Pull request #121 : Add hint annotation by reg-ex and expand to hint annotation by reg-ex

Merge in RED/redaction-service from dismissGen to master * commit '0f263c69b8f2eb7201f062cc263c8d06b7a710a1': Add hint annotation by reg-ex and expand to hint annotation by reg-ex
Add hint annotation by reg-ex and expand to hint annotation by reg-ex
2021-02-11 17:22:55 +01:00 · 2021-02-11 16:53:07 +01:00 · 2021-02-11 14:11:22 +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
19 changed files with 247 additions and 68 deletions
@@ -72,14 +72,6 @@ public class PlanSpec {
                        .jobs(new Job("Default Job",
                                new BambooKey("JOB1"))
                                .tasks(
-                                        new ScriptTask()
-                                                .description("Clean")
-                                                .inlineBody("#!/bin/bash\n" +
-                                                        "set -e\n" +
-                                                        "rm -rf ./*"),
-                                        new VcsCheckoutTask()
-                                                .description("Checkout Default Repository")
-                                                .checkoutItems(new CheckoutItem().defaultRepository()),
                                        new ScriptTask()
                                                .description("Clean")
                                                .inlineBody("#!/bin/bash\n" +
@@ -5,7 +5,7 @@
    <parent>
        <artifactId>platform-dependency</artifactId>
        <groupId>com.iqser.red</groupId>
-        <version>1.0.2</version>
+        <version>1.0.3</version>
    </parent>
    <modelVersion>4.0.0</modelVersion>

@@ -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;
+    }
+
+}
@@ -54,7 +54,16 @@ public class PDFLinesTextStripper extends PDFTextStripper {
    protected PDPage pdpage;

    @Getter
-    private int minCharWidths;
+    private int minCharWidth;
+
+    @Getter
+    private int maxCharWidth;
+    
+    @Getter
+    private int minCharHeight;
+
+    @Getter
+    private int maxCharHeight;

    @Getter
    private final List<TextPositionSequence> textPositionSequences = new ArrayList<>();
@@ -280,8 +289,19 @@ public class PDFLinesTextStripper extends PDFTextStripper {
        for (int i = 0; i <= textPositions.size() - 1; i++) {

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

            if (i == 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
@@ -333,7 +353,10 @@ public class PDFLinesTextStripper extends PDFTextStripper {
    @Override
    public String getText(PDDocument doc) throws IOException {

-        minCharWidths = Integer.MAX_VALUE;
+        minCharWidth = Integer.MAX_VALUE;
+        maxCharWidth = 0;
+        minCharHeight = Integer.MAX_VALUE;
+        maxCharHeight = 0;
        textPositionSequences.clear();
        imageBounds = new ArrayList<>();
        rulings.clear();
@@ -20,4 +20,5 @@ public class ParsedElements {
    private boolean rotated;

    private float minCharWidth;
+    private float maxCharWidth;
 }
@@ -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;
@@ -140,6 +140,48 @@ public class Section {
    }


+    public void expandToHintAnnotationByRegEx(String type, String pattern, boolean patternCaseInsensitive, int group, String asType) {
+
+        Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
+
+        Set<Entity> expanded = new HashSet<>();
+        for (Entity entity : entities) {
+
+            if (!entity.getType().equals(type) || entity.getTextAfter() == null) {
+                continue;
+            }
+
+            Matcher matcher = compiledPattern.matcher(entity.getTextAfter());
+
+            while (matcher.find()) {
+                String match = matcher.group(group);
+                if (StringUtils.isNotBlank(match)) {
+                    expanded.addAll(findEntities(entity.getWord() + match, asType, false, false, 0, null, null));
+                }
+            }
+        }
+
+        EntitySearchUtils.addEntitiesWithHigherRank(entities, expanded, dictionary);
+        EntitySearchUtils.removeEntitiesContainedInLarger(entities);
+    }
+
+
+    public void addHintAnnotationByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType) {
+
+        Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
+
+        Matcher matcher = compiledPattern.matcher(searchText);
+
+        while (matcher.find()) {
+            String match = matcher.group(group);
+            if (StringUtils.isNotBlank(match)) {
+                Set<Entity> found = findEntities(match.trim(), asType, false, false, 0, null, null);
+                EntitySearchUtils.addEntitiesWithHigherRank(entities, found, dictionary);
+            }
+        }
+    }
+
+
    public void redactIfPrecededBy(String prefix, String type, int ruleNumber, String reason, String legalBasis) {

        entities.forEach(entity -> {
@@ -156,7 +198,7 @@ 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);
    }


@@ -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;
@@ -113,7 +115,18 @@ 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 +170,6 @@ 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 +182,14 @@ 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 +220,8 @@ public class EntityRedactionService {
                    .build(), searchableRow));

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

@@ -251,14 +266,13 @@ 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 +283,7 @@ 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)
@@ -194,7 +194,7 @@ 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);
+    }
 }
@@ -61,13 +61,13 @@ public class PdfSegmentationService {
                    .rulings(stripper.getRulings())
                    .sequences(stripper.getTextPositionSequences())
                    .imageBounds(stripper.getImageBounds())
-                    .minCharWidth(stripper.getMinCharWidths())
+                    .minCharWidth(stripper.getMinCharWidth())
+                    .maxCharWidth(stripper.getMaxCharWidth())
                    .landscape(isLandscape)
                    .rotated(isRotated)
                    .build();

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

            Page page = blockificationService.blockify(parsedElements.getSequences(), cleanRulings.getHorizontal(), cleanRulings
                    .getVertical());
@@ -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;
@@ -24,25 +26,39 @@ public class SectionsBuilderService {

        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 +76,8 @@ 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 +85,8 @@ public class SectionsBuilderService {
        chunkBlockList.add(chunkBlock);

        document.setParagraphs(chunkBlockList);
+        document.setHeaders(headers);
+        document.setFooters(footers);
    }


@@ -175,9 +195,9 @@ 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 +208,9 @@ 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);
@@ -18,9 +18,10 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;
@Service
 public class RulingCleaningService {

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

        List<Ruling> vrs = new ArrayList<>();
@@ -39,13 +40,10 @@ public class RulingCleaningService {
        }
        List<Ruling> horizontalRulingLines = collapseOrientedRulings(hrs);

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

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

        // collect points and keep a Line -> p1,p2 map
@@ -122,12 +120,14 @@ public class RulingCleaningService {


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


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

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

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

                final float newStart = lastFlipped ? Math.max(nextS, lastStart) : Math.min(nextS, lastStart);
-                final float newEnd   = lastFlipped ? Math.min(nextE, lastEnd)   : Math.max(nextE, lastEnd);
+                final float newEnd = lastFlipped ? Math.min(nextE, lastEnd) : Math.max(nextE, lastEnd);
                last.setStartEnd(newStart, newEnd);
                assert !last.oblique();
-            }
-            else if (next_line.length() == 0) {
+            } else if (next_line.length() == 0) {
                continue;
-            }
-            else {
+            } else {
                rv.add(next_line);
            }
        }
        return rv;
    }
+
 }
@@ -418,6 +418,7 @@ public class RedactionIntegrationTest {
    }


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

        List<File> result = new ArrayList<>();
@@ -439,9 +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_01_Volume_1_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)
@@ -590,7 +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()))
@@ -316,9 +316,9 @@ rule "26: Redact On behalf of Syngenta Ltd.:"

 // --------------------------------------- other rules -------------------------------------------------------------------

-rule "27: Redact Purity"
+rule "27: Purity Hint"
    when
-        Section(searchText.contains("purity"))
+        Section(searchText.toLowerCase().contains("purity"))
    then
-	    section.redactByRegEx("purity ?:? (([\\d\\.]+)( .{0,4}\\.)? ?%)", true, 1, "purity", 27, "Purity found", "Reg (EC) No 1107/2009 Art. 63 (2a)");
+	    section.addHintAnnotationByRegEx("(purity( of)?( ?:)?) [\\d\\.]+( .{0,4}\\.)? ?%", true, 1, "hint_only");
    end
@@ -10,7 +10,7 @@
        <Root level="warn">
            <AppenderRef ref="CONSOLE" />
        </Root>
-        <Logger name="com.iqser.gin4" level="info" />
+        <Logger name="com.iqser" level="info" />
    </Loggers>

 </Configuration>
Author	SHA1	Message	Date
Lena Maldacker	8989191314	Pull request #121 : Add hint annotation by reg-ex and expand to hint annotation by reg-ex Merge in RED/redaction-service from dismissGen to master * commit '0f263c69b8f2eb7201f062cc263c8d06b7a710a1': Add hint annotation by reg-ex and expand to hint annotation by reg-ex	2021-02-11 17:22:55 +01:00
lmaldacker	0f263c69b8	Add hint annotation by reg-ex and expand to hint annotation by reg-ex	2021-02-11 16:53:07 +01:00
Kresnadi Budisantoso	ea1c2898bc	Remove duplicate build tasks in bamboo specs	2021-02-11 14:11:22 +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