{1, 1, 0});
+        typeColorMap.put(SPONSOR, new float[]{.5f, .5f, .5f});
        typeColorMap.put(NO_REDACTION_INDICATOR, new float[]{0.8f, 0, 0.8f});
        typeColorMap.put(REDACTION_INDICATOR, new float[]{1, 0.502f, 0.1f});
        typeColorMap.put(HINT_ONLY, new float[]{0.8f, 1, 0.8f});
@@ -183,6 +192,7 @@ public class RedactionIntegrationTest {
        hintTypeMap.put(VERTEBRATES_CODE, true);
        hintTypeMap.put(ADDRESS_CODE, false);
        hintTypeMap.put(NAME_CODE, false);
+        hintTypeMap.put(SPONSOR, false);
        hintTypeMap.put(NO_REDACTION_INDICATOR, true);
        hintTypeMap.put(REDACTION_INDICATOR, true);
        hintTypeMap.put(HINT_ONLY, true);
@@ -191,6 +201,7 @@ public class RedactionIntegrationTest {
        caseInSensitiveMap.put(VERTEBRATES_CODE, true);
        caseInSensitiveMap.put(ADDRESS_CODE, false);
        caseInSensitiveMap.put(NAME_CODE, false);
+        caseInSensitiveMap.put(SPONSOR, false);
        caseInSensitiveMap.put(NO_REDACTION_INDICATOR, true);
        caseInSensitiveMap.put(REDACTION_INDICATOR, true);
        caseInSensitiveMap.put(HINT_ONLY, true);
@@ -227,6 +238,7 @@ public class RedactionIntegrationTest {
    @Test
    public void noExceptionShouldBeThrownForAnyFiles() throws IOException {

+        System.out.println("noExceptionShouldBeThrownForAnyFiles");
        ClassLoader loader = getClass().getClassLoader();
        URL url = loader.getResource("files");
        File[] files = new File(url.getPath()).listFiles();
@@ -266,8 +278,9 @@ public class RedactionIntegrationTest {
    @Test
    public void redactionTest() throws IOException {

+        System.out.println("redactionTest");
        long start = System.currentTimeMillis();
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Trinexapac/96 Trinexapac-ethyl_RAR_09_Volume_3CA_B-7_2018-02-23.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()))
@@ -289,8 +302,9 @@ public class RedactionIntegrationTest {
    @Test
    public void testTableRedaction() throws IOException {

+        System.out.println("testTableRedaction");
        long start = System.currentTimeMillis();
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Single Table.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()))
@@ -311,6 +325,7 @@ public class RedactionIntegrationTest {
    @Test
    public void testManualRedaction() throws IOException {

+        System.out.println("testManualRedaction");
        long start = System.currentTimeMillis();
        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Single Table.pdf");

@@ -345,6 +360,7 @@ public class RedactionIntegrationTest {
    @Test
    public void classificationTest() throws IOException {

+        System.out.println("classificationTest");
        ClassPathResource pdfFileResource = new ClassPathResource("files/Fludioxonil/51 " +
                "Fludioxonil_RAR_02_Volume_2_2018-02-21.pdf");

@@ -363,6 +379,7 @@ public class RedactionIntegrationTest {
    @Test
    public void sectionsTest() throws IOException {

+        System.out.println("sectionsTest");
        ClassPathResource pdfFileResource = new ClassPathResource("files/Fludioxonil/51 " +
                "Fludioxonil_RAR_02_Volume_2_2018-02-21.pdf");

@@ -381,8 +398,8 @@ public class RedactionIntegrationTest {
    @Test
    public void htmlTablesTest() throws IOException {

-        ClassPathResource pdfFileResource = new ClassPathResource("files/Fludioxonil/51 " +
-                "Fludioxonil_RAR_02_Volume_2_2018-02-21.pdf");
+        System.out.println("htmlTablesTest");
+        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()))
@@ -399,6 +416,7 @@ public class RedactionIntegrationTest {
    @Test
    public void htmlTableRotationTest() throws IOException {

+        System.out.println("htmlTableRotationTest");
        ClassPathResource pdfFileResource = new ClassPathResource("files/Metolachlor/S" +
                "-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");

@@ -414,6 +432,49 @@ public class RedactionIntegrationTest {
    }


+    @Test
+    public void phantomCellsDocumentTest() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Phantom Cells.pdf");
+
+        RedactionRequest request = RedactionRequest.builder()
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .build();
+        request.setFlatRedaction(false);
+
+        RedactionResult result = redactionController.redact(request);
+
+        result.getRedactionLog().getRedactionLogEntry().forEach(entry -> {
+            if (!entry.isHint()) {
+                assertThat(entry.getReason()).isEqualTo("Not redacted because row is not a vertebrate study");
+            }
+        });
+    }
+
+
+    @Test
+    public void sponsorCompanyTest() throws IOException {
+
+        long start = System.currentTimeMillis();
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/sponsor_companies.pdf");
+
+        RedactionRequest request = RedactionRequest.builder()
+                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
+                .build();
+        request.setFlatRedaction(false);
+
+        RedactionResult result = redactionController.redact(request);
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Redacted.pdf")) {
+            fileOutputStream.write(result.getDocument());
+        }
+        long end = System.currentTimeMillis();
+
+        System.out.println("duration: " + (end - start));
+        System.out.println("numberOfPages: " + result.getNumberOfPages());
+    }
+
+
    private static String loadFromClassPath(String path) {

        URL resource = ResourceLoader.class.getClassLoader().getResource(path);
@@ -10,13 +10,16 @@ import java.io.InputStream;
 import java.io.InputStreamReader;
 import java.net.URL;
 import java.nio.charset.StandardCharsets;
+import java.util.ArrayList;
 import java.util.Arrays;
 import java.util.Collections;
 import java.util.HashSet;
 import java.util.Set;
+import java.util.concurrent.atomic.AtomicLong;

 import org.apache.commons.io.IOUtils;
 import org.apache.pdfbox.pdmodel.PDDocument;
+import org.junit.Before;
 import org.junit.Test;
 import org.junit.runner.RunWith;
 import org.kie.api.KieServices;
@@ -45,14 +48,16 @@ import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
 import com.iqser.red.service.redaction.v1.server.segmentation.PdfSegmentationService;

-@RunWith(SpringRunner.class)
@SpringBootTest
+@RunWith(SpringRunner.class)
 public class EntityRedactionServiceTest {

    private static final String DEFAULT_RULES = loadFromClassPath("drools/rules.drl");
    private static final String NAME_CODE = "name";
    private static final String ADDRESS_CODE = "address";

+    private static final AtomicLong DICTIONARY_VERSION = new AtomicLong();
+    private static final AtomicLong RULES_VERSION = new AtomicLong();
    @MockBean
    private DictionaryClient dictionaryClient;

@@ -65,6 +70,9 @@ public class EntityRedactionServiceTest {
    @Autowired
    private PdfSegmentationService pdfSegmentationService;

+    @Autowired
+    private DroolsExecutionService droolsExecutionService;
+
    @TestConfiguration
    public static class RedactionIntegrationTestConfiguration {

@@ -112,45 +120,244 @@ public class EntityRedactionServiceTest {
                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
                .build();

+        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
+                .entries(Arrays.asList("Casey, H.W.", "O’Loughlin,  C.K.", "Salamon, C.M.", "Smith, S.H."))
+                .build();
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        DictionaryResponse addressResponse = DictionaryResponse.builder()
+                .entries(Collections.singletonList("Toxigenics, Inc., Decatur, IL 62526, USA"))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+
+        try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(redactionRequest.getDocument()))) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+            assertThat(classifiedDoc.getEntities().get(1)).hasSize(7);// 3 names, 1 address, 1 Y and 2 N entities
+        }
+    }
+
+
+    @Test
+    public void testTrueNegativesInTable() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Cyprodinil/40 Cyprodinil - EU AIR3 - LCA Section 1" +
+                " Supplement - Identity of the active substance - Reference list.pdf");
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
+                .entries(new ArrayList<>(ResourceLoader.load("dictionaries/names.txt")))
+                .build();
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        DictionaryResponse addressResponse = DictionaryResponse.builder()
+                .entries(new ArrayList<>(ResourceLoader.load("dictionaries/addresses.txt")))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()
+                    .entrySet()
+                    .stream()
+                    .noneMatch(entry -> entry.getValue().stream().anyMatch(e -> e.getMatchedRule() == 9))).isTrue();
+        }
+        pdfFileResource = new ClassPathResource("files/Compounds/27 A8637C - EU AIR3 - MCP Section 1 - Identity of " +
+                "the plant protection product.pdf");
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()
+                    .entrySet()
+                    .stream()
+                    .noneMatch(entry -> entry.getValue().stream().anyMatch(e -> e.getMatchedRule() == 9))).isTrue();
+        }
+    }
+
+    @Test
+    public void testFalsePositiveInWrongCell() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Row With Ambiguous Redaction.pdf");
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
+                .entries(new ArrayList<>(ResourceLoader.load("dictionaries/names.txt")))
+                .build();
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        DictionaryResponse addressResponse = DictionaryResponse.builder()
+                .entries(new ArrayList<>(ResourceLoader.load("dictionaries/addresses.txt")))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+            assertThat(classifiedDoc.getEntities().get(1).stream()
+                    .filter(entity -> entity.getMatchedRule() == 9)
+                    .count()).isEqualTo(10);
+        }
+
+    }
+
+
+    @Test
+    public void testApplicantInTableRedaction() throws IOException {
+
        String tableRules = "package drools\n" +
                "\n" +
                "import com.iqser.red.service.redaction.v1.server.redaction.model.Section\n" +
                "\n" +
                "global Section section\n" +
+                "rule \"6: Redact contact information if applicant is found\"\n" +
+                "   when\n" +
+                "       eval(section.headlineContainsWord(\"applicant\") || section.getText().contains(\"Applicant\"));\n" +
+                "   then\n" +
+                "       section.redactLineAfter(\"Name:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactBetween(\"Address:\", \"Contact\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Contact point:\",  \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Phone:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Fax:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Tel.:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Tel:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"E-mail:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Email:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Contact:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Telephone number:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Fax number:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Telephone:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactBetween(\"No:\", \"Fax\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactBetween(\"Contact:\", \"Tel.:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "   end";
+        when(rulesClient.getVersion()).thenReturn(RULES_VERSION.incrementAndGet());
+        when(rulesClient.getRules()).thenReturn(new RulesResponse(tableRules));
+        droolsExecutionService.updateRules();
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Applicant Producer Table.pdf");
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
+                .entries(new ArrayList<>(ResourceLoader.load("dictionaries/names.txt")))
+                .build();
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        DictionaryResponse addressResponse = DictionaryResponse.builder()
+                .entries(new ArrayList<>(ResourceLoader.load("dictionaries/addresses.txt")))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+            assertThat(classifiedDoc.getEntities().get(1).stream()
+                    .filter(entity -> entity.getMatchedRule() == 6)
+                    .count()).isEqualTo(18);
+        }
+
+    }
+
+
+    @Test
+    public void headerPropagation() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Header Propagation.pdf");
+
+        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
+                .entries(Arrays.asList("Bissig R.", "Thanei P."))
+                .build();
+
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        DictionaryResponse addressResponse = DictionaryResponse.builder()
+                .entries(Collections.singletonList("Novartis Crop Protection AG, Basel, Switzerland"))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(2); // two pages
+            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(8);
+            assertThat(classifiedDoc.getEntities().get(2).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(4);
+        }
+
+        pdfFileResource = new ClassPathResource("files/Minimal Examples/Header Propagation2.pdf");
+
+        dictionaryResponse = DictionaryResponse.builder()
+                .entries(Arrays.asList("Tribolet, R.", "Muir, G.", "Kühne-Thu, H.", "Close, C."))
+                .build();
+
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        addressResponse = DictionaryResponse.builder()
+                .entries(Collections.singletonList("Novartis Crop Protection AG, Basel, Switzerland"))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(3);
+            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(8);
+        }
+    }
+
+
+    @Test
+    public void testNGuideline() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Empty Tabular Data.pdf");
+
+        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
+                .entries(Collections.singletonList("Aldershof S."))
+                .build();
+
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        DictionaryResponse addressResponse = DictionaryResponse.builder()
+                .entries(Collections.singletonList("Novartis Crop Protection AG, Basel, Switzerland"))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(6);
+        }
+    }
+
+
+    @Before
+    public void stubRedaction() {
+        String tableRules = "package drools\n" +
+                "\n" +
+                "import com.iqser.red.service.redaction.v1.server.redaction.model.Section\n" +
+                "\n" +
+                "global Section section\n" +
+                "rule \"8: Not redacted because Vertebrate Study = N\"\n" +
+                "    when\n" +
+                "        Section(rowEquals(\"Vertebrate study Y/N\", \"N\") || rowEquals(\"Vertebrate study Y/N\", \"No\"))\n" +
+                "    then\n" +
+                "        section.redactNot(\"name\", 8, \"Not redacted because row is not a vertebrate study\");\n" +
+                "        section.redactNot(\"address\", 8, \"Not redacted because row is not a vertebrate study\");\n" +
+                "        section.highlightCell(\"Vertebrate study Y/N\", 8, \"hint_only\");\n" +
+                "    end\n" +
                "rule \"9: Redact Authors and Addresses in Reference Table, if it is a Vertebrate study\"\n" +
                "    when\n" +
-                "        Section(tabularData != null && tabularData.size() > 0\n" +
-                "            && tabularData.containsKey(\"Vertebrate study Y/N\")\n" +
-                "            && tabularData.get(\"Vertebrate study Y/N\").equals(\"Y\")\n" +
-                "        )\n" +
+                "        Section(rowEquals(\"Vertebrate study Y/N\", \"Y\") || rowEquals(\"Vertebrate study Y/N\", " +
+                "\"Yes\"))\n" +
                "    then\n" +
                "        section.redact(\"name\", 9, \"Redacted because row is a vertebrate study\");\n" +
-                "        section.redact(\"address\", 9, \"Redacted because rows is a vertebrate study\");\n" +
-                "        section.highlightCell(\"Vertebrate study Y/N\", 9);\n" +
+                "        section.redact(\"address\", 9, \"Redacted because row is a vertebrate study\");\n" +
+                "        section.highlightCell(\"Vertebrate study Y/N\", 9, \"must_redact\");\n" +
                "    end";
-        when(rulesClient.getVersion()).thenReturn(1L);
+        when(rulesClient.getVersion()).thenReturn(RULES_VERSION.incrementAndGet());
        when(rulesClient.getRules()).thenReturn(new RulesResponse(tableRules));
        TypeResponse typeResponse = TypeResponse.builder()
                .types(Arrays.asList(
                        TypeResult.builder().type(NAME_CODE).color(new float[]{1, 1, 0}).build(),
                        TypeResult.builder().type(ADDRESS_CODE).color(new float[]{0, 1, 1}).build()))
                .build();
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
        when(dictionaryClient.getAllTypes()).thenReturn(typeResponse);
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(Arrays.asList("Casey, H.W.", "O’Loughlin,  C.K.", "Salamon, C.M.", "Smith, S.H."))
-                .build();
-        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(Collections.singletonList("Toxigenics, Inc., Decatur, IL 62526, USA"))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
        when(dictionaryClient.getDefaultColor()).thenReturn(new DefaultColor());
-        try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(redactionRequest.getDocument()))) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, null);
-            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-            assertThat(classifiedDoc.getEntities().get(1)).hasSize(5); // 4 out of 5 entities recognized on page 1
-        }
    }


@@ -0,0 +1,189 @@
+package com.iqser.red.service.redaction.v1.server.segmentation;
+
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.io.IOException;
+import java.util.Collections;
+import java.util.List;
+import java.util.stream.Collectors;
+
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.junit.Test;
+import org.junit.runner.RunWith;
+import org.kie.api.runtime.KieContainer;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.boot.test.context.SpringBootTest;
+import org.springframework.boot.test.mock.mockito.MockBean;
+import org.springframework.core.io.ClassPathResource;
+import org.springframework.test.context.junit4.SpringRunner;
+
+import com.iqser.red.service.redaction.v1.server.classification.model.Document;
+import com.iqser.red.service.redaction.v1.server.classification.service.BlockificationService;
+import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
+import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
+import com.iqser.red.service.redaction.v1.server.tableextraction.service.RulingCleaningService;
+import com.iqser.red.service.redaction.v1.server.tableextraction.service.TableExtractionService;
+
+@SpringBootTest
+@RunWith(SpringRunner.class)
+public class PdfSegmentationServiceTest {
+
+    @Autowired
+    private PdfSegmentationService pdfSegmentationService;
+
+    @Autowired
+    private RulingCleaningService rulingCleaningService;
+
+    @Autowired
+    private TableExtractionService tableExtractionService;
+
+    @Autowired
+    private BlockificationService blockificationService;
+
+    @MockBean
+    private KieContainer kieContainer;
+
+
+    @Test
+    public void testPDFSegmentationWithComplexTable() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Spanning Cells.pdf");
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document document = pdfSegmentationService.parseDocument(pdDocument);
+            assertThat(document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())).isNotEmpty();
+            Table table = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(0);
+            assertThat(table.getColCount()).isEqualTo(6);
+            assertThat(table.getRowCount()).isEqualTo(13);
+            assertThat(table.getRows().stream().mapToInt(List::size).sum()).isEqualTo(6 * 13);
+        }
+    }
+
+
+    @Test
+    public void testTableExtraction() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Merge Table.pdf");
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document document = pdfSegmentationService.parseDocument(pdDocument);
+            assertThat(document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())).isNotEmpty();
+            Table firstTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(0);
+            assertThat(firstTable.getColCount()).isEqualTo(8);
+            assertThat(firstTable.getRowCount()).isEqualTo(1);
+            Table secondTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(1);
+            assertThat(secondTable.getColCount()).isEqualTo(8);
+            assertThat(secondTable.getRowCount()).isEqualTo(2);
+            List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
+                    .get(0)
+                    .stream()
+                    .map(Collections::singletonList)
+                    .collect(Collectors.toList());
+            assertThat(secondTable.getRows().stream()
+                    .allMatch(row -> row.stream()
+                            .map(Cell::getHeaderCells)
+                            .collect(Collectors.toList())
+                            .equals(firstTableHeaderCells)))
+                    .isTrue();
+        }
+    }
+
+
+    @Test
+    public void testMultiPageMetadataPropagation() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Merge Multi Page Table.pdf");
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document document = pdfSegmentationService.parseDocument(pdDocument);
+            assertThat(document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())).isNotEmpty();
+            Table firstTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(0);
+            assertThat(firstTable.getColCount()).isEqualTo(9);
+            assertThat(firstTable.getRowCount()).isEqualTo(5);
+            Table secondTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(1);
+            assertThat(secondTable.getColCount()).isEqualTo(9);
+            assertThat(secondTable.getRowCount()).isEqualTo(6);
+            List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
+                    .get(firstTable.getRowCount() - 1)
+                    .stream()
+                    .map(Cell::getHeaderCells)
+                    .collect(Collectors.toList());
+            assertThat(secondTable.getRows().stream()
+                    .allMatch(row -> row.stream()
+                            .map(Cell::getHeaderCells)
+                            .collect(Collectors.toList())
+                            .equals(firstTableHeaderCells)))
+                    .isTrue();
+        }
+    }
+
+
+    @Test
+    public void testHeaderCellsForRotatedTable() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Rotated Table Headers.pdf");
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document document = pdfSegmentationService.parseDocument(pdDocument);
+            assertThat(document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())).isNotEmpty();
+            Table firstTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(0);
+            assertThat(firstTable.getColCount()).isEqualTo(8);
+            assertThat(firstTable.getRowCount()).isEqualTo(1);
+            Table secondTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(1);
+            assertThat(secondTable.getColCount()).isEqualTo(8);
+            assertThat(secondTable.getRowCount()).isEqualTo(6);
+            List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
+                    .get(0)
+                    .stream()
+                    .map(Collections::singletonList)
+                    .collect(Collectors.toList());
+            assertThat(secondTable.getRows().stream()
+                    .allMatch(row -> row.stream()
+                            .map(Cell::getHeaderCells)
+                            .collect(Collectors.toList())
+                            .equals(firstTableHeaderCells)))
+                    .isTrue();
+        }
+    }
+
+}
@@ -227,6 +227,7 @@ CIBA-GEIGY Limited, Toxicology Services, Short-term Toxicology, 4332 Stein, Swit
 CIBA-GEIGY Ltd. CH-4002 Basle, Switzerland
 CIBA-GEIGY Ltd., Product Safety, Ecotoxicology, CH-4002 Basel, Switzerland
 CIBA-GEIGY Ltd., Switzerland
+CibaGeigy Ltd., CH-4002, Basel, Switzerland
 CIP Chemisches Institut Pforzheim GmbH, Pforzheim, Germany
 CL PHARMA AG
 CL PHARMA AG, Linz, Austria
@@ -1563,4 +1564,36 @@ Zyma SA
 Zyma SA, Nyon, Switzerland
 “Bayer CropScience AG” Monheim
 “Biologische Bundesanstalt (BBA)”, Berlin-Dahlem
-“W. Neudorff GmbH KG”, An der Mühle 3, D- 31860 Emmertal
+“W. Neudorff GmbH KG”, An der Mühle 3, D- 31860 Emmertal
+Syngenta Ltd Jealott’s Hill Int. Research Centre Bracknell, Berkshire, RG42 6EY, United Kingdom
+MITOX Consultants Science Park 408, 1098XH Amsterdam, NL
+Eurofins Agrosciences Services Eco-Chem GmbH, Eutinger Str. 24, 75233 Niefern-Öschelbronn, Germany
+Fraunhofer Institute for Molecular Biology and Applied Ecology (IME), 57392 Schmallenberg, Germany
+Syngenta Crop Protection AGRCC Cytotest Cell Research GmbH, Germany
+BASF SE, Ludwigshafen, Germany
+AGRCC Ltd., Switzerland
+AGWIL Research Lab. Inc., USA
+Charles River Laboratories, UK
+Ciba-Geigy Corporation Environmental Health Centre, Farmington
+Syngenta Crop Protection, Inc., Greensboro, United States
+Experimental Toxicology and Ecology, BASF SE
+Wildlife International, Ltd. 8598 Commerce Drive Easton, MD 21601 USA
+Küberich, Wiesentheid / Geesdorf, Germany
+CiToxLAB Hungary Ltd. H8200 Veszprém, Szabadságpuszta, Hungary
+Forellenzucht Trostadt GbR, Trostadt, Germany
+Osage Catfisheries Inc., Osage Beach, MO, USA
+Fischzucht Rhönforelle, Gersfeld, Germany
+Aquatic Biosystems, Ft. Collins, Colorado, USA
+West Aquarium GmbH, 37431 Bad Lauterberg, Germany
+Aquatic BioSystems, Fort Collins, Colorado, USA
+“U.S. Environmental Research Laboratory”, Duluth, Minnesota, USA
+Continuous laboratory cultures
+Wildlife International, Ltd. 8598 Commerce Drive Easton, MD 21601 USA
+Osage Catfisheries, Inc., Osage Beach, MO, USA
+Osage catfisheries, Osage Beach (MO), USA
+Ciba-Geigy Corporation, Environmental Health Centre, Farmington
+Novartis Crop Protection Inc., Greensboro, United States
+Syngenta, Jealott’s Hill International Research Station, UK
+Hillandale Farms, Kenansville, Florida, USA
+Jealott’s Hill International Research Station, UK
+L & C Dairy, Avon Park, Florida, UK
@@ -1,3 +1,2 @@
-Batches Produced at
 CTL
-for determination of residues
+determination of residues
@@ -5290,6 +5290,7 @@ Naeb O.
 Nagai K.
 Nagashima Yoshikazu
 Nagata H.|Tanaka T.
+Nagra, B
 Nagra B.S
 Nagra B.|Kingdom N.
 Nagy K
@@ -5640,6 +5641,7 @@ Pedersen Carol
 Pedersen Carol|DuCharme Darlene
 Pedersen C|Ducharme D.
 Pederson C.A.
+Pederson
 Peffer
 Peffer R
 Peffer R.
@@ -6249,6 +6251,7 @@ Roth M.
 Roth Markus
 Roth RN|Heilman RD
 Roth, M.
+Roth
 Rotroff
 Rouaud, J. L.
 Roussel C
@@ -6848,7 +6851,7 @@ Sokolowshi Andrea
 Sokolowski A
 Sokolowski A.
 Sokolowski Andrea
-Sokolowski,
+Sokolowski
 Sole
 Sole C
 Sole, C.
@@ -7248,6 +7251,7 @@ Thanei, P.
 Tharp BE
 Thede B.
 Theophilidis G
+Thevenaz
 Thevenaz P
 Thevenaz P.
 Thevenaz Ph.
@@ -8193,4 +8197,113 @@ Zoriki Hosomi R.
 Zoriki Hosomi Rosana
 Zuberer D
 Zubrod J
-Zwicker R.E.
+Zwicker R.E.
+Ott
+Argus et al
+Mihail
+Clapp
+Murfitt R
+Yang
+Swarbrick
+Hertl
+Breitwieser
+Seville, A
+Trumper, C.
+Gasso-Brown, D.
+Váliczkó, É
+Váliczkó, É.
+Durando, J.
+Tarcai, Z.
+Pooles, A
+Roig, J.
+Schäfers, C.
+Nixon, W.
+Zok, S.
+Durjava M.K. et al
+Jatzek J
+Salinas E
+Sousa J. V
+Caferella, M.A.
+Salina, E.
+Walraven J
+Kaspers et al
+Schneider et al
+Lowrie C
+Macdougall J
+Roberts S
+Madrid, S.O.
+Maynard, M.S.
+Ray, W.
+Ray, W.J.
+Váliczkó É
+Tarcai Z
+Orovecz B
+Kaspers
+Tarcai Z.
+Gizler, A
+Whitlow
+Johnson et al
+Bryden et al
+Váliczkó
+Höger
+Dorgerloh
+Schäfers C
+Caferella M.A
+Salinas, E
+Zok S
+Tummon OJ
+Eckstein CL
+Francis PD
+Ott M
+Richeux F
+Rajsekhar PV
+Webbley
+O’Hagan
+MacDonald
+Jewkes
+Harris
+Tomlinson
+Tomlinson et al
+Nagy
+Török-Bathó
+Hargitai
+Shearer
+Robertson
+Strepka
+Hackford
+Blunt
+Sieber
+Elcombe
+Haines
+Kang
+Sayers
+Harris J
+Baxter A
+Macdonald M
+PetusArpasy M
+TorokBatho M.
+van Ravenzwaa y B
+Cords S
+Dymarkows ka K.
+Bahig M
+Pekari K
+Bercz J
+Russell L
+Kitchen K
+Miyagawa M.
+Sasaki Y
+Yin D
+Huff J
+Blackburn K
+Carlson G
+Pereira M
+Stoner G
+Zelenak V
+Valiczko E
+Tomlinso n J
+Trumper C
+Senciuc M
+Vance C.
+Przeorska J.
+Melville S.
+Wicksted G.
@@ -0,0 +1,7 @@
+Monthey Syngenta Crop Protection AG, Basel, Switzerland
+Syngenta Crop Protection, Monthey, Switzerland
+Fine Organics Limited, Middlesbrough, United Kingdom
+Syngenta Monthey Switzerland
+Hunan Haili Chemical Industry Co., Ltd., Hunan, China
+Syngenta, Switzerland
+Syngenta Nantong, China
@@ -7,7 +7,7 @@ global Section section

 rule "1: Redacted because Section contains Vertebrate"
 	when
-		eval(section.contains("vertebrate")==true);
+		Section(matchesType("vertebrate"))
 	then
 		section.redact("name", 1, "Redacted because Section contains Vertebrate");
 		section.redact("address", 1, "Redacted because Section contains Vertebrate");
@@ -16,7 +16,7 @@ rule "1: Redacted because Section contains Vertebrate"

 rule "2: Not Redacted because Section contains no Vertebrate"
 	when
-		eval(section.contains("vertebrate")==false);
+		Section(matchesType("vertebrate"))
 	then
 		section.redactNot("name", 2, "Not Redacted because Section contains no Vertebrate");
 		section.redactNot("address", 2, "Not Redacted because Section contains no Vertebrate");
@@ -25,7 +25,7 @@ rule "2: Not Redacted because Section contains no Vertebrate"

 rule "3: Do not redact Names and Addresses if no redaction Indicator is contained"
 	when
-		eval(section.contains("vertebrate")==true && section.contains("no_redaction_indicator")==true);
+		Section(matchesType("vertebrate"), matchesType("no_redaction_indicator"))
 	then
 		section.redactNot("name", 3, "Vertebrate was found, but also a no redaction indicator");
 		section.redactNot("address", 3, "Vertebrate was found, but also a no redaction indicator");
@@ -34,7 +34,7 @@ rule "3: Do not redact Names and Addresses if no redaction Indicator is containe

 rule "4: Redact Names and Addresses if no_redaction_indicator and redaction_indicator is contained"
 	when
-		eval(section.contains("vertebrate")==true && section.contains("no_redaction_indicator")==true && section.contains("redaction_indicator")==true);
+		Section(matchesType("vertebrate"), matchesType("no_redaction_indicator"), matchesType("redaction_indicator"))
 	then
 		section.redact("name", 4, "Vertebrate was found and no_redaction_indicator and redaction_indicator");
 		section.redact("address", 4, "Vertebrate was found and no_redaction_indicator and redaction_indicator");
@@ -43,70 +43,92 @@ rule "4: Redact Names and Addresses if no_redaction_indicator and redaction_indi

 rule "5: Do not redact in guideline sections"
 	when
-		eval(section.headlineContainsWord("guideline") || section.headlineContainsWord("Guidance"));
+		Section(headlineContainsWord("guideline") || headlineContainsWord("Guidance"))
 	then
 		section.redactNot("name", 5, "Section is a guideline section.");
 		section.redactNot("address", 5, "Section is a guideline section.");
 	end

-rule "6: Redact if must redact entry is found"
+
+rule "6: Redact contact information if applicant is found"
 	when
-		eval(section.contains("must_redact")==true);
+		Section(headlineContainsWord("applicant") || text.contains("Applicant") || headlineContainsWord("Primary contact") || headlineContainsWord("Alternative contact"))
 	then
-		section.redact("name", 6, "must_redact entry was found.");
-		section.redact("address", 6, "must_redact entry was found.");
+		section.redactLineAfter("Name:", "address", 6, "Applicant information was found");
+		section.redactBetween("Address:", "Contact", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Contact point:",  "address", 6, "Applicant information was found");
+		section.redactLineAfter("Phone:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Fax:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Tel.:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Tel:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("E-mail:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Email:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("e-mail:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("E-mail address:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Contact:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Alternative contact:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Telephone number:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Telephone No:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Fax number:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Telephone:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Company:", "address", 6, "Applicant information was found");
+		section.redactBetween("No:", "Fax", "address", 6, "Applicant information was found");
+		section.redactBetween("Contact:", "Tel.:", "address", 6, "Applicant information was found");
 	end


-rule "7: Redact contact information, if applicant is found"
+rule "7: Redact contact information if Producer is found"
 	when
-		eval(section.headlineContainsWord("applicant") || section.getText().contains("Applicant"));
+		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("Name:", "address", 7, "Applicant information was found");
-		section.redactBetween("Address:", "Contact", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Contact point:",  "address", 7, "Applicant information was found");
-		section.redactLineAfter("Phone:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Fax:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Tel.:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Tel:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("E-mail:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Email:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Contact:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Telephone number:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Fax number:", "address", 7, "Applicant information was found");
-		section.redactLineAfter("Telephone:", "address", 7, "Applicant information was found");
-		section.redactBetween("No:", "Fax", "address", 7, "Applicant information was found");
-		section.redactBetween("Contact:", "Tel.:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Name:", "address", 7, "Producer was found");
+		section.redactBetween("Address:", "Contact", "address", 7, "Producer was found");
+		section.redactBetween("Contact:", "Phone", "address", 7, "Producer was found");
+		section.redactBetween("Contact:", "Telephone number:", "address", 7, "Producer was found");
+		section.redactBetween("Address:", "Manufacturing", "address", 7, "Producer was found");
+		section.redactLineAfter("Telephone:", "address", 7, "Producer was found");
+		section.redactLineAfter("Phone:", "address", 7, "Producer was found");
+		section.redactLineAfter("Fax:", "address", 7, "Producer was found");
+		section.redactLineAfter("E-mail:", "address", 7, "Producer was found");
+		section.redactLineAfter("Contact:", "address", 7, "Producer was found");
+		section.redactLineAfter("Fax number:", "address", 7, "Producer was found");
+		section.redactLineAfter("Telephone number:", "address", 7, "Producer was found");
+		section.redactLineAfter("Tel:", "address", 7, "Producer was found");
+		section.redactBetween("No:", "Fax", "address", 7, "Producer was found");
 	end

-rule "8: Redact contact information, if Producer is found"
-	when
-		eval(section.getText().toLowerCase().contains("producer of the plant protection") || section.getText().toLowerCase().contains("producer of the active substance") || section.getText().contains("Manufacturer of the active substance") || section.getText().contains("Manufacturer:") || section.getText().contains("Producer or producers of the active substance"));
-	then
-		section.redactLineAfter("Name:", "address", 8, "Producer was found");
-		section.redactBetween("Address:", "Contact", "address", 8, "Producer was found");
-		section.redactBetween("Contact:", "Phone", "address", 8, "Producer was found");
-		section.redactBetween("Contact:", "Telephone number:", "address", 8, "Producer was found");
-		section.redactBetween("Address:", "Manufacturing", "address", 8, "Producer was found");
-		section.redactLineAfter("Telephone:", "address", 8, "Producer was found");
-		section.redactLineAfter("Phone:", "address", 8, "Producer was found");
-		section.redactLineAfter("Fax:", "address", 8, "Producer was found");
-		section.redactLineAfter("E-mail:", "address", 8, "Producer was found");
-		section.redactLineAfter("Contact:", "address", 8, "Producer was found");
-		section.redactLineAfter("Fax number:", "address", 8, "Producer was found");
-		section.redactLineAfter("Telephone number:", "address", 8, "Producer was found");
-		section.redactLineAfter("Tel:", "address", 8, "Producer was found");
-		section.redactBetween("No:", "Fax", "address", 8, "Producer was found");
-	end

-rule "9: Redact Authors and Addresses in Reference Table, if it is a Vertebrate study"
+rule "8: Not redacted because Vertebrate Study = N"
    when
-        Section(tabularData != null
-            && tabularData.containsKey("Vertebrate study Y/N")
-            && tabularData.get("Vertebrate study Y/N").equals("Y")
-        )
+        Section(rowEquals("Vertebrate study Y/N", "N") || rowEquals("Vertebrate study Y/N", "No"))
    then
-        section.redact("name", 9, "Redacted because row is a vertebrate study");
-        section.redact("address", 9, "Redacted because rows is a vertebrate study");
-        section.highlightCell("Vertebrate study Y/N", 9);
+        section.redactNotCell("Author(s)", 8, "name", "Not redacted because row is not a vertebrate study");
+        section.redactNot("address", 8, "Not redacted because row is not a vertebrate study");
+        section.highlightCell("Vertebrate study Y/N", 8, "hint_only");
+    end
+
+
+rule "9: Redact if must redact entry is found"
+	when
+		Section(matchesType("must_redact"))
+	then
+		section.redact("name", 9, "must_redact entry was found.");
+		section.redact("address", 9, "must_redact entry was found.");
+	end
+
+
+rule "10: Redact Authors and Addresses in Reference Table if it is a Vertebrate study"
+    when
+        Section(rowEquals("Vertebrate study Y/N", "Y") || rowEquals("Vertebrate study Y/N", "Yes"))
+    then
+        section.redactCell("Author(s)", 10, "name", "Redacted because row is a vertebrate study");
+        section.redact("address", 10, "Redacted because row is a vertebrate study");
+        section.highlightCell("Vertebrate study Y/N", 10, "must_redact");
+    end
+
+rule "11: Redact sponsor company"
+    when
+        Section(text.toLowerCase().contains("batches produced at"))
+    then
+        section.redactIfPrecededBy("batches produced at", "sponsor", 11, "Redacted because it represents a sponsor company");
    end
Author	SHA1	Message	Date
Dominique Eiflaender	0c1d8ba33d	Pull request #48 : RED-372: Suffix cross page annotations with -1, -2, etc., to avoid duplicate annotation ids Merge in RED/redaction-service from RED-372 to master * commit 'ecdfd29803e09ac57cc74b19408db54d9a8247a7': RED-372: Suffix cross page annotations with -1, -2, etc., to avoid duplicate annotation ids	2020-10-01 13:01:24 +02:00
deiflaender	ecdfd29803	RED-372: Suffix cross page annotations with -1, -2, etc., to avoid duplicate annotation ids	2020-10-01 12:48:44 +02:00
Dominique Eiflaender	107dbe1910	Pull request #47 : RED-373: Use a single multiline annotation instead of seperate annotations per line for one Entity Merge in RED/redaction-service from RED-373 to master * commit 'dd02296face3c32bf7675301cf9969103e65f52f': Changed formating RED-373: Use a single multiline annotation instead of seperate annotations per line for one Entity	2020-10-01 12:47:58 +02:00
deiflaender	dd02296fac	Changed formating	2020-10-01 11:55:38 +02:00
Thierry Goeckel	589458a112	Pull request #46 : Add rule redacting sponsor companies if preceded by prefix Merge in RED/redaction-service from RED-301 to master * commit '6a8f3665198ec2d7452ed0dac43b9a7d220c8bc5': Remove entries from must_redact dict, add test and refactor rules Add rule redacting sponsor companies if preceded by prefix	2020-10-01 11:44:41 +02:00
deiflaender	ce22121802	RED-373: Use a single multiline annotation instead of seperate annotations per line for one Entity	2020-10-01 11:15:28 +02:00
Thierry Göckel	6a8f366519	Remove entries from must_redact dict, add test and refactor rules	2020-09-30 16:08:56 +02:00
Thierry Göckel	cec5fd3d5e	Add rule redacting sponsor companies if preceded by prefix	2020-09-30 14:13:17 +02:00
Dominique Eiflaender	99bde2956f	Pull request #45 : RED-277: Upgraded to pdfbox 2.0.21 Merge in RED/redaction-service from RED-277 to master * commit 'ecc303e79e82ed698f1c8bdd37dbd521335428bc': RED-277: Upgraded to pdfbox 2.0.21	2020-09-29 16:17:32 +02:00
deiflaender	ecc303e79e	RED-277: Upgraded to pdfbox 2.0.21	2020-09-29 15:18:58 +02:00
Dominique Eiflaender	384cd1d0dc	Pull request #44 : Fix annotating single letter entities Merge in RED/redaction-service from bugfix/single-letter-annotations to master * commit '8e5ff318106e700e28abd14a0eb1337d3a44c109': Fix annotating single letter entities	2020-09-29 15:17:51 +02:00
Thierry Göckel	8e5ff31810	Fix annotating single letter entities	2020-09-29 15:10:52 +02:00
Dominique Eiflaender	8b1574845c	Pull request #43 : RED-264: Avoid phantom cells, by merging line till rounded value of biggest char height/width Merge in RED/redaction-service from RED-264 to master * commit 'e1adcfb02c4c7d46140884b3d38f8daf1e1aae92': Added missing unittest RED-264: Avoid phantom cells, by merging line till rounded value of biggest char height/width	2020-09-29 15:04:20 +02:00
deiflaender	e1adcfb02c	Added missing unittest	2020-09-29 14:54:35 +02:00
deiflaender	11164219c5	RED-264: Avoid phantom cells, by merging line till rounded value of biggest char height/width	2020-09-29 14:21:42 +02:00
Thierry Goeckel	c9516205fe	Pull request #42 : RED-318 Merge in RED/redaction-service from RED-318 to master * commit 'f5790bccab3fb8dce66aeb9fbe501655ae6b86c3': Log entries and manual redaction require one position per line Annotate line by line Refactor multiple positions to one Merge rectangles for single annotation/redaction log entry	2020-09-29 14:15:21 +02:00
Thierry Göckel	f5790bccab	Log entries and manual redaction require one position per line	2020-09-29 13:54:32 +02:00
Thierry Göckel	74e63ca292	Annotate line by line	2020-09-29 13:43:16 +02:00
Thierry Göckel	6638f0cf9e	Refactor multiple positions to one	2020-09-28 17:47:55 +02:00
Thierry Göckel	5448153096	Merge rectangles for single annotation/redaction log entry	2020-09-28 17:31:40 +02:00
Dominique Eiflaender	03274ba1ab	Pull request #41 : RED-299: Redact complete Author(s) Field in Vertebrate study tables Merge in RED/redaction-service from RED-299 to master * commit '664b9b420605cb1adf1113b34c52a45e8024dbbc': RED-299: Redact complete Author(s) Field in Vertebrate study tables	2020-09-23 15:34:46 +02:00
deiflaender	664b9b4206	RED-299: Redact complete Author(s) Field in Vertebrate study tables	2020-09-23 15:20:52 +02:00
Thierry Goeckel	564d74e39d	Pull request #40 : Test modified rules 8 and 10 Merge in RED/redaction-service from RED-288 to master * commit 'cfb7554c618b9482f529b8183bc0f1f95df7d00f': Test modified rules 8 and 9	2020-09-23 11:44:35 +02:00
Thierry Göckel	cfb7554c61	Test modified rules 8 and 9	2020-09-22 21:36:05 +02:00
Christoph Schabert	ee40064b32	Pull request #39 : RED-316: Switch to new red base image Merge in RED/redaction-service from RED-316 to master * commit 'f2fafc3db2ab8825eb4a2136e8eab28a12dbc2e6': Switch to rootless base-image Upgrade bamboo-specs to 7.1.2	2020-09-16 08:59:45 +02:00
Christoph Schabert	f2fafc3db2	Switch to rootless base-image	2020-09-15 16:02:08 +02:00
Christoph Schabert	d5b0638f11	Upgrade bamboo-specs to 7.1.2	2020-09-15 16:01:26 +02:00
Dominique Eiflaender	130cac0b77	Pull request #38 : RED-293: Add unkown Textblocks to paragraphs Merge in RED/redaction-service from RED-293 to master * commit 'de310e8a654b601cb4bf258675904a25d4233783': RED-293: Add unkown Textblocks to paragraphs	2020-09-14 14:39:36 +02:00
deiflaender	de310e8a65	RED-293: Add unkown Textblocks to paragraphs	2020-09-14 13:52:16 +02:00
Dominique Eiflaender	a0a78440d8	Pull request #37 : RED-305: Added ids to manual redaction entries, RED-306: set redacet to false in redactionLog for manual deletions Merge in RED/redaction-service from RED-305 to master * commit 'e52a54b0a3eb8491ed3e79b3f514364a5cf22f76': RED-305: Added ids to manual redaction entries, RED-306: set redacet to false in redactionLog for manual deletions	2020-09-11 11:18:42 +02:00
deiflaender	e52a54b0a3	RED-305: Added ids to manual redaction entries, RED-306: set redacet to false in redactionLog for manual deletions	2020-09-11 11:10:10 +02:00
deiflaender	65a29fb3a5	Added new files and missing dictionary entries	2020-09-07 16:06:25 +02:00
Dominique Eiflaender	ff0a73c635	Pull request #36 : Added filename to redactionLog Merge in RED/redaction-service from filename to master * commit 'c27c9b2f5902d66f654982d7e5f005ecc176e01b': Added missing constuctor for RedactionLog Added filename to redactionLog	2020-09-02 15:22:34 +02:00
deiflaender	c27c9b2f59	Added missing constuctor for RedactionLog	2020-09-02 14:56:48 +02:00
deiflaender	d95ff5e5cf	Added filename to redactionLog	2020-09-02 14:47:58 +02:00
Thierry Goeckel	34e058c4e4	Pull request #35 : Fix redaction in single cell tables Merge in RED/redaction-service from RED-279 to master * commit '3d5455d7297e5805ad30a578d00d6d1737820c7d': Fix redaction in single cell tables	2020-09-02 14:39:08 +02:00
Thierry Göckel	3d5455d729	Fix redaction in single cell tables	2020-09-02 14:31:38 +02:00
Thierry Goeckel	b07ebf78d2	Pull request #34 : RED-267: Return dictionary and rules version in redactionLog Merge in RED/redaction-service from RED-267 to master * commit 'b5fa771c285554132edc32e1df59836fd7748b9a': RED-267: Return dictionary and rules version in redactionLog	2020-09-02 08:54:43 +02:00
deiflaender	b5fa771c28	RED-267: Return dictionary and rules version in redactionLog	2020-09-02 08:47:05 +02:00
Thierry Goeckel	81ea2e91ef	Pull request #33 : Fix entity span in table rows and detection of headers in rotated tables Merge in RED/redaction-service from bugfix/rowspan-and-header-in-rotated-table-fix to master * commit '4954aafed78e06484531d6264bdf215176ee0ef2': Reduce log level. Remove unused import Adjust test to added rule and fix vertical header propagation for row > 2 Fix entity span in table rows and detection of headers in rotated tables	2020-08-25 17:30:57 +02:00
Thierry Göckel	4954aafed7	Reduce log level.	2020-08-25 17:21:56 +02:00
Thierry Göckel	0baf48698d	Remove unused import	2020-08-25 15:45:24 +02:00
Thierry Göckel	3c590dcf1d	Adjust test to added rule and fix vertical header propagation for row > 2	2020-08-25 15:41:44 +02:00
Thierry Göckel	848c506c3f	Fix entity span in table rows and detection of headers in rotated tables	2020-08-25 13:50:34 +02:00
Thierry Goeckel	6483e637c6	Pull request #32 : Remove '-' in headernames Merge in RED/redaction-service from tablesHeader3 to master * commit '272c7cc2287b146664b8bb277c3731262ac78a19': Remove '-' in headernames	2020-08-24 16:15:44 +02:00
deiflaender	272c7cc228	Remove '-' in headernames	2020-08-24 15:39:50 +02:00
Thierry Goeckel	5586531c33	Pull request #31 : Made rules for table more stabil and flexible Merge in RED/redaction-service from tablesRules2 to master * commit '559c42154276ea861c09cf29b71d9bb9c17e6e87': Made rules for table more stabil and flexible	2020-08-24 14:52:29 +02:00
deiflaender	559c421542	Made rules for table more stabil and flexible	2020-08-24 14:45:30 +02:00
Dominique Eiflaender	d026f4e1db	Pull request #30 : Fix test and logic of same page table merge for one row tables Merge in RED/redaction-service from bugfix/fix-test-table-merge to master * commit '38ddb1d9c8dbf0e83d40ec916f1ebeb6d0f86f42': Fix test and logic of same page table merge for one row tables	2020-08-24 14:22:41 +02:00
Thierry Göckel	38ddb1d9c8	Fix test and logic of same page table merge for one row tables	2020-08-24 14:06:34 +02:00
Thierry Goeckel	92f0af9d21	Pull request #28 : Highlight N in vertebrate study tables and fixed rule order Merge in RED/redaction-service from tableRules to master * commit 'afa26895c8e78f261ec3700c515f412d8cf605ce': Fixed test problem Fixed pmd Highlight N in vertebrate study tables and fixed rule order	2020-08-24 13:15:11 +02:00
Dominique Eiflaender	b24922aa60	Pull request #29 : Fix multi-page table merge Merge in RED/redaction-service from bugfix/multi-page-table-merge to master * commit 'baa703928f50e53a66f925fcf72082df188482a1': Fix multi-page table merge	2020-08-24 13:09:43 +02:00
deiflaender	afa26895c8	Fixed test problem	2020-08-24 13:05:23 +02:00
Thierry Göckel	baa703928f	Fix multi-page table merge	2020-08-24 13:03:03 +02:00
deiflaender	85667c91e4	Fixed pmd	2020-08-24 12:36:49 +02:00
deiflaender	8a471d0771	Highlight N in vertebrate study tables and fixed rule order	2020-08-24 12:24:48 +02:00
Dominique Eiflaender	bb1112d0d7	Pull request #27 : Fix merging of tables with single header row table Merge in RED/redaction-service from bugfix/merge-single-row-tables to master * commit '00a960ee2308e7ed7da23ebd3e85f5ef9db14610': Fix merging of tables with single header row table	2020-08-24 12:10:53 +02:00
Thierry Göckel	00a960ee23	Fix merging of tables with single header row table	2020-08-24 11:43:19 +02:00
Dominique Eiflaender	8c08bb3664	Pull request #25 : RED-101 & RED-102: Make table structure detection more robust Merge in RED/redaction-service from bugfix/RED-101 to master * commit '76369f13f8d41154f3dd3690af81b9567ca1133e': Remove redundant comment RED-101: Add features as requested in PR Add unit test for table structure requirements Make table structure detection more robust	2020-08-24 11:09:17 +02:00
Cheng Zhu	14b4f4ab8a	Pull request #26 : Feature/RED-116: replace gin4-commons with red-commons Merge in RED/redaction-service from feature/RED-116 to master * commit '4206e708a20a862083f03d27a992e8661c44abc2': RED-116: replace gin4-commons with red-commons RED-116: replace gin4-commons with red-commons	2020-08-20 17:22:36 +02:00
cheng	4206e708a2	RED-116: replace gin4-commons with red-commons	2020-08-20 16:52:56 +02:00
cheng	c4164b57fb	RED-116: replace gin4-commons with red-commons	2020-08-20 16:51:02 +02:00
Thierry Göckel	76369f13f8	Remove redundant comment	2020-08-18 20:53:41 +02:00
Thierry Göckel	1fff4f7eb0	RED-101: Add features as requested in PR - Add rule for non-vertebrate studies - Don't highlight ambiguous cell values - Rewrite rules in more human-readable fashion	2020-08-18 20:35:40 +02:00
Thierry Göckel	c7f5b4a280	Add unit test for table structure requirements	2020-08-18 20:35:17 +02:00
Thierry Göckel	4236fa05cc	Make table structure detection more robust	2020-08-13 17:48:37 +02:00