Fix pmd

Merge in RED/redaction-service from RED-1039 to master

* commit '00b0cb160342f1857ac0e523f994918057d5fc6b':
  RED-1039: Fixed finding textpositions, RED-1042: Fixed get rectangles per line

redaction-service-v1/redaction-service-server-v1/pom.xml

@@ -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>

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

@@ -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();

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

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

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

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

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

@@ -51,6 +51,9 @@ public class RedactionController implements RedactionResource {
     @Override
     public AnalyzeResult analyze(@RequestBody AnalyzeRequest analyzeRequest) {
 
+        log.info("Starting redaction analysis...");
+        long startTime = System.currentTimeMillis();
+
         try (PDDocument pdDocument = PDDocument.load(new ByteArrayInputStream(analyzeRequest.getDocument()))) {
             pdDocument.setAllSecurityToBeRemoved(true);
 
@@ -64,6 +67,8 @@ public class RedactionController implements RedactionResource {
 
             log.info("Redaction analysis successful...");
 
+            log.debug("Finishing redaction analysis took: {}.", System.currentTimeMillis() - startTime);
+
             return AnalyzeResult.builder()
                     .sectionGrid(classifiedDoc.getSectionGrid())
                     .redactionLog(new RedactionLog(classifiedDoc.getRedactionLogEntities(), classifiedDoc.getDictionaryVersion(), classifiedDoc

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

@@ -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();

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

@@ -20,4 +20,5 @@ public class ParsedElements {
     private boolean rotated;
 
     private float minCharWidth;
+    private float maxCharWidth;
 }

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

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

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

@@ -156,7 +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);
     }
 
 

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

@@ -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 {
@@ -40,6 +42,9 @@ public class AnnotationService {
 
     public void annotate(PDDocument document, RedactionLog redactionLog, SectionGrid sectionGrid) throws IOException {
 
+        log.info("Annotating document...");
+        long startTime = System.currentTimeMillis();
+
         Map<Integer, List<RedactionLogEntry>> redactionLogPerPage = convertRedactionLog(redactionLog);
 
         for (int page = 1; page <= document.getNumberOfPages(); page++) {
@@ -56,6 +61,7 @@ public class AnnotationService {
                 addAnnotations(logEntries, pdPage, page);
             }
         }
+        log.info("Finishing document annotation took: {}.", System.currentTimeMillis() - startTime);
     }
 
 

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

@@ -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;
@@ -48,14 +50,23 @@ public class EntityRedactionService {
 
     public void processDocument(Document classifiedDoc, String ruleSetId, ManualRedactions manualRedactions) {
 
+        log.info("Processing document...");
+        long startTime = System.currentTimeMillis();
+
         dictionaryService.updateDictionary(ruleSetId);
         KieContainer container = droolsExecutionService.updateRules(ruleSetId);
         long rulesVersion = droolsExecutionService.getRulesVersion(ruleSetId);
 
         Dictionary dictionary = dictionaryService.getDeepCopyDictionary(ruleSetId);
 
+        long current = System.currentTimeMillis() - startTime;
+        log.info("Adding image entries took: {}.", current);
+
         Set<Entity> documentEntities = new HashSet<>(findEntities(classifiedDoc, container, manualRedactions, dictionary, false, null));
 
+        current = System.currentTimeMillis() - startTime - current;
+        log.info("Finding entries took: {}.", current);
+
         if (dictionary.hasLocalEntries()) {
 
             Map<Integer, Set<Entity>> hintsPerSectionNumber = new HashMap<>();
@@ -69,6 +80,9 @@ public class EntityRedactionService {
             Set<Entity> foundByLocal = findEntities(classifiedDoc, container, manualRedactions, dictionary, true, hintsPerSectionNumber);
             EntitySearchUtils.addEntitiesWithHigherRank(documentEntities, foundByLocal, dictionary);
             EntitySearchUtils.removeEntitiesContainedInLarger(documentEntities);
+
+            current = System.currentTimeMillis() - startTime - current;
+            log.info("Adding local dictionary annotations took: {}.", current);
         }
 
         for (Entity entity : documentEntities) {
@@ -87,10 +101,19 @@ public class EntityRedactionService {
             }
         }
 
+        current = System.currentTimeMillis() - startTime - current;
+        log.info("Adding entity position sequence took: {}.", current);
+
         dictionaryService.updateExternalDictionary(dictionary, ruleSetId);
 
         classifiedDoc.setDictionaryVersion(dictionary.getVersion());
         classifiedDoc.setRulesVersion(rulesVersion);
+
+        current = System.currentTimeMillis() - startTime - current;
+        log.info("Adjusting rules set took: {}.", current);
+
+        log.info("Finishing document processing took: {}.", System.currentTimeMillis() - startTime);
+
     }
 
 
@@ -113,7 +136,18 @@ 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 +191,6 @@ 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 +203,14 @@ 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 +241,8 @@ public class EntityRedactionService {
                     .build(), searchableRow));
 
             sectionNumber.incrementAndGet();
-            rowNumber++;
         }
+
         return sectionSearchableTextPairs;
     }
 
@@ -251,14 +287,13 @@ 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 +304,7 @@ 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)

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

@@ -34,7 +34,9 @@ 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 {
@@ -47,6 +49,9 @@ public class RedactionLogCreatorService {
     public void createRedactionLog(Document classifiedDoc, int numberOfPages, ManualRedactions manualRedactions,
                                    String ruleSetId) {
 
+        log.info("Creating redaction log...");
+        long startTime = System.currentTimeMillis();
+
         Set<Integer> manualRedactionPages = getManualRedactionPages(manualRedactions);
 
         for (int page = 1; page <= numberOfPages; page++) {
@@ -57,14 +62,25 @@ public class RedactionLogCreatorService {
                 addEntries(classifiedDoc, manualRedactions, page, ruleSetId);
             }
 
+            long current = System.currentTimeMillis() - startTime;
+            log.info("Adding entries took: {}.", current);
+
             if (manualRedactionPages.contains(page)) {
                 addManualEntries(classifiedDoc, manualRedactions, page, ruleSetId);
             }
 
+            current = System.currentTimeMillis() - startTime - current;
+            log.info("Adding manual entries took: {}.", current);
+
             if (!classifiedDoc.getPages().get(page - 1).getImageBounds().isEmpty()) {
                 addImageEntries(classifiedDoc, page, ruleSetId);
             }
+
+            current = System.currentTimeMillis() - startTime - current;
+            log.info("Adding image entries took: {}.", current);
         }
+
+        log.info("Finishing redaction log creation took: {}.", System.currentTimeMillis() - startTime);
     }
 
 
@@ -194,7 +210,7 @@ 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());
             }
         }

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

@@ -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);
+    }
 }

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

@@ -39,6 +39,9 @@ public class PdfSegmentationService {
 
     public Document parseDocument(PDDocument pdDocument) throws IOException {
 
+        log.info("Parsing document...");
+        long startTime = System.currentTimeMillis();
+
         Document document = new Document();
 
         List<Page> pages = new ArrayList<>();
@@ -61,13 +64,13 @@ 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());
@@ -91,6 +94,8 @@ public class PdfSegmentationService {
 
         sectionsBuilderService.buildSections(document);
 
+        log.info("Finishing document parsing took: {}.", System.currentTimeMillis() - startTime);
+
         return document;
     }
 

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

@@ -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 +19,52 @@ 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.debug("Building sections...");
+        long startTime = System.currentTimeMillis();
+
         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 +82,8 @@ 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 +91,11 @@ public class SectionsBuilderService {
         chunkBlockList.add(chunkBlock);
 
         document.setParagraphs(chunkBlockList);
+        document.setHeaders(headers);
+        document.setFooters(footers);
+
+        log.debug("Finishing section building took: {}.", System.currentTimeMillis() - startTime);
+
     }
 
 
@@ -175,9 +204,9 @@ 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 +217,9 @@ public class SectionsBuilderService {
 
         for (int i = table.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
             List<Cell> row = table.getRows().get(i);
+            if (row.size() == 1) {
+                continue;
+            }
             boolean allNonHeader = true;
             for (Cell cell : row) {
                 if (cell.isHeaderCell()) {

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

@@ -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);

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

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

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

@@ -23,11 +23,17 @@ 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.debug("Extracting tables...");
+        long startTime = System.currentTimeMillis();
+
         List<Cell> cells = findCells(cleanRulings.getHorizontal(), cleanRulings.getVertical());
 
         List<TextBlock> toBeRemoved = new ArrayList<>();
@@ -78,6 +84,9 @@ public class TableExtractionService {
         }
 
         page.getTextBlocks().removeAll(toBeRemoved);
+
+        log.debug("Finishing table extraction took: {}.", System.currentTimeMillis() - startTime);
+
     }
 
 

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

@@ -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()))