Adjust imports

Adjust pom
Add logs
2021-02-09 10:36:58 +01:00 · 2021-02-09 10:21:34 +01:00 · 2021-02-08 16:22:26 +01:00 · 2021-02-08 14:50:02 +01:00 · 2021-02-08 14:09:11 +01:00 · 2021-02-05 15:08:15 +01:00
25 changed files with 509 additions and 99 deletions
@@ -10,6 +10,9 @@
    </parent>

    <artifactId>redaction-service-server-v1</artifactId>
+    <properties>
+        <slf4j.version>1.7.30</slf4j.version>
+    </properties>

    <dependencies>
        <dependency>
@@ -68,6 +71,11 @@
            <groupId>org.apache.pdfbox</groupId>
            <artifactId>pdfbox-tools</artifactId>
        </dependency>
+        <dependency>
+            <groupId>org.slf4j</groupId>
+            <artifactId>slf4j-api</artifactId>
+            <version>${slf4j.version}</version>
+        </dependency>
        <!-- spring -->
        <dependency>
            <groupId>org.springframework.cloud</groupId>
@@ -18,6 +18,8 @@ public class Document {

    private List<Page> pages = new ArrayList<>();
    private List<Paragraph> paragraphs = new ArrayList<>();
+    private List<Header> headers = new ArrayList<>();
+    private List<Footer> footers = new ArrayList<>();
    private Map<Integer, List<Entity>> entities = new HashMap<>();
    private FloatFrequencyCounter textHeightCounter = new FloatFrequencyCounter();
    private FloatFrequencyCounter fontSizeCounter= new FloatFrequencyCounter();
@@ -0,0 +1,24 @@
+package com.iqser.red.service.redaction.v1.server.classification.model;
+
+import java.util.List;
+
+import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+
+@Data
+@AllArgsConstructor
+public class Footer {
+
+    private List<TextBlock> textBlocks;
+
+
+    public SearchableText getSearchableText() {
+
+        SearchableText searchableText = new SearchableText();
+        textBlocks.forEach(block -> searchableText.addAll(block.getSequences()));
+        return searchableText;
+    }
+
+}
@@ -0,0 +1,24 @@
+package com.iqser.red.service.redaction.v1.server.classification.model;
+
+import java.util.List;
+
+import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+
+@Data
+@AllArgsConstructor
+public class Header {
+
+    private List<TextBlock> textBlocks;
+
+
+    public SearchableText getSearchableText() {
+
+        SearchableText searchableText = new SearchableText();
+        textBlocks.forEach(block -> searchableText.addAll(block.getSequences()));
+        return searchableText;
+    }
+
+}
@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.classification.model;

+import java.awt.geom.Rectangle2D;
 import java.util.List;

 import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
@@ -16,6 +17,8 @@ public class Page {
    @NonNull
    private List<AbstractTextContainer> textBlocks;

+    private List<Rectangle2D> imageBounds;
+
    private Rectangle bodyTextFrame;

    private boolean landscape;
@@ -51,6 +51,8 @@ public class RedactionController implements RedactionResource {
    @Override
    public AnalyzeResult analyze(@RequestBody AnalyzeRequest analyzeRequest) {

+        log.info("Starting redaction analysis...");
+
        try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(analyzeRequest.getDocument()))) {
            pdDocument.setAllSecurityToBeRemoved(true);

@@ -64,8 +66,12 @@ public class RedactionController implements RedactionResource {

            log.info("Redaction analysis successful...");

-            return AnalyzeResult.builder().sectionGrid(classifiedDoc.getSectionGrid()).redactionLog(new RedactionLog(classifiedDoc.getRedactionLogEntities(), classifiedDoc.getDictionaryVersion(), classifiedDoc
-                    .getRulesVersion(), analyzeRequest.getRuleSetId())).build();
+            return AnalyzeResult.builder()
+                    .sectionGrid(classifiedDoc.getSectionGrid())
+                    .redactionLog(new RedactionLog(classifiedDoc.getRedactionLogEntities(), classifiedDoc.getDictionaryVersion(), classifiedDoc
+                    .getRulesVersion(), analyzeRequest.getRuleSetId()))
+                    .numberOfPages(classifiedDoc.getPages().size())
+                    .build();

        } catch (Exception e) {
            throw new RedactionException(e);
@@ -1,6 +1,8 @@
 package com.iqser.red.service.redaction.v1.server.parsing;

+import java.awt.geom.AffineTransform;
 import java.awt.geom.Point2D;
+import java.awt.geom.Rectangle2D;
 import java.io.IOException;
 import java.util.ArrayList;
 import java.util.List;
@@ -28,10 +30,15 @@ import org.apache.pdfbox.contentstream.operator.state.SetLineWidth;
 import org.apache.pdfbox.contentstream.operator.state.SetRenderingIntent;
 import org.apache.pdfbox.contentstream.operator.text.SetFontAndSize;
 import org.apache.pdfbox.cos.COSBase;
+import org.apache.pdfbox.cos.COSName;
 import org.apache.pdfbox.cos.COSNumber;
 import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.pdmodel.graphics.PDXObject;
+import org.apache.pdfbox.pdmodel.graphics.image.PDImageXObject;
 import org.apache.pdfbox.text.PDFTextStripper;
 import org.apache.pdfbox.text.TextPosition;
+import org.apache.pdfbox.util.Matrix;

 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Ruling;
@@ -43,8 +50,17 @@ import lombok.extern.slf4j.Slf4j;
@Slf4j
 public class PDFLinesTextStripper extends PDFTextStripper {

+    @Setter
+    protected PDPage pdpage;
+
    @Getter
-    private int maxCharWidths;
+    private int minCharWidth;
+
+    @Getter
+    private int maxCharWidth;
+    
+    @Getter
+    private int minCharHeight;

    @Getter
    private int maxCharHeight;
@@ -57,13 +73,18 @@ public class PDFLinesTextStripper extends PDFTextStripper {

    private final List<Ruling> graphicsPath = new ArrayList<>();

+    @Getter
+    private List<Rectangle2D> imageBounds = new ArrayList<>();
+
    private float path_x;
    private float path_y;

    @Setter
    private int pageNumber;

+
    public PDFLinesTextStripper() throws IOException {
+
        super();
        this.addOperator(new SetStrokingColorSpace());
        this.addOperator(new SetNonStrokingColorSpace());
@@ -87,9 +108,9 @@ public class PDFLinesTextStripper extends PDFTextStripper {
        this.addOperator(new SetLineWidth());
    }

+
    @Override
-    protected void processOperator(Operator operator, List<COSBase> arguments)
-            throws IOException {
+    protected void processOperator(Operator operator, List<COSBase> arguments) throws IOException {

        String operation = operator.getName();

@@ -110,9 +131,11 @@ public class PDFLinesTextStripper extends PDFTextStripper {

                    // The direction of vertical lines must always be from bottom to top for the table extraction algorithm.
                    if (pos.getY() > path_y) {
-                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, path_y), new Point2D.Float((float) pos.getX(), (float) pos.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, path_y), new Point2D.Float((float) pos.getX(), (float) pos
+                                .getY())));
                    } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, (float) pos.getY()), new Point2D.Float((float) pos.getX(), path_y)));
+                        graphicsPath.add(new Ruling(new Point2D.Float(path_x, (float) pos.getY()), new Point2D.Float((float) pos
+                                .getX(), path_y)));
                    }

                    path_x = (float) pos.getX();
@@ -133,19 +156,25 @@ public class PDFLinesTextStripper extends PDFTextStripper {
                    Point2D p2 = transformPosition(x + width, y + height);

                    // Horizontal lines
-                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p2.getX(), (float) p1.getY())));
-                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p2.getX(), (float) p2.getY())));
+                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p2
+                            .getX(), (float) p1.getY())));
+                    graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p2
+                            .getX(), (float) p2.getY())));

                    // Vertical lines, direction must always be from bottom to top for the table extraction algorithm.
                    if (p2.getY() > p1.getY()) {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p1.getY()), new Point2D.Float((float) p2.getX(), (float) p2.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p1.getY()), new Point2D.Float((float) p2
+                                .getX(), (float) p2.getY())));
                    } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p2.getY()), new Point2D.Float((float) p2.getX(), (float) p1.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float((float) p2.getX(), (float) p2.getY()), new Point2D.Float((float) p2
+                                .getX(), (float) p1.getY())));
                    }
                    if (p2.getY() > p1.getY()) {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p1.getX(), (float) p2.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p1.getY()), new Point2D.Float((float) p1
+                                .getX(), (float) p2.getY())));
                    } else {
-                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p1.getX(), (float) p1.getY())));
+                        graphicsPath.add(new Ruling(new Point2D.Float((float) p1.getX(), (float) p2.getY()), new Point2D.Float((float) p1
+                                .getX(), (float) p1.getY())));
                    }
                }
                break;
@@ -168,12 +197,59 @@ public class PDFLinesTextStripper extends PDFTextStripper {
            case OperatorName.ENDPATH:
                graphicsPath.clear();
                break;
+
+            case OperatorName.DRAW_OBJECT:
+                processImageOperation(arguments);
+                break;
+
        }

        super.processOperator(operator, arguments);
    }

+
+    protected void processImageOperation(List<COSBase> arguments) {
+
+        try {
+            COSName objectName = (COSName) arguments.get(0);
+            PDXObject xobject = getResources().getXObject(objectName);
+            if (xobject instanceof PDImageXObject) {
+                PDImageXObject pdfImage = (PDImageXObject) xobject;
+
+                Rectangle2D imageBounds = calculateImagePosition(pdfImage);
+
+                Rectangle2D rect = new Rectangle2D.Float((float) imageBounds.getX(), (float) imageBounds.getY(), (float) imageBounds
+                        .getWidth(), (float) imageBounds.getHeight());
+
+                if (rect.getHeight() > 2 && rect.getWidth() > 2) {
+                    this.imageBounds.add(rect);
+                }
+            }
+        } catch (Exception e) {
+            log.warn("Problem during image extraction: {}", e.getMessage());
+        }
+    }
+
+
+    private Rectangle2D calculateImagePosition(PDImageXObject pdfImage) throws IOException {
+
+        Matrix ctm = getGraphicsState().getCurrentTransformationMatrix();
+
+        Rectangle2D imageBounds = pdfImage.getImage().getRaster().getBounds();
+
+        AffineTransform imageTransform = new AffineTransform(ctm.createAffineTransform());
+        imageTransform.scale(1.0 / pdfImage.getWidth(), -1.0 / pdfImage.getHeight());
+        imageTransform.translate(0, -pdfImage.getHeight());
+
+        AffineTransform pageTransform = new AffineTransform();
+        pageTransform.concatenate(imageTransform);
+
+        return pageTransform.createTransformedShape(imageBounds).getBounds2D();
+    }
+
+
    private float floatValue(COSBase value) {
+
        if (value instanceof COSNumber) {
            return ((COSNumber) value).floatValue();
        } else {
@@ -181,38 +257,56 @@ public class PDFLinesTextStripper extends PDFTextStripper {
        }
    }

+
    private Point2D.Float transformPosition(float x, float y) {
+
        return super.transformedPoint(x, y);
    }

+
    private void addVisibleRulings(List<Ruling> path, boolean stroke) throws IOException {

        try {
-            if (stroke && !getGraphicsState().getStrokingColor().isPattern() && getGraphicsState().getStrokingColor().toRGB() == 0 || !stroke && !getGraphicsState().getNonStrokingColor().isPattern() && getGraphicsState().getNonStrokingColor().toRGB() == 0) {
+            if (stroke && !getGraphicsState().getStrokingColor().isPattern() && getGraphicsState().getStrokingColor()
+                    .toRGB() == 0 || !stroke && !getGraphicsState().getNonStrokingColor()
+                    .isPattern() && getGraphicsState().getNonStrokingColor().toRGB() == 0) {
                rulings.addAll(path);
            }
        } catch (UnsupportedOperationException e) {
-            log.error("UnsupportedOperationException: " + getGraphicsState().getStrokingColor().getColorSpace().getName() + " or " + getGraphicsState().getNonStrokingColor().getColorSpace().getName() + " does not support toRGB");
+            log.error("UnsupportedOperationException: " + getGraphicsState().getStrokingColor()
+                    .getColorSpace()
+                    .getName() + " or " + getGraphicsState().getNonStrokingColor()
+                    .getColorSpace()
+                    .getName() + " does not support toRGB");
        }
    }

+
    @Override
    public void writeString(String text, List<TextPosition> textPositions) throws IOException {

        int startIndex = 0;
        for (int i = 0; i <= textPositions.size() - 1; i++) {

+            int charWidth = (int) textPositions.get(i).getWidthDirAdj();
+            if (charWidth < minCharWidth) {
+                minCharWidth = charWidth;
+            }
+            if (charWidth > maxCharWidth) {
+                maxCharWidth = charWidth;
+            }
+
            int charHeight = (int) textPositions.get(i).getHeightDir();
-            if(charHeight > maxCharHeight){
+            if (charHeight < minCharHeight) {
+                minCharHeight = charHeight;
+            }
+            if (charWidth > maxCharHeight) {
                maxCharHeight = charHeight;
            }

-            int charWidth = (int) textPositions.get(i).getWidthDirAdj();
-            if(charWidth > maxCharWidths){
-                maxCharWidths = charWidth;
-            }
-
-            if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0"))) {
+            if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
+                    .getUnicode()
+                    .equals("\u00A0"))) {
                startIndex++;
                continue;
            }
@@ -220,15 +314,21 @@ public class PDFLinesTextStripper extends PDFTextStripper {
            // Strange but sometimes this is happening, for example: Metolachlor2.pdf
            if (i > 0 && textPositions.get(i).getX() < textPositions.get(i - 1).getX()) {
                List<TextPosition> sublist = textPositions.subList(startIndex, i);
-                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0)
+                        .getUnicode()
+                        .equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
                    textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
                }
                startIndex = i;
            }

-            if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i).getUnicode().equals("\u00A0")) && i <= textPositions.size() - 2) {
+            if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
+                    .getUnicode()
+                    .equals("\u00A0")) && i <= textPositions.size() - 2) {
                List<TextPosition> sublist = textPositions.subList(startIndex, i);
-                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0)
+                        .getUnicode()
+                        .equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
                    textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
                }
                startIndex = i + 1;
@@ -236,21 +336,29 @@ public class PDFLinesTextStripper extends PDFTextStripper {
        }

        List<TextPosition> sublist = textPositions.subList(startIndex, textPositions.size());
-        if (!sublist.isEmpty() && (sublist.get(sublist.size() - 1).getUnicode().equals(" ") || sublist.get(sublist.size() - 1).getUnicode().equals("\u00A0"))) {
+        if (!sublist.isEmpty() && (sublist.get(sublist.size() - 1)
+                .getUnicode()
+                .equals(" ") || sublist.get(sublist.size() - 1).getUnicode().equals("\u00A0"))) {
            sublist = sublist.subList(0, sublist.size() - 1);
        }
-        if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+        if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0)
+                .getUnicode()
+                .equals("\u00A0")))) {
            textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
        }
        super.writeString(text);
    }

+
    @Override
    public String getText(PDDocument doc) throws IOException {

-        maxCharWidths = 0;
-        maxCharWidths = 0;
+        minCharWidth = Integer.MAX_VALUE;
+        maxCharWidth = 0;
+        minCharHeight = Integer.MAX_VALUE;
+        maxCharHeight = 0;
        textPositionSequences.clear();
+        imageBounds = new ArrayList<>();
        rulings.clear();
        graphicsPath.clear();
        path_x = 0.0f;
@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.parsing.model;

+import java.awt.geom.Rectangle2D;
 import java.util.List;

 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Ruling;
@@ -13,10 +14,11 @@ public class ParsedElements {

    private List<TextPositionSequence> sequences;
    private List<Ruling> rulings;
+    private List<Rectangle2D> imageBounds;

    private boolean landscape;
    private boolean rotated;

+    private float minCharWidth;
    private float maxCharWidth;
-    private float maxCharHeight;
 }
@@ -32,6 +32,7 @@ public class SearchableText {

    }

+
    @SuppressWarnings("checkstyle:ModifiedControlVariable")
    public List<EntityPositionSequence> getSequences(String searchString, boolean caseInsensitive,
                                                     List<TextPositionSequence> sequencesSubList) {
@@ -66,9 +67,12 @@ public class SearchableText {
            for (int j = 0; j < searchSpace.get(i).length(); j++) {

                if (i > 0 && j == 0 && searchSpace.get(i).charAt(0, caseInsensitive) == ' ' && searchSpace.get(i - 1)
-                        .charAt(searchSpace.get(i - 1).length() - 1, caseInsensitive) == ' ' || j > 0 && searchSpace.get(i)
-                        .charAt(j, caseInsensitive) == ' ' && searchSpace.get(i).charAt(j - 1, caseInsensitive) == ' ') {
-                    if (j == searchSpace.get(i).length() - 1 && counter != 0 && !partMatch.getTextPositions().isEmpty()) {
+                        .charAt(searchSpace.get(i - 1)
+                                .length() - 1, caseInsensitive) == ' ' || j > 0 && searchSpace.get(i)
+                        .charAt(j, caseInsensitive) == ' ' && searchSpace.get(i)
+                        .charAt(j - 1, caseInsensitive) == ' ') {
+                    if (j == searchSpace.get(i).length() - 1 && counter != 0 && !partMatch.getTextPositions()
+                            .isEmpty()) {
                        crossSequenceParts.add(partMatch);
                    }
                    continue;
@@ -80,8 +84,8 @@ public class SearchableText {
                    counter++;
                }

-                if (searchSpace.get(i)
-                        .charAt(j, caseInsensitive) == searchChars[counter] || counter != 0 && searchSpace.get(i)
+                if (searchSpace.get(i).charAt(j, caseInsensitive) == searchChars[counter] || counter != 0 && searchSpace
+                        .get(i)
                        .charAt(j, caseInsensitive) == '-') {

                    if (counter != 0 || i == 0 && j == 0 || j != 0 && isSeparator(searchSpace.get(i)
@@ -100,14 +104,15 @@ public class SearchableText {
                    if (counter == searchString.length()) {
                        crossSequenceParts.add(partMatch);

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

                        counter = 0;
@@ -130,15 +135,21 @@ public class SearchableText {
        }

        return finalMatches;
-
    }


-    private List<EntityPositionSequence> buildEntityPositionSequence(List<TextPositionSequence> crossSequenceParts) {
+    private List<EntityPositionSequence> buildEntityPositionSequence(List<TextPositionSequence> crossSequenceParts,
+                                                                     String searchString) {
+
+        List<EntityPositionSequence> result = new ArrayList<>();
+        String asString = buildString(crossSequenceParts);
+        if (!asString.equalsIgnoreCase(searchString)) {
+            return result;
+        }

        String plainId = IdBuilder.buildId(crossSequenceParts);
        String id = plainId;
-        List<EntityPositionSequence> result = new ArrayList<>();
+
        int currentPage = -1;
        int idDiffentPageSuffix = 1;
        EntityPositionSequence entityPositionSequence = new EntityPositionSequence(id);
@@ -173,6 +184,12 @@ public class SearchableText {
    @Override
    public String toString() {

+        return buildString(sequences);
+    }
+
+
+    public String buildString(List<TextPositionSequence> sequences) {
+
        StringBuilder sb = new StringBuilder();

        TextPositionSequence previous = null;
@@ -62,9 +62,10 @@ public class Section {
    }


-    public boolean hasTableHeader(String headerName){
-       String cleanHeaderName = headerName.replaceAll("\n", "").replaceAll(" ", "").replaceAll("-", "");
-       return tabularData != null && tabularData.containsKey(cleanHeaderName);
+    public boolean hasTableHeader(String headerName) {
+
+        String cleanHeaderName = headerName.replaceAll("\n", "").replaceAll(" ", "").replaceAll("-", "");
+        return tabularData != null && tabularData.containsKey(cleanHeaderName);
    }


@@ -80,6 +81,34 @@ public class Section {
    }


+    public void expandByRegEx(String type, String pattern, boolean patternCaseInsensitive, int group) {
+
+        Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
+
+        Set<Entity> expanded = new HashSet<>();
+        for (Entity entity : entities) {
+
+            if (!entity.getType().equals(type) || entity.getTextAfter() == null) {
+                continue;
+            }
+
+            Matcher matcher = compiledPattern.matcher(entity.getTextAfter());
+
+            while (matcher.find()) {
+                String match = matcher.group(group);
+                if (StringUtils.isNotBlank(match)) {
+                    expanded.addAll(findEntities(entity.getWord() + match, type, false, entity.isRedaction(), entity.getMatchedRule(), entity
+                            .getRedactionReason(), entity.getLegalBasis()));
+
+                }
+            }
+        }
+
+        EntitySearchUtils.addEntitiesWithHigherRank(entities, expanded, dictionary);
+        EntitySearchUtils.removeEntitiesContainedInLarger(entities);
+    }
+
+
    public void redact(String type, int ruleNumber, String reason, String legalBasis) {

        boolean hasRecommendationDictionary = dictionaryTypes.contains(RECOMMENDATION_PREFIX + type);
@@ -127,7 +156,7 @@ public class Section {
    public void addHintAnnotation(String value, String asType) {

        Set<Entity> found = findEntities(value.trim(), asType, true, false, 0, null, null);
-        EntitySearchUtils.addEntitiesWithHigherRank(entities, found, dictionary);
+        EntitySearchUtils.addEntitiesIgnoreRank(entities, found);
    }


@@ -160,7 +189,7 @@ public class Section {

                String trimmedValue = value.trim();
                String cleanValue;
-                if(trimmedValue.startsWith(":")){
+                if (trimmedValue.startsWith(":")) {
                    cleanValue = trimmedValue.substring(1).trim();
                } else {
                    cleanValue = trimmedValue;
@@ -207,8 +236,8 @@ public class Section {
    }


-    public void redactAndRecommendByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType, int ruleNumber,
-                              String reason, String legalBasis) {
+    public void redactAndRecommendByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType,
+                                          int ruleNumber, String reason, String legalBasis) {

        Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);

@@ -363,6 +392,7 @@ public class Section {
            }
        }
    }
+
 }


@@ -30,7 +30,9 @@ import com.iqser.red.service.redaction.v1.model.SectionGrid;
 import com.iqser.red.service.redaction.v1.model.SectionRectangle;

 import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;

+@Slf4j
@Service
@RequiredArgsConstructor
 public class AnnotationService {
@@ -39,7 +41,7 @@ public class AnnotationService {


    public void annotate(PDDocument document, RedactionLog redactionLog, SectionGrid sectionGrid) throws IOException {
-
+        log.info("Annotating document.");
        Map<Integer, List<RedactionLogEntry>> redactionLogPerPage = convertRedactionLog(redactionLog);

        for (int page = 1; page <= document.getNumberOfPages(); page++) {
@@ -53,25 +55,25 @@ public class AnnotationService {

            List<RedactionLogEntry> logEntries = redactionLogPerPage.get(page);
            if (logEntries != null && !logEntries.isEmpty()) {
-                addAnnotations(logEntries, pdPage, page, redactionLog.getRuleSetId());
+                addAnnotations(logEntries, pdPage, page);
            }
        }
+        log.info("Finished document annotation.");
    }


-    private void addAnnotations(List<RedactionLogEntry> logEntries, PDPage pdPage, int page,
-                                String ruleSetId) throws IOException {
+    private void addAnnotations(List<RedactionLogEntry> logEntries, PDPage pdPage, int page) throws IOException {

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

        for (RedactionLogEntry entry : logEntries) {
-            annotations.addAll(createAnnotation(entry, page, ruleSetId, pdPage.getMediaBox(), pdPage.getCropBox()));
+            annotations.addAll(createAnnotation(entry, page, pdPage.getMediaBox(), pdPage.getCropBox()));
        }
    }


-    private List<PDAnnotation> createAnnotation(RedactionLogEntry redactionLogEntry, int page, String ruleSetId,
-                                                PDRectangle mediaBox, PDRectangle cropBox) {
+    private List<PDAnnotation> createAnnotation(RedactionLogEntry redactionLogEntry, int page, PDRectangle mediaBox,
+                                                PDRectangle cropBox) {

        List<PDAnnotation> annotations = new ArrayList<>();

@@ -89,7 +91,7 @@ public class AnnotationService {
        PDRectangle pdRectangle = toPDRectangle(rectangles, mediaBox, cropBox);
        annotation.setRectangle(pdRectangle);
        annotation.setQuadPoints(toQuadPoints(rectangles, mediaBox, cropBox));
-        if (!dictionaryService.isHint(redactionLogEntry.getType(), ruleSetId)) {
+        if (!redactionLogEntry.isHint()) {
            annotation.setContents(createAnnotationContent(redactionLogEntry));
        }
        annotation.setTitlePopup(redactionLogEntry.getId());
@@ -19,6 +19,8 @@ import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
 import com.iqser.red.service.redaction.v1.model.ManualRedactions;
 import com.iqser.red.service.redaction.v1.model.Rectangle;
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
+import com.iqser.red.service.redaction.v1.server.classification.model.Footer;
+import com.iqser.red.service.redaction.v1.server.classification.model.Header;
 import com.iqser.red.service.redaction.v1.server.classification.model.Paragraph;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
 import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
@@ -48,6 +50,8 @@ public class EntityRedactionService {

    public void processDocument(Document classifiedDoc, String ruleSetId, ManualRedactions manualRedactions) {

+        log.info("Processing document.");
+
        dictionaryService.updateDictionary(ruleSetId);
        KieContainer container = droolsExecutionService.updateRules(ruleSetId);
        long rulesVersion = droolsExecutionService.getRulesVersion(ruleSetId);
@@ -91,6 +95,9 @@ public class EntityRedactionService {

        classifiedDoc.setDictionaryVersion(dictionary.getVersion());
        classifiedDoc.setRulesVersion(rulesVersion);
+
+        log.info("Finished document processing.");
+
    }


@@ -113,7 +120,18 @@ public class EntityRedactionService {
                }
                sectionNumber.incrementAndGet();
            }
-            sectionSearchableTextPairs.add(processText(paragraph, manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+            sectionSearchableTextPairs.add(processText(paragraph.getSearchableText(), paragraph.getTextBlocks(), paragraph
+                    .getHeadline(), manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+            sectionNumber.incrementAndGet();
+        }
+
+        for (Header header : classifiedDoc.getHeaders()) {
+            sectionSearchableTextPairs.add(processText(header.getSearchableText(), header.getTextBlocks(), "Header", manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
+            sectionNumber.incrementAndGet();
+        }
+
+        for (Footer footer : classifiedDoc.getFooters()) {
+            sectionSearchableTextPairs.add(processText(footer.getSearchableText(), footer.getTextBlocks(), "Footer", manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
            sectionNumber.incrementAndGet();
        }

@@ -157,7 +175,6 @@ public class EntityRedactionService {

        List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();

-        int rowNumber = 0;
        for (List<Cell> row : table.getRows()) {
            SearchableText searchableRow = new SearchableText();
            Map<String, CellValue> tabularData = new HashMap<>();
@@ -170,11 +187,14 @@ public class EntityRedactionService {
                addSectionToManualRedactions(cell.getTextBlocks(), manualRedactions, table.getHeadline(), sectionNumber.intValue());
                int cellStart = start;

-                if(rowNumber != 0) {
+                if (!cell.isHeaderCell()) {
                    cell.getHeaderCells().forEach(headerCell -> {
                        StringBuilder headerBuilder = new StringBuilder();
                        headerCell.getTextBlocks().forEach(textBlock -> headerBuilder.append(textBlock.getText()));
-                        String headerName = headerBuilder.toString().replaceAll("\n", "").replaceAll(" ", "").replaceAll("-", "");
+                        String headerName = headerBuilder.toString()
+                                .replaceAll("\n", "")
+                                .replaceAll(" ", "")
+                                .replaceAll("-", "");
                        tabularData.put(headerName, new CellValue(cell.getTextBlocks(), cellStart));
                    });
                }
@@ -205,8 +225,8 @@ public class EntityRedactionService {
                    .build(), searchableRow));

            sectionNumber.incrementAndGet();
-            rowNumber++;
        }
+
        return sectionSearchableTextPairs;
    }

@@ -251,14 +271,13 @@ public class EntityRedactionService {
    }


-    private SectionSearchableTextPair processText(Paragraph paragraph, ManualRedactions manualRedactions,
+    private SectionSearchableTextPair processText(SearchableText searchableText, List<TextBlock> paragraphTextBlocks,
+                                                  String headline, ManualRedactions manualRedactions,
                                                  AtomicInteger sectionNumber, Dictionary dictionary, boolean local,
                                                  Map<Integer, Set<Entity>> hintsPerSectionNumber) {

-        SearchableText searchableText = paragraph.getSearchableText();
-        addSectionToManualRedactions(paragraph.getTextBlocks(), manualRedactions, paragraph.getHeadline(), sectionNumber
-                .intValue());
-        Set<Entity> entities = findEntities(searchableText, paragraph.getHeadline(), sectionNumber.intValue(), dictionary, local);
+        addSectionToManualRedactions(paragraphTextBlocks, manualRedactions, headline, sectionNumber.intValue());
+        Set<Entity> entities = findEntities(searchableText, headline, sectionNumber.intValue(), dictionary, local);
        surroundingWordsService.addSurroundingText(entities, searchableText, dictionary);

        return new SectionSearchableTextPair(Section.builder()
@@ -269,7 +288,7 @@ public class EntityRedactionService {
                        .collect(Collectors.toSet()) : entities)
                .text(searchableText.getAsStringWithLinebreaks())
                .searchText(searchableText.toString())
-                .headline(paragraph.getHeadline())
+                .headline(headline)
                .sectionNumber(sectionNumber.intValue())
                .searchableText(searchableText)
                .dictionary(dictionary)
@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;

+import java.awt.geom.Rectangle2D;
 import java.util.ArrayList;
 import java.util.HashSet;
 import java.util.List;
@@ -27,22 +28,29 @@ import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
 import com.iqser.red.service.redaction.v1.server.redaction.model.EntityPositionSequence;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.IdBuilder;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;

 import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;

+@Slf4j
@Service
@RequiredArgsConstructor
 public class RedactionLogCreatorService {

+    private static final String IMAGE = "image";
+
    private final DictionaryService dictionaryService;


    public void createRedactionLog(Document classifiedDoc, int numberOfPages, ManualRedactions manualRedactions,
                                   String ruleSetId) {

+        log.info("Creating redaction log.");
+
        Set<Integer> manualRedactionPages = getManualRedactionPages(manualRedactions);

        for (int page = 1; page <= numberOfPages; page++) {
@@ -56,6 +64,32 @@ public class RedactionLogCreatorService {
            if (manualRedactionPages.contains(page)) {
                addManualEntries(classifiedDoc, manualRedactions, page, ruleSetId);
            }
+
+            if (!classifiedDoc.getPages().get(page - 1).getImageBounds().isEmpty()) {
+                addImageEntries(classifiedDoc, page, ruleSetId);
+            }
+        }
+
+        log.info("Finished redaction log creation.");
+    }
+
+
+    private void addImageEntries(Document classifiedDoc, int pageNumber, String ruleSetId) {
+
+        for (Rectangle2D imageBounds : classifiedDoc.getPages().get(pageNumber - 1).getImageBounds()) {
+            RedactionLogEntry redactionLogEntry = RedactionLogEntry.builder()
+                    .id(IdBuilder.buildId(imageBounds, pageNumber))
+                    .color(getColor(IMAGE, ruleSetId))
+                    .type(IMAGE)
+                    .redacted(false)
+                    .isHint(true)
+                    .manual(false)
+                    .isDictionaryEntry(false)
+                    .isRecommendation(false)
+                    .positions(List.of(new Rectangle(new Point((float) imageBounds.getX(), (float) imageBounds.getY()), (float) imageBounds
+                            .getWidth(), (float) imageBounds.getHeight(), pageNumber)))
+                    .build();
+            classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
        }
    }

@@ -166,7 +200,7 @@ public class RedactionLogCreatorService {
                    startIndex = i;
                }
            }
-            if (startIndex != textPositions.size() - 1) {
+            if (startIndex != textPositions.size()) {
                rectangles.add(new TextPositionSequence(textPositions.subList(startIndex, textPositions.size()), page).getRectangle());
            }
        }
@@ -121,4 +121,10 @@ public class EntitySearchUtils {
        }
        entities.add(found);
    }
+
+    public void addEntitiesIgnoreRank(Set<Entity> entities, Set<Entity> found){
+        // HashSet keeps old value but we want the new.
+        entities.removeAll(found);
+        entities.addAll(found);
+    }
 }
@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.redaction.utils;

+import java.awt.geom.Rectangle2D;
 import java.nio.charset.StandardCharsets;
 import java.util.List;

@@ -23,4 +24,16 @@ public class IdBuilder {

        return hashFunction.hashString(sb.toString(), StandardCharsets.UTF_8).toString();
    }
+
+
+    public String buildId(Rectangle2D rectangle2D, int page){
+
+        StringBuilder sb = new StringBuilder();
+        sb.append("x").append(rectangle2D.getX()).append("y").append(rectangle2D.getY()).append("h").append(rectangle2D.getHeight()).append("w").append(rectangle2D.getWidth()).append("p").append(page);
+
+        return hashFunction.hashString(sb.toString(), StandardCharsets.UTF_8).toString();
+    }
+
+
+
 }
@@ -39,6 +39,8 @@ public class PdfSegmentationService {

    public Document parseDocument(PDDocument pdDocument) throws IOException {

+        log.info("Parsing document.");
+
        Document document = new Document();

        List<Page> pages = new ArrayList<>();
@@ -48,6 +50,7 @@ public class PdfSegmentationService {
            stripper.setPageNumber(pageNumber);
            stripper.setStartPage(pageNumber);
            stripper.setEndPage(pageNumber);
+            stripper.setPdpage(pdPage);
            stripper.getText(pdDocument);

            PDRectangle pdr = pdPage.getMediaBox();
@@ -56,18 +59,17 @@ public class PdfSegmentationService {
            int rotation = pdPage.getRotation();
            boolean isRotated = rotation != 0 && rotation != 360;

-
            ParsedElements parsedElements = ParsedElements.builder()
                    .rulings(stripper.getRulings())
                    .sequences(stripper.getTextPositionSequences())
-                    .maxCharWidth(stripper.getMaxCharWidths())
-                    .maxCharHeight(stripper.getMaxCharWidths())
+                    .imageBounds(stripper.getImageBounds())
+                    .minCharWidth(stripper.getMinCharWidth())
+                    .maxCharWidth(stripper.getMaxCharWidth())
                    .landscape(isLandscape)
                    .rotated(isRotated)
                    .build();

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

            Page page = blockificationService.blockify(parsedElements.getSequences(), cleanRulings.getHorizontal(), cleanRulings
                    .getVertical());
@@ -81,20 +83,22 @@ public class PdfSegmentationService {

            page.setPageNumber(pageNumber);
            increaseDocumentStatistics(page, document);
+            page.setImageBounds(parsedElements.getImageBounds());
            pages.add(page);
        }
+
        document.setPages(pages);

        classificationService.classifyDocument(document);

        sectionsBuilderService.buildSections(document);

+        log.info("Finished document parsing.");
+
        return document;
-
    }


-
    private void increaseDocumentStatistics(Page page, Document document) {

        if (!page.isLandscape()) {
@@ -10,6 +10,8 @@ import org.apache.commons.collections4.CollectionUtils;
 import org.springframework.stereotype.Service;

 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
+import com.iqser.red.service.redaction.v1.server.classification.model.Footer;
+import com.iqser.red.service.redaction.v1.server.classification.model.Header;
 import com.iqser.red.service.redaction.v1.server.classification.model.Page;
 import com.iqser.red.service.redaction.v1.server.classification.model.Paragraph;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
@@ -17,32 +19,51 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractT
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;

+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
@Service
 public class SectionsBuilderService {

    public void buildSections(Document document) {

+        log.info("Building sections.");
+
        List<AbstractTextContainer> chunkWords = new ArrayList<>();
        List<Paragraph> chunkBlockList = new ArrayList<>();
+        List<Header> headers = new ArrayList<>();
+        List<Footer> footers = new ArrayList<>();

        AbstractTextContainer prev = null;

        String lastHeadline = "";
        Table previousTable = null;
        for (Page page : document.getPages()) {
+            List<TextBlock> header = new ArrayList<>();
+            List<TextBlock> footer = new ArrayList<>();
            for (AbstractTextContainer current : page.getTextBlocks()) {

-                if (current.getClassification() == null || current.getClassification()
-                        .equals("Header") || current.getClassification().equals("Footer")) {
+                if (current.getClassification() == null) {
                    continue;
                }

                current.setPage(page.getPageNumber());

-                if (prev != null && current.getClassification().startsWith("H ") && !prev.getClassification().startsWith("H ") || !document.isHeadlines()) {
+                if (current.getClassification().equals("Header")) {
+                    header.add((TextBlock) current);
+                    continue;
+                }
+
+                if (current.getClassification().equals("Footer")) {
+                    footer.add((TextBlock) current);
+                    continue;
+                }
+
+                if (prev != null && current.getClassification().startsWith("H ") && !prev.getClassification()
+                        .startsWith("H ") || !document.isHeadlines()) {
                    Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
                    chunkBlock.setHeadline(lastHeadline);
-                    if(document.isHeadlines()) {
+                    if (document.isHeadlines()) {
                        lastHeadline = current.getText();
                    }
                    chunkBlockList.add(chunkBlock);
@@ -60,6 +81,8 @@ public class SectionsBuilderService {
                chunkWords.add(current);
                prev = current;
            }
+            headers.add(new Header(header));
+            footers.add(new Footer(footer));
        }

        Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
@@ -67,6 +90,11 @@ public class SectionsBuilderService {
        chunkBlockList.add(chunkBlock);

        document.setParagraphs(chunkBlockList);
+        document.setHeaders(headers);
+        document.setFooters(footers);
+
+        log.info("Finished section building.");
+
    }


@@ -175,9 +203,9 @@ public class SectionsBuilderService {

    private boolean hasInvalidHeaderInformation(Table table) {

-        return table.getRows().stream()
-                .flatMap(row -> row.stream()
-                        .filter(cell -> CollectionUtils.isNotEmpty(cell.getHeaderCells())))
+        return table.getRows()
+                .stream()
+                .flatMap(row -> row.stream().filter(cell -> CollectionUtils.isNotEmpty(cell.getHeaderCells())))
                .findAny()
                .isEmpty();

@@ -188,6 +216,9 @@ public class SectionsBuilderService {

        for (int i = table.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
            List<Cell> row = table.getRows().get(i);
+            if (row.size() == 1) {
+                continue;
+            }
            boolean allNonHeader = true;
            for (Cell cell : row) {
                if (cell.isHeaderCell()) {
@@ -110,6 +110,10 @@ public class Table extends AbstractTextContainer {
        // we move from left to right and top to bottom
        for (int rowIndex = 0; rowIndex < rows.size(); rowIndex++) {
            List<Cell> rowCells = rows.get(rowIndex);
+            if(rowCells.size() == 1){
+                continue;
+            }
+
            for (int colIndex = 0; colIndex < rowCells.size(); colIndex++) {
                Cell cell = rowCells.get(colIndex);
                List<Cell> cellsToTheLeft = rowCells.subList(0, colIndex);
@@ -15,12 +15,18 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.CleanRuli
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Ruling;
 import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;

+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
@Service
 public class RulingCleaningService {

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

        List<Ruling> vrs = new ArrayList<>();
@@ -39,13 +45,10 @@ public class RulingCleaningService {
        }
        List<Ruling> horizontalRulingLines = collapseOrientedRulings(hrs);

-        return CleanRulings
-                .builder()
-                .vertical(verticalRulingLines)
-                .horizontal(horizontalRulingLines)
-                .build();
+        return CleanRulings.builder().vertical(verticalRulingLines).horizontal(horizontalRulingLines).build();
    }

+
    public void snapPoints(List<? extends Line2D.Float> rulings, float xThreshold, float yThreshold) {

        // collect points and keep a Line -> p1,p2 map
@@ -122,12 +125,14 @@ public class RulingCleaningService {


    private List<Ruling> collapseOrientedRulings(List<Ruling> lines) {
+
        int COLINEAR_OR_PARALLEL_PIXEL_EXPAND_AMOUNT = 1;
        return collapseOrientedRulings(lines, COLINEAR_OR_PARALLEL_PIXEL_EXPAND_AMOUNT);
    }


    private List<Ruling> collapseOrientedRulings(List<Ruling> lines, int expandAmount) {
+
        ArrayList<Ruling> rv = new ArrayList<>();
        lines.sort((a, b) -> {
            final float diff = a.getPosition() - b.getPosition();
@@ -141,25 +146,24 @@ public class RulingCleaningService {
                final float lastStart = last.getStart();
                final float lastEnd = last.getEnd();

-                final boolean lastFlipped = lastStart            > lastEnd;
+                final boolean lastFlipped = lastStart > lastEnd;
                final boolean nextFlipped = next_line.getStart() > next_line.getEnd();

                boolean differentDirections = nextFlipped != lastFlipped;
-                float nextS = differentDirections ? next_line.getEnd()   : next_line.getStart();
+                float nextS = differentDirections ? next_line.getEnd() : next_line.getStart();
                float nextE = differentDirections ? next_line.getStart() : next_line.getEnd();

                final float newStart = lastFlipped ? Math.max(nextS, lastStart) : Math.min(nextS, lastStart);
-                final float newEnd   = lastFlipped ? Math.min(nextE, lastEnd)   : Math.max(nextE, lastEnd);
+                final float newEnd = lastFlipped ? Math.min(nextE, lastEnd) : Math.max(nextE, lastEnd);
                last.setStartEnd(newStart, newEnd);
                assert !last.oblique();
-            }
-            else if (next_line.length() == 0) {
+            } else if (next_line.length() == 0) {
                continue;
-            }
-            else {
+            } else {
                rv.add(next_line);
            }
        }
        return rv;
    }
+
 }
@@ -23,11 +23,16 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.Ruling;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
 import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;

+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
@Service
 public class TableExtractionService {

    public void extractTables(CleanRulings cleanRulings, Page page) {

+        log.info("Extracting tables.");
+
        List<Cell> cells = findCells(cleanRulings.getHorizontal(), cleanRulings.getVertical());

        List<TextBlock> toBeRemoved = new ArrayList<>();
@@ -78,6 +83,9 @@ public class TableExtractionService {
        }

        page.getTextBlocks().removeAll(toBeRemoved);
+
+        log.info("Finished table extraction.");
+
    }


@@ -81,6 +81,7 @@ public class RedactionIntegrationTest {
    private static final String PUBLISHED_INFORMATION = "published_information";
    private static final String TEST_METHOD = "test_method";
    private static final String PURITY = "purity";
+    private static final String IMAGE = "image";

    private static final String RECOMMENDATION_AUTHOR = "recommendation_CBI_author";
    private static final String RECOMMENDATION_ADDRESS = "recommendation_CBI_address";
@@ -157,6 +158,7 @@ public class RedactionIntegrationTest {
        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_ADDRESS, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(RECOMMENDATION_ADDRESS));
        when(dictionaryClient.getDictionaryForType(FALSE_POSITIVE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(FALSE_POSITIVE));
        when(dictionaryClient.getDictionaryForType(PURITY, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(PURITY));
+        when(dictionaryClient.getDictionaryForType(IMAGE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(IMAGE));
        when(dictionaryClient.getColors(TEST_RULESET_ID)).thenReturn(colors);
    }

@@ -238,6 +240,11 @@ public class RedactionIntegrationTest {
                        .stream()
                        .map(this::cleanDictionaryEntry)
                        .collect(Collectors.toSet()));
+        dictionary.computeIfAbsent(IMAGE, v -> new ArrayList<>())
+                .addAll(ResourceLoader.load("dictionaries/image.txt")
+                        .stream()
+                        .map(this::cleanDictionaryEntry)
+                        .collect(Collectors.toSet()));
    }


@@ -264,6 +271,7 @@ public class RedactionIntegrationTest {
        typeColorMap.put(RECOMMENDATION_ADDRESS, "#8df06c");
        typeColorMap.put(FALSE_POSITIVE, "#ffffff");
        typeColorMap.put(PURITY, "#ffe187");
+        typeColorMap.put(IMAGE, "#fcc5fb");

        hintTypeMap.put(VERTEBRATE, true);
        hintTypeMap.put(ADDRESS, false);
@@ -280,6 +288,7 @@ public class RedactionIntegrationTest {
        hintTypeMap.put(RECOMMENDATION_ADDRESS, false);
        hintTypeMap.put(FALSE_POSITIVE, true);
        hintTypeMap.put(PURITY, false);
+        hintTypeMap.put(IMAGE, true);

        caseInSensitiveMap.put(VERTEBRATE, true);
        caseInSensitiveMap.put(ADDRESS, false);
@@ -296,6 +305,7 @@ public class RedactionIntegrationTest {
        caseInSensitiveMap.put(RECOMMENDATION_ADDRESS, false);
        caseInSensitiveMap.put(FALSE_POSITIVE, false);
        caseInSensitiveMap.put(PURITY, false);
+        caseInSensitiveMap.put(IMAGE, true);

        recommendationTypeMap.put(VERTEBRATE, false);
        recommendationTypeMap.put(ADDRESS, false);
@@ -312,6 +322,8 @@ public class RedactionIntegrationTest {
        recommendationTypeMap.put(RECOMMENDATION_ADDRESS, true);
        recommendationTypeMap.put(FALSE_POSITIVE, false);
        recommendationTypeMap.put(PURITY, false);
+        recommendationTypeMap.put(IMAGE, false);
+

        rankTypeMap.put(FALSE_POSITIVE, 160);
        rankTypeMap.put(PURITY, 155);
@@ -328,6 +340,8 @@ public class RedactionIntegrationTest {
        rankTypeMap.put(HINT_ONLY, 50);
        rankTypeMap.put(RECOMMENDATION_AUTHOR, 40);
        rankTypeMap.put(RECOMMENDATION_ADDRESS, 30);
+        rankTypeMap.put(IMAGE, 30);
+

        colors.setDefaultColor("#acfc00");
        colors.setNotRedacted("#cccccc");
@@ -404,6 +418,7 @@ public class RedactionIntegrationTest {
    }


+
    private List<File> getPathsRecursively(File path) {

        List<File> result = new ArrayList<>();
@@ -425,9 +440,16 @@ public class RedactionIntegrationTest {
    @Test
    public void redactionTest() throws IOException {

+//        49 Cyprodinil - EU AIR3 - MCA Section 8 Supplement - Ecotoxicological studies on the active substance.pdf
+//        182 Fludioxonil - EU AIR3 - MCA Section 8 Supplement - Ecotoxicological studies on the active substance.pdf
+//        38 A14325E - EU AIR3 - MCP Section 10 - Ecotoxicological studies on the plant protection product.pdf
+//        91 Trinexapac-ethyl_RAR_01_Volume_1_2018-02-23.pdf
+//        95 Trinexapac-ethyl_RAR_08_Volume_3CA_B-6_2018-01-10.pdf
+
+
        System.out.println("redactionTest");
        long start = System.currentTimeMillis();
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Cyprodinil/49 Cyprodinil - EU AIR3 - MCA Section 8 Supplement - Ecotoxicological studies on the active substance.pdf");

        AnalyzeRequest request = AnalyzeRequest.builder()
                .ruleSetId(TEST_RULESET_ID)
@@ -576,7 +598,7 @@ public class RedactionIntegrationTest {
    public void htmlTablesTest() throws IOException {

        System.out.println("htmlTablesTest");
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Fludioxonil/52 Fludioxonil_RAR_07_Volume_3CA_B-5_2018-02-21.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");

        RedactionRequest request = RedactionRequest.builder()
                .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))
@@ -1,15 +1,36 @@
 package com.iqser.red.service.redaction.v1.server.redaction.utils;

+import org.apache.commons.lang3.StringUtils;
 import org.junit.Test;

 import java.util.ArrayList;
 import java.util.List;
+import java.util.Set;
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;

+import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
+
 public class RegExPatternTest {


+    @Test
+    public void testExpand(){
+        String pattern = "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)";
+        String text = ", G.R., Simoneaux,";
+        Pattern compiledPattern = Pattern.compile(pattern, 0);
+        Matcher matcher = compiledPattern.matcher(text);
+
+        while (matcher.find()) {
+            String match = matcher.group(1);
+            if (StringUtils.isNotBlank(match)) {
+                System.out.println(match);
+            }
+        }
+    }
+
+
+
    @Test
    public void testEmailRegEx(){
        String text = "Address: Schwarzwaldalle " +
@@ -7,6 +7,15 @@ global Section section

 // --------------------------------------- CBI rules -------------------------------------------------------------------

+rule "0: Expand CBI Authors with firstname initials"
+    when
+        Section(matchesType("CBI_author") || matchesType("recommendation_CBI_author"))
+    then
+        section.expandByRegEx("CBI_author", "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)", false, 1);
+        section.expandByRegEx("recommendation_CBI_author", "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)", false, 1);
+    end
+
+
 rule "1: Redact CBI Authors"
    when
        Section(matchesType("CBI_author") || matchesType("recommendation_CBI_author"))
@@ -7,6 +7,15 @@ global Section section

 // --------------------------------------- CBI rules -------------------------------------------------------------------

+rule "0: Expand CBI Authors with firstname initials"
+    when
+        Section(matchesType("CBI_author") || matchesType("recommendation_CBI_author"))
+    then
+        section.expandByRegEx("CBI_author", "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)", false, 1);
+        section.expandByRegEx("recommendation_CBI_author", "(,? [A-Z]\\.?( ?[A-Z]\\.?)?( ?[A-Z]\\.?)?\\b\\.?)", false, 1);
+    end
+
+
 rule "1: Redacted because Section contains Vertebrate"
    when
        Section(matchesType("vertebrate"))
Author	SHA1	Message	Date
lmaldacker	9ceba4abf3	Adjust imports	2021-02-09 10:36:58 +01:00
lmaldacker	ebca1fc750	Adjust pom	2021-02-09 10:21:34 +01:00
lmaldacker	37125f14fd	Add logs	2021-02-08 16:22:26 +01:00
Dominique Eiflaender	6d9ed080ce	Pull request #120 : RED-1039: Fixed finding textpositions, RED-1042: Fixed get rectangles per line Merge in RED/redaction-service from RED-1039 to master * commit '00b0cb160342f1857ac0e523f994918057d5fc6b': RED-1039: Fixed finding textpositions, RED-1042: Fixed get rectangles per line	2021-02-08 14:50:02 +01:00
Dominique Eifländer	00b0cb1603	RED-1039: Fixed finding textpositions, RED-1042: Fixed get rectangles per line	2021-02-08 14:09:11 +01:00
Dominique Eiflaender	5ecf21290c	Pull request #119 : RED-1046: Ignore dictionary rank for words that are explicitly set in the rules Merge in RED/redaction-service from RED-1046 to master * commit '8965e7654867ff666c54c841319376c5899e7326': RED-1046: Ignore dictionary rank for words that are explicitly set in the rules	2021-02-05 15:08:15 +01:00
Dominique Eifländer	8965e76548	RED-1046: Ignore dictionary rank for words that are explicitly set in the rules	2021-02-05 14:55:16 +01:00
Dominique Eiflaender	ba8c4dc240	Pull request #118 : RED-1045: Enabled to redact in headers and footers Merge in RED/redaction-service from RED-1045 to master * commit '577db37b11dfd0243d22986a6d224700f0b49693': RED-1045: Enabled to redact in headers and footers	2021-02-05 13:16:55 +01:00
Dominique Eifländer	577db37b11	RED-1045: Enabled to redact in headers and footers	2021-02-05 13:05:12 +01:00
Dominique Eiflaender	154e09b843	Pull request #117 : Fixed several table extraction problems Merge in RED/redaction-service from tableExtractionImprovement to master * commit 'a101b98a400b2a35635f2c5e0e894ac849288c3b': Fixed several table extraction problems	2021-02-04 15:21:12 +01:00
Dominique Eifländer	a101b98a40	Fixed several table extraction problems	2021-02-04 15:08:48 +01:00
Dominique Eiflaender	76ecfdabd3	Pull request #116 : Fixed table extraction problems Merge in RED/redaction-service from TableExtractionProbemFix to master * commit 'fc2ac03691871a8a1f3c8a29756e3a60e63e16bf': Fixed table extraction problems	2021-02-03 14:49:01 +01:00
Dominique Eifländer	fc2ac03691	Fixed table extraction problems	2021-02-03 14:34:29 +01:00
Dominique Eiflaender	7898f6a30f	Pull request #115 : Expand CBI Authors with firstname initials Merge in RED/redaction-service from ExpandAuthors to master * commit '35f3582d0850053c1dc38ef0b149daa47bd9cc56': Expand CBI Authors with firstname initials	2021-02-02 12:35:07 +01:00
Dominique Eifländer	35f3582d08	Expand CBI Authors with firstname initials	2021-02-02 12:20:40 +01:00
Dominique Eiflaender	b4699caaae	Pull request #114 : Fixed position problem for rotated images Merge in RED/redaction-service from rotatedImages to master * commit 'd8e444280b5a025e19785c70c9fcbb11480b5042': Fixed position problem for rotated images	2021-02-01 16:17:19 +01:00
Dominique Eifländer	d8e444280b	Fixed position problem for rotated images	2021-02-01 16:04:45 +01:00
Dominique Eiflaender	ec4db538a7	Pull request #113 : Ignore too small images Merge in RED/redaction-service from IgnoreTooSmall to master * commit '39ca191b9cb5dec5fbb01e8319cbbe361c7cd85e': Ignore too small images	2021-02-01 15:46:57 +01:00
Dominique Eifländer	39ca191b9c	Ignore too small images	2021-02-01 15:14:46 +01:00
Dominique Eiflaender	0cec88f1b4	Pull request #112 : Annotate images Merge in RED/redaction-service from annotateImages to master * commit 'acddfafa5b60d0120a48bdf47fe218bf59e359d2': Annotate images	2021-02-01 14:12:28 +01:00
Dominique Eifländer	acddfafa5b	Annotate images	2021-02-01 13:52:18 +01:00
Dominique Eiflaender	6b68890fcf	Pull request #111 : RED-1019: Fixed returning numberOfPages Merge in RED/redaction-service from RED-1019 to master * commit '1ed99412595b560d92e542d949e52d19d58beb2c': RED-1019: Fixed returning numberOfPages	2021-01-29 13:32:05 +01:00
Dominique Eifländer	1ed9941259	RED-1019: Fixed returning numberOfPages	2021-01-29 13:17:25 +01:00
Timo Bejan	1e7d535a33	Pull request #110 : fixed issue affecting service on restarts Merge in RED/redaction-service from bugfix-annotate to master * commit '1091d6a886243387f23da32d0c678b66b0c21f26': fixed issue affecting service on restarts	2021-01-28 10:17:02 +01:00