Pull request #40: Test modified rules 8 and 10

Merge in RED/redaction-service from RED-288 to master

* commit 'cfb7554c618b9482f529b8183bc0f1f95df7d00f':
  Test modified rules 8 and 9

bamboo-specs/pom.xml

@@ -5,7 +5,7 @@
   <parent>
     <groupId>com.atlassian.bamboo</groupId>
     <artifactId>bamboo-specs-parent</artifactId>
-    <version>7.0.4</version>
+    <version>7.1.2</version>
     <relativePath/>
   </parent>
 
@@ -34,4 +34,4 @@
 
   <!-- run 'mvn test' to perform offline validation of the plan -->
   <!-- run 'mvn -Ppublish-specs' to upload the plan to your Bamboo server -->
-</project>
+</project>

redaction-service-image-v1/src/main/docker/Dockerfile

@@ -1,4 +1,4 @@
-FROM gin5/platform-base:5.2.0
+FROM red/base-image:1.0.0
 
 ARG PLATFORM_JAR
 
@@ -6,4 +6,4 @@ ENV PLATFORM_JAR ${PLATFORM_JAR}
 
 ENV USES_ELASTICSEARCH false
 
-COPY ["${PLATFORM_JAR}", "/"]
+COPY ["${PLATFORM_JAR}", "/"]

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

@@ -11,6 +11,19 @@ import lombok.NoArgsConstructor;
 @NoArgsConstructor
 public class RedactionLog {
 
+    public RedactionLog(List<RedactionLogEntry> redactionLogEntry, long dictionaryVersion, long rulesVersion) {
+
+        this.redactionLogEntry = redactionLogEntry;
+        this.dictionaryVersion = dictionaryVersion;
+        this.rulesVersion = rulesVersion;
+    }
+
     private List<RedactionLogEntry> redactionLogEntry;
 
+    private long dictionaryVersion = -1;
+
+    private long rulesVersion = -1;
+
+    private String filename;
+
 }

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

@@ -84,7 +84,7 @@ public class ClassificationService {
         }
         else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getMostPopularWordFontSize() == document.getFontSizeCounter().getMostPopular() && textBlock.getMostPopularWordStyle().equals("italic") && !document.getFontStyleCounter().getMostPopular().equals("italic") && PositionUtils.getApproxLineCount(textBlock) < 2.9) {
             textBlock.setClassification("TextBlock Italic");
-        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock) && textBlock.getSequences().size() > 3){
+        } else if (PositionUtils.isWithinBodyTextFrame(bodyTextFrame, textBlock)){
             textBlock.setClassification("TextBlock Unknown");
         }
     }

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

@@ -3,18 +3,21 @@ package com.iqser.red.service.redaction.v1.server.controller;
 import java.io.ByteArrayInputStream;
 import java.io.ByteArrayOutputStream;
 import java.io.IOException;
+import java.util.List;
 
 import org.apache.pdfbox.pdmodel.PDDocument;
 import org.springframework.web.bind.annotation.RequestBody;
 import org.springframework.web.bind.annotation.RestController;
 
 import com.iqser.red.service.redaction.v1.model.RedactionLog;
+import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
 import com.iqser.red.service.redaction.v1.model.RedactionRequest;
 import com.iqser.red.service.redaction.v1.model.RedactionResult;
 import com.iqser.red.service.redaction.v1.resources.RedactionResource;
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
 import com.iqser.red.service.redaction.v1.server.classification.model.Page;
 import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
+import com.iqser.red.service.redaction.v1.server.redaction.service.DictionaryService;
 import com.iqser.red.service.redaction.v1.server.redaction.service.DroolsExecutionService;
 import com.iqser.red.service.redaction.v1.server.redaction.service.EntityRedactionService;
 import com.iqser.red.service.redaction.v1.server.segmentation.PdfSegmentationService;
@@ -36,6 +39,7 @@ public class RedactionController implements RedactionResource {
     private final EntityRedactionService entityRedactionService;
     private final PdfFlattenService pdfFlattenService;
     private final DroolsExecutionService droolsExecutionService;
+    private final DictionaryService dictionaryService;
 
     @Override
     public RedactionResult redact(@RequestBody RedactionRequest redactionRequest) {
@@ -49,10 +53,10 @@ public class RedactionController implements RedactionResource {
 
             if (redactionRequest.isFlatRedaction()) {
                 PDDocument flatDocument = pdfFlattenService.flattenPDF(pdDocument);
-                return convert(flatDocument, classifiedDoc.getPages().size(), new RedactionLog(classifiedDoc.getRedactionLogEntities()));
+                return convert(flatDocument, classifiedDoc.getPages().size(), classifiedDoc.getRedactionLogEntities());
             }
 
-            return convert(pdDocument, classifiedDoc.getPages().size(), new RedactionLog(classifiedDoc.getRedactionLogEntities()));
+            return convert(pdDocument, classifiedDoc.getPages().size(), classifiedDoc.getRedactionLogEntities());
 
         } catch (IOException e) {
             throw new RedactionException(e);
@@ -130,14 +134,14 @@ public class RedactionController implements RedactionResource {
         return convert(document, numberOfPages, null);
     }
 
-    private RedactionResult convert(PDDocument document, int numberOfPages, RedactionLog redactionLog) throws IOException {
+    private RedactionResult convert(PDDocument document, int numberOfPages, List<RedactionLogEntry> redactionLogEntities) throws IOException {
 
         try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream()) {
             document.save(byteArrayOutputStream);
             return RedactionResult.builder()
                     .document(byteArrayOutputStream.toByteArray())
                     .numberOfPages(numberOfPages)
-                    .redactionLog(redactionLog)
+                    .redactionLog(new RedactionLog(redactionLogEntities,  dictionaryService.getDictionaryVersion(), droolsExecutionService.getRulesVersion()))
                     .build();
         }
 

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

@@ -0,0 +1,16 @@
+package com.iqser.red.service.redaction.v1.server.redaction.model;
+
+import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+
+import lombok.RequiredArgsConstructor;
+import lombok.Value;
+
+@Value
+@RequiredArgsConstructor
+public class CellValue {
+
+    TextBlock textBlock;
+
+    int rowSpanStart;
+
+}

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

@@ -9,8 +9,6 @@ import java.util.regex.Pattern;
 
 import org.apache.commons.lang3.StringUtils;
 
-import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
-
 import lombok.Builder;
 import lombok.Data;
 import lombok.extern.slf4j.Slf4j;
@@ -32,24 +30,16 @@ public class Section {
 
     private int sectionNumber;
 
-    private Map<String, TextBlock> tabularData;
+    private Map<String, CellValue> tabularData;
 
 
-    public boolean isVertebrateStudy() {
-        return tabularData != null
-                && (tabularData.containsKey("Vertebrate study Y/N")
-                && tabularData.get("Vertebrate study Y/N").getText().equals("Y")
-                || tabularData.containsKey("Verte brate study Y/N")
-                && tabularData.get("Verte brate study Y/N").getText().equals("Y"));
-    }
+    public boolean rowEquals(String headerName, String value){
+        String cleanHeaderName = headerName.replaceAll("\n", "")
+                .replaceAll(" ", "")
+                .replaceAll("-", "");
 
-
-    public boolean isNotVertebrateStudy() {
-        return tabularData != null
-                && (tabularData.containsKey("Vertebrate study Y/N")
-                && tabularData.get("Vertebrate study Y/N").getText().equals("N")
-                || tabularData.containsKey("Verte brate study Y/N")
-                && tabularData.get("Verte brate study Y/N").getText().equals("N"));
+        return tabularData != null && tabularData.containsKey(cleanHeaderName)
+                && tabularData.get(cleanHeaderName).getTextBlock().getText().equals(value);
     }
 
 
@@ -182,18 +172,25 @@ public class Section {
 
     public void highlightCell(String cellHeader, int ruleNumber, String type) {
 
-        TextBlock value = tabularData.get(cellHeader);
+        String cleanHeaderName = cellHeader.replaceAll("\n", "")
+                .replaceAll(" ", "")
+                .replaceAll("-", "");
+
+        CellValue value = tabularData.get(cleanHeaderName);
         if (value == null) {
             log.warn("Could not find any data for {}.", cellHeader);
         } else {
-            Entity entity = new Entity(value.getText(), type, 0, value.getText().length(), headline, sectionNumber);
+            Entity entity = new Entity(value.getTextBlock()
+                    .getText(), type, value.getRowSpanStart(), value.getRowSpanStart() + value.getTextBlock()
+                    .getText()
+                    .length(), headline, sectionNumber);
             entity.setRedaction(false);
             entity.setMatchedRule(ruleNumber);
             entity.setRedactionReason(cellHeader);
-            entity.setTargetSequences(value.getSequences()); // Make sure no other cells with same content are highlighted
+            entity.setTargetSequences(value.getTextBlock().getSequences()); // Make sure no other cells with same content are highlighted
             entities.add(entity);
         }
 
     }
 
-}
+}

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

@@ -27,6 +27,7 @@ public class DictionaryService {
 
     private final DictionaryClient dictionaryClient;
 
+    @Getter
     private long dictionaryVersion = -1;
 
     @Getter

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

@@ -18,6 +18,7 @@ import com.iqser.red.service.redaction.v1.server.client.RulesClient;
 import com.iqser.red.service.redaction.v1.server.exception.RulesValidationException;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Section;
 
+import lombok.Getter;
 import lombok.RequiredArgsConstructor;
 
 @Service
@@ -29,6 +30,7 @@ public class DroolsExecutionService {
     @Autowired
     private KieContainer kieContainer;
 
+    @Getter
     private long rulesVersion = -1;
 
     public Section executeRules(Section section) {

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

@@ -18,6 +18,7 @@ import com.iqser.red.service.redaction.v1.model.Rectangle;
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
 import com.iqser.red.service.redaction.v1.server.classification.model.Paragraph;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
 import com.iqser.red.service.redaction.v1.server.redaction.model.EntityPositionSequence;
 import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
@@ -51,24 +52,30 @@ public class EntityRedactionService {
             List<Table> tables = paragraph.getTables();
 
             for (Table table : tables) {
+                boolean singleCellTable = table.getRowCount() == 1 && table.getColCount() == 1;
                 for (List<Cell> row : table.getRows()) {
                     SearchableText searchableRow = new SearchableText();
-                    Map<String, TextBlock> tabularData = new HashMap<>();
+                    Map<String, CellValue> tabularData = new HashMap<>();
+                    int start = 0;
                     for (Cell cell : row) {
-                        if (cell.isHeaderCell() || CollectionUtils.isEmpty(cell.getTextBlocks())) {
+                        if (!singleCellTable && cell.isHeaderCell() || CollectionUtils.isEmpty(cell.getTextBlocks())) {
                             continue;
                         }
                         addSectionToManualRedactions(cell.getTextBlocks(), manualRedactions, table.getHeadline(), sectionNumber);
+                        int cellStart = start;
                         cell.getHeaderCells().forEach(headerCell -> {
-                            String headerName = headerCell.getTextBlocks().get(0).getText()
-                                    .replaceAll("\n", " ")
-                                    .replaceAll(" ", " ");
-                            tabularData.put(headerName, cell.getTextBlocks().get(0));
+                            StringBuilder headerBuilder = new StringBuilder();
+                            headerCell.getTextBlocks().forEach(textBlock -> headerBuilder.append(textBlock.getText()));
+                            String headerName = headerBuilder.toString()
+                                    .replaceAll("\n", "")
+                                    .replaceAll(" ", "")
+                                    .replaceAll("-", "");
+                            tabularData.put(headerName, new CellValue(cell.getTextBlocks().get(0), cellStart));
                         });
+                        start = start + cell.getTextBlocks().get(0).toString().length();
                         for (TextBlock textBlock : cell.getTextBlocks()) {
                             searchableRow.addAll(textBlock.getSequences());
                         }
-
                     }
                     Set<Entity> rowEntities = findEntities(searchableRow, table.getHeadline(), sectionNumber);
 
@@ -138,7 +145,7 @@ public class EntityRedactionService {
     private Set<Entity> findEntities(SearchableText searchableText, String headline, int sectionNumber) {
 
         Set<Entity> found = new HashSet<>();
-        if (StringUtils.isEmpty(searchableText.toString()) && StringUtils.isEmpty(headline)) {
+        if (StringUtils.isEmpty(searchableText.toString())) {
             return found;
         }
 

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

@@ -4,6 +4,7 @@ import java.util.ArrayList;
 import java.util.Collections;
 import java.util.Iterator;
 import java.util.List;
+import java.util.stream.Collectors;
 
 import org.apache.commons.collections4.CollectionUtils;
 import org.springframework.stereotype.Service;
@@ -39,22 +40,27 @@ public class SectionsBuilderService {
                 current.setPage(page.getPageNumber());
 
                 if (prev != null && current.getClassification().startsWith("H ") || !document.isHeadlines()) {
-                    Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline, previousTable);
+                    Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
                     chunkBlock.setHeadline(lastHeadline);
                     lastHeadline = current.getText();
-                    if (CollectionUtils.isNotEmpty(chunkBlock.getTables())) {
-                        previousTable = chunkBlock.getTables().get(0);
-                    }
                     chunkBlockList.add(chunkBlock);
                     chunkWords = new ArrayList<>();
+                    if (CollectionUtils.isNotEmpty(chunkBlock.getTables())) {
+                        previousTable = chunkBlock.getTables().get(chunkBlock.getTables().size() - 1);
+                    }
+                }
+                if (current instanceof Table) {
+                    Table table = (Table) current;
+                    // Distribute header information for subsequent tables
+                    mergeTableMetadata(table, previousTable);
+                    previousTable = table;
                 }
-
                 chunkWords.add(current);
                 prev = current;
             }
         }
 
-        Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline, previousTable);
+        Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
         chunkBlock.setHeadline(lastHeadline);
         chunkBlockList.add(chunkBlock);
 
@@ -62,7 +68,38 @@ public class SectionsBuilderService {
     }
 
 
-    private Paragraph buildTextBlock(List<AbstractTextContainer> wordBlockList, String lastHeadline, Table previousTable) {
+    private void mergeTableMetadata(Table currentTable, Table previousTable) {
+
+        // Distribute header information for subsequent tables
+        if (previousTable != null && hasInvalidHeaderInformation(currentTable) && hasValidHeaderInformation(previousTable)) {
+            List<Cell> previousTableNonHeaderRow = getRowWithNonHeaderCells(previousTable);
+            List<Cell> tableNonHeaderRow = getRowWithNonHeaderCells(currentTable);
+            // Allow merging of tables if header row is separated from first logical non-header row
+            if (previousTableNonHeaderRow.isEmpty() && previousTable.getRowCount() == 1 && previousTable.getRows()
+                    .get(0)
+                    .size() == tableNonHeaderRow.size()) {
+                previousTableNonHeaderRow = previousTable.getRows().get(0).stream().map(cell -> {
+                    Cell fakeCell = new Cell(cell.getPoints()[0], cell.getPoints()[2]);
+                    fakeCell.setHeaderCells(Collections.singletonList(cell));
+                    return fakeCell;
+                }).collect(Collectors.toList());
+            }
+            if (previousTableNonHeaderRow.size() == tableNonHeaderRow.size()) {
+                for (int i = currentTable.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
+                    List<Cell> row = currentTable.getRows().get(i);
+                    if (row.size() == tableNonHeaderRow.size() && row.stream()
+                            .allMatch(cell -> cell.getHeaderCells().isEmpty())) {
+                        for (int j = 0; j < row.size(); j++) {
+                            row.get(j).setHeaderCells(previousTableNonHeaderRow.get(j).getHeaderCells());
+                        }
+                    }
+                }
+            }
+        }
+    }
+
+
+    private Paragraph buildTextBlock(List<AbstractTextContainer> wordBlockList, String lastHeadline) {
 
         Paragraph paragraph = new Paragraph();
         TextBlock textBlock = null;
@@ -85,27 +122,6 @@ public class SectionsBuilderService {
                 } else {
                     table.setHeadline("Table in: " + lastHeadline);
                 }
-                // Distribute header information for subsequent tables
-                if (previousTable != null && hasInvalidHeaderInformation(table) && hasValidHeaderInformation(previousTable)) {
-                    List<Cell> previousTableNonHeaderRow = getRowWithNonHeaderCells(previousTable);
-                    List<Cell> tableNonHeaderRow = getRowWithNonHeaderCells(table);
-                    // Allow merging of tables if header row is separated from first logical non-header row
-                    if (previousTableNonHeaderRow.isEmpty() && previousTable.getRowCount() == 1
-                            && previousTable.getRows().get(0).size() == tableNonHeaderRow.size()) {
-                        previousTableNonHeaderRow = previousTable.getRows().get(0);
-                    }
-                    if (previousTableNonHeaderRow.size() == tableNonHeaderRow.size()) {
-                        for (int i = table.getRows().size() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
-                            List<Cell> row = table.getRows().get(i);
-                            if (row.size() == tableNonHeaderRow.size()
-                                    && row.stream().allMatch(cell -> cell.getHeaderCells().isEmpty())) {
-                                for (int j = 0; j < row.size(); j++) {
-                                    row.get(j).setHeaderCells(previousTableNonHeaderRow.get(j).getHeaderCells());
-                                }
-                            }
-                        }
-                    }
-                }
 
                 if (textBlock != null && !alreadyAdded) {
                     paragraph.getPageBlocks().add(textBlock);
@@ -168,7 +184,7 @@ public class SectionsBuilderService {
 
     private List<Cell> getRowWithNonHeaderCells(Table table) {
 
-        for (int i = table.getRows().size() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
+        for (int i = table.getRowCount() - 1; i >= 0; i--) { // Non header rows are most likely at bottom of table
             List<Cell> row = table.getRows().get(i);
             boolean allNonHeader = true;
             for (Cell cell : row) {

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

@@ -29,11 +29,13 @@ public class Table extends AbstractTextContainer {
     @Setter
     private String headline;
 
-    @Getter
-    private int rowCount;
+    private int unrotatedRowCount;
 
-    @Getter
-    private int colCount;
+    private int unrotatedColCount;
+
+    private int rowCount = -1;
+
+    private int colCount = -1;
 
     private final int rotation;
 
@@ -65,6 +67,25 @@ public class Table extends AbstractTextContainer {
     }
 
 
+    public int getRowCount() {
+
+        if (rowCount == -1) {
+            rowCount = getRows().size();
+        }
+        return rowCount;
+    }
+
+
+    public int getColCount() {
+
+        if (colCount == -1) {
+            colCount = getRows().stream().mapToInt(List::size).max().orElse(0);
+        }
+        return colCount;
+
+    }
+
+
     /**
      * Detect header cells (either first row or first column):
      * Column is marked as header if cell text is bold and row cell text is not bold.
@@ -72,100 +93,54 @@ public class Table extends AbstractTextContainer {
      */
     private void computeHeaders() {
 
+        if (rows == null) {
+            rows = computeRows();
+        }
         // A bold cell is a header cell as long as every cell to the left/top is bold, too
-        cells.forEach((position, cell) -> {
-            List<Cell> cellsToTheLeft = getCellsToTheLeft(position);
-            Cell lastHeaderCell = null;
-            for (Cell leftCell : cellsToTheLeft) {
-                if (CollectionUtils.isNotEmpty(leftCell.getTextBlocks()) && leftCell.getTextBlocks()
+        // we move from left to right and top to bottom
+        for (int rowIndex = 0; rowIndex < rows.size(); rowIndex++) {
+            List<Cell> rowCells = rows.get(rowIndex);
+            for (int colIndex = 0; colIndex < rowCells.size(); colIndex++) {
+                Cell cell = rowCells.get(colIndex);
+                List<Cell> cellsToTheLeft = rowCells.subList(0, colIndex);
+                Cell lastHeaderCell = null;
+                for (Cell leftCell : cellsToTheLeft) {
+                    if (leftCell.isHeaderCell()) {
+                        lastHeaderCell = leftCell;
+                    } else {
+                        break;
+                    }
+                }
+                if (lastHeaderCell != null) {
+                    cell.getHeaderCells().add(lastHeaderCell);
+                }
+                List<Cell> cellsToTheTop = new ArrayList<>();
+                for (int i = 0; i < rowIndex; i++) {
+                    try {
+                        cellsToTheTop.add(rows.get(i).get(colIndex));
+                    } catch (IndexOutOfBoundsException e) {
+                        log.debug("No cell {} in row {}, ignoring.", colIndex, rowIndex);
+                    }
+                }
+                for (Cell topCell : cellsToTheTop) {
+                    if (topCell.isHeaderCell()) {
+                        lastHeaderCell = topCell;
+                    } else {
+                        break;
+                    }
+                }
+                if (lastHeaderCell != null) {
+                    cell.getHeaderCells().add(lastHeaderCell);
+                }
+                if (CollectionUtils.isNotEmpty(cell.getTextBlocks()) && cell.getTextBlocks()
                         .get(0)
                         .getMostPopularWordStyle()
                         .equals("bold")) {
-                    lastHeaderCell = leftCell;
-                } else {
-                    break;
+                    cell.setHeaderCell(true);
                 }
             }
-            if (lastHeaderCell != null) {
-                cell.getHeaderCells().add(lastHeaderCell);
-            }
-            lastHeaderCell = null;
-            List<Cell> cellsToTheTop = getCellToTheTop(position);
-            for (Cell topCell : cellsToTheTop) {
-                if (CollectionUtils.isNotEmpty(topCell.getTextBlocks()) && topCell.getTextBlocks()
-                        .get(0)
-                        .getMostPopularWordStyle()
-                        .equals("bold")) {
-                    lastHeaderCell = topCell;
-                } else {
-                    break;
-                }
-            }
-            if (lastHeaderCell != null) {
-                cell.getHeaderCells().add(lastHeaderCell);
-            }
-            if (CollectionUtils.isNotEmpty(cell.getTextBlocks()) && cell.getTextBlocks()
-                    .get(0)
-                    .getMostPopularWordStyle()
-                    .equals("bold")) {
-                cell.setHeaderCell(true);
-            }
-        });
-
-    }
-
-
-    private List<Cell> getCellsToTheLeft(CellPosition cellPosition) {
-
-        List<Cell> result = new ArrayList<>();
-        if (cellPosition.getCol() == 0) {
-            return result;
-        }
-        int row = cellPosition.getRow();
-        for (int i = cellPosition.getCol() - 1; i >= 0; i--) {
-            if (cells.get(new CellPosition(row, i)) != null) {
-                result.add(cells.get(new CellPosition(row, i)));
-            } else {
-                Cell spanningCell = null;
-                while (spanningCell == null && row >= 0) {
-                    row--;
-                    spanningCell = cells.get(new CellPosition(row, i));
-                }
-                if (spanningCell != null) {
-                    result.add(spanningCell);
-                }
-                row = cellPosition.getRow();
-            }
         }
-        Collections.reverse(result);
-        return result;
-    }
 
-
-    private List<Cell> getCellToTheTop(CellPosition cellPosition) {
-
-        List<Cell> result = new ArrayList<>();
-        if (cellPosition.getRow() == 0) {
-            return result;
-        }
-        int col = cellPosition.getCol();
-        for (int i = cellPosition.getRow() - 1; i >= 0; i--) {
-            if (cells.get(new CellPosition(i, col)) != null) {
-                result.add(cells.get(new CellPosition(i, col)));
-            } else {
-                Cell spanningCell = null;
-                while (spanningCell == null && col >= 0) {
-                    col--;
-                    spanningCell = cells.get(new CellPosition(i, col));
-                }
-                if (spanningCell != null) {
-                    result.add(spanningCell);
-                }
-                col = cellPosition.getCol();
-            }
-        }
-        Collections.reverse(result);
-        return result;
     }
 
 
@@ -173,9 +148,9 @@ public class Table extends AbstractTextContainer {
 
         List<List<Cell>> rows = new ArrayList<>();
         if (rotation == 90) {
-            for (int i = 0; i < colCount; i++) { // rows
+            for (int i = 0; i < unrotatedColCount; i++) { // rows
                 List<Cell> lastRow = new ArrayList<>();
-                for (int j = rowCount - 1; j >= 0; j--) { // cols
+                for (int j = unrotatedRowCount - 1; j >= 0; j--) { // cols
                     Cell cell = cells.get(new CellPosition(j, i));
                     if (cell != null) {
                         lastRow.add(cell);
@@ -184,9 +159,9 @@ public class Table extends AbstractTextContainer {
                 rows.add(lastRow);
             }
         } else if (rotation == 270) {
-            for (int i = colCount - 1; i >= 0; i--) { // rows
+            for (int i = unrotatedColCount - 1; i >= 0; i--) { // rows
                 List<Cell> lastRow = new ArrayList<>();
-                for (int j = 0; j < rowCount; j++) { // cols
+                for (int j = 0; j < unrotatedRowCount; j++) { // cols
                     Cell cell = cells.get(new CellPosition(i, j));
                     if (cell != null) {
                         lastRow.add(cell);
@@ -195,9 +170,9 @@ public class Table extends AbstractTextContainer {
                 rows.add(lastRow);
             }
         } else {
-            for (int i = 0; i < rowCount; i++) {
+            for (int i = 0; i < unrotatedRowCount; i++) {
                 List<Cell> lastRow = new ArrayList<>();
-                for (int j = 0; j < colCount; j++) {
+                for (int j = 0; j < unrotatedColCount; j++) {
                     Cell cell = cells.get(new CellPosition(i, j)); // JAVA_8 use getOrDefault()
                     if (cell != null) {
                         lastRow.add(cell);
@@ -214,8 +189,8 @@ public class Table extends AbstractTextContainer {
 
     private void add(Cell chunk, int row, int col) {
 
-        rowCount = Math.max(rowCount, row + 1);
-        colCount = Math.max(colCount, col + 1);
+        unrotatedRowCount = Math.max(unrotatedRowCount, row + 1);
+        unrotatedColCount = Math.max(unrotatedColCount, col + 1);
 
         CellPosition cp = new CellPosition(row, col);
         cells.put(cp, chunk);

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

@@ -2,7 +2,9 @@ package com.iqser.red.service.redaction.v1.server.visualization.service;
 
 import java.awt.Color;
 import java.io.IOException;
+import java.util.HashSet;
 import java.util.List;
+import java.util.Set;
 
 import org.apache.commons.collections4.CollectionUtils;
 import org.apache.pdfbox.pdmodel.PDDocument;
@@ -45,22 +47,42 @@ public class AnnotationHighlightService {
 
     public void highlight(PDDocument document, Document classifiedDoc, boolean flatRedaction, ManualRedactions manualRedactions) throws IOException {
 
+        Set<Integer> manualRedactionPages = getManualRedactionPages(manualRedactions);
+
         for (int page = 1; page <= document.getNumberOfPages(); page++) {
 
             PDPage pdPage = document.getPage(page - 1);
 
             drawSectionFrames(document, classifiedDoc, flatRedaction, pdPage, page);
 
-            if (classifiedDoc.getEntities().get(page) == null) {
-                continue;
+            if (classifiedDoc.getEntities().get(page) != null) {
+                addAnnotations(pdPage, classifiedDoc, flatRedaction, manualRedactions, page);
             }
 
-            addAnnotations(pdPage, classifiedDoc, flatRedaction, manualRedactions, page);
-            addManualAnnotations(pdPage, classifiedDoc, manualRedactions, page);
+            if (manualRedactionPages.contains(page)) {
+                addManualAnnotations(pdPage, classifiedDoc, manualRedactions, page);
+            }
         }
     }
 
 
+    private Set<Integer> getManualRedactionPages(ManualRedactions manualRedactions) {
+
+        Set<Integer> manualRedactionPages = new HashSet<>();
+
+        if (manualRedactions == null) {
+            return manualRedactionPages;
+        }
+
+        manualRedactions.getEntriesToAdd().forEach(entry -> {
+            entry.getPositions().forEach(pos -> {
+                manualRedactionPages.add(pos.getPage());
+            });
+        });
+        return manualRedactionPages;
+    }
+
+
     private void addAnnotations(PDPage pdPage, Document classifiedDoc, boolean flatRedaction, ManualRedactions manualRedactions, int page) throws IOException {
 
         List<PDAnnotation> annotations = pdPage.getAnnotations();
@@ -77,8 +99,11 @@ public class AnnotationHighlightService {
 
                 if (manualRedactions != null && manualRedactions.getIdsToRemove()
                         .contains(entityPositionSequence.getId())) {
+                    String manualOverrideReason = entity.getRedactionReason() + ", removed by manual override";
                     entity.setRedaction(false);
-                    entity.setRedactionReason(entity.getRedactionReason() + ", removed by manual override");
+                    entity.setRedactionReason(manualOverrideReason);
+                    redactionLogEntry.setRedacted(false);
+                    redactionLogEntry.setReason(manualOverrideReason);
                     redactionLogEntry.setManual(true);
                 }
 
@@ -107,28 +132,34 @@ public class AnnotationHighlightService {
 
             String id = IdBuilder.buildId(manualRedactionEntry);
 
-            RedactionLogEntry redactionLogEntry = createRedactionLogEntry(manualRedactionEntry);
+            RedactionLogEntry redactionLogEntry = createRedactionLogEntry(manualRedactionEntry, id);
 
+            boolean foundOnPage = false;
             for (Rectangle rectangle : manualRedactionEntry.getPositions()) {
 
                 if (page != rectangle.getPage()) {
                     continue;
                 }
 
+                foundOnPage = true;
+
                 PDAnnotationTextMarkup highlight = createAnnotation(rectangle, id, createAnnotationContent(manualRedactionEntry), getColor(manualRedactionEntry
                         .getType()), true);
                 annotations.add(highlight);
 
                 redactionLogEntry.getPositions().add(rectangle);
             }
-            classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
+            if (foundOnPage) {
+                classifiedDoc.getRedactionLogEntities().add(redactionLogEntry);
+            }
         }
     }
 
 
-    private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry) {
+    private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry, String id) {
 
         return RedactionLogEntry.builder()
+                .id(id)
                 .color(getColor(manualRedactionEntry.getType()))
                 .reason(manualRedactionEntry.getReason())
                 .value(manualRedactionEntry.getValue())

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

@@ -269,7 +269,7 @@ public class RedactionIntegrationTest {
 
         System.out.println("redactionTest");
         long start = System.currentTimeMillis();
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Trinexapac/96 Trinexapac-ethyl_RAR_09_Volume_3CA_B-7_2018-02-23.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/new/Thiabendazole DAR Addendum for ED_April_2020.pdf");
 
         RedactionRequest request = RedactionRequest.builder()
                 .document(IOUtils.toByteArray(pdfFileResource.getInputStream()))

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

@@ -57,6 +57,7 @@ public class EntityRedactionServiceTest {
     private static final String ADDRESS_CODE = "address";
 
     private static final AtomicLong DICTIONARY_VERSION = new AtomicLong();
+    private static final AtomicLong RULES_VERSION = new AtomicLong();
     @MockBean
     private DictionaryClient dictionaryClient;
 
@@ -69,6 +70,9 @@ public class EntityRedactionServiceTest {
     @Autowired
     private PdfSegmentationService pdfSegmentationService;
 
+    @Autowired
+    private DroolsExecutionService droolsExecutionService;
+
     @TestConfiguration
     public static class RedactionIntegrationTestConfiguration {
 
@@ -130,7 +134,7 @@ public class EntityRedactionServiceTest {
             Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
             entityRedactionService.processDocument(classifiedDoc, null);
             assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-            assertThat(classifiedDoc.getEntities().get(1)).hasSize(5); // 4 out of 5 entities recognized on page 1
+            assertThat(classifiedDoc.getEntities().get(1)).hasSize(7);// 3 names, 1 address, 1 Y and 2 N entities
         }
     }
 
@@ -185,7 +189,7 @@ public class EntityRedactionServiceTest {
         try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
             Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
             entityRedactionService.processDocument(classifiedDoc, null);
-            assertThat(classifiedDoc.getEntities()).hasSize(1); // two pages
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
             assertThat(classifiedDoc.getEntities().get(1).stream()
                     .filter(entity -> entity.getMatchedRule() == 9)
                     .count()).isEqualTo(10);
@@ -193,6 +197,61 @@ public class EntityRedactionServiceTest {
 
     }
 
+
+    @Test
+    public void testApplicantInTableRedaction() throws IOException {
+
+        String tableRules = "package drools\n" +
+                "\n" +
+                "import com.iqser.red.service.redaction.v1.server.redaction.model.Section\n" +
+                "\n" +
+                "global Section section\n" +
+                "rule \"6: Redact contact information if applicant is found\"\n" +
+                "   when\n" +
+                "       eval(section.headlineContainsWord(\"applicant\") || section.getText().contains(\"Applicant\"));\n" +
+                "   then\n" +
+                "       section.redactLineAfter(\"Name:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactBetween(\"Address:\", \"Contact\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Contact point:\",  \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Phone:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Fax:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Tel.:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Tel:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"E-mail:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Email:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Contact:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Telephone number:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Fax number:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactLineAfter(\"Telephone:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactBetween(\"No:\", \"Fax\", \"address\", 6, \"Applicant information was found\");\n" +
+                "       section.redactBetween(\"Contact:\", \"Tel.:\", \"address\", 6, \"Applicant information was found\");\n" +
+                "   end";
+        when(rulesClient.getVersion()).thenReturn(RULES_VERSION.incrementAndGet());
+        when(rulesClient.getRules()).thenReturn(new RulesResponse(tableRules));
+        droolsExecutionService.updateRules();
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Applicant Producer Table.pdf");
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
+                .entries(new ArrayList<>(ResourceLoader.load("dictionaries/names.txt")))
+                .build();
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        DictionaryResponse addressResponse = DictionaryResponse.builder()
+                .entries(new ArrayList<>(ResourceLoader.load("dictionaries/addresses.txt")))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+            assertThat(classifiedDoc.getEntities().get(1).stream()
+                    .filter(entity -> entity.getMatchedRule() == 6)
+                    .count()).isEqualTo(18);
+        }
+
+    }
+
+
     @Test
     public void headerPropagation() throws IOException {
 
@@ -216,6 +275,52 @@ public class EntityRedactionServiceTest {
             assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(8);
             assertThat(classifiedDoc.getEntities().get(2).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(4);
         }
+
+        pdfFileResource = new ClassPathResource("files/Minimal Examples/Header Propagation2.pdf");
+
+        dictionaryResponse = DictionaryResponse.builder()
+                .entries(Arrays.asList("Tribolet, R.", "Muir, G.", "Kühne-Thu, H.", "Close, C."))
+                .build();
+
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        addressResponse = DictionaryResponse.builder()
+                .entries(Collections.singletonList("Novartis Crop Protection AG, Basel, Switzerland"))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(3);
+            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(8);
+        }
+    }
+
+
+    @Test
+    public void testNGuideline() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Empty Tabular Data.pdf");
+
+        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
+                .entries(Collections.singletonList("Aldershof S."))
+                .build();
+
+        when(dictionaryClient.getVersion()).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(dictionaryResponse);
+        DictionaryResponse addressResponse = DictionaryResponse.builder()
+                .entries(Collections.singletonList("Novartis Crop Protection AG, Basel, Switzerland"))
+                .build();
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(addressResponse);
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+            entityRedactionService.processDocument(classifiedDoc, null);
+            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(6);
+        }
     }
 
 
@@ -226,15 +331,24 @@ public class EntityRedactionServiceTest {
                 "import com.iqser.red.service.redaction.v1.server.redaction.model.Section\n" +
                 "\n" +
                 "global Section section\n" +
+                "rule \"8: Not redacted because Vertebrate Study = N\"\n" +
+                "    when\n" +
+                "        Section(rowEquals(\"Vertebrate study Y/N\", \"N\") || rowEquals(\"Vertebrate study Y/N\", \"No\"))\n" +
+                "    then\n" +
+                "        section.redactNot(\"name\", 8, \"Not redacted because row is not a vertebrate study\");\n" +
+                "        section.redactNot(\"address\", 8, \"Not redacted because row is not a vertebrate study\");\n" +
+                "        section.highlightCell(\"Vertebrate study Y/N\", 8, \"hint_only\");\n" +
+                "    end\n" +
                 "rule \"9: Redact Authors and Addresses in Reference Table, if it is a Vertebrate study\"\n" +
                 "    when\n" +
-                "        Section(isVertebrateStudy())\n" +
+                "        Section(rowEquals(\"Vertebrate study Y/N\", \"Y\") || rowEquals(\"Vertebrate study Y/N\", " +
+                "\"Yes\"))\n" +
                 "    then\n" +
                 "        section.redact(\"name\", 9, \"Redacted because row is a vertebrate study\");\n" +
-                "        section.redact(\"address\", 9, \"Redacted because rows is a vertebrate study\");\n" +
+                "        section.redact(\"address\", 9, \"Redacted because row is a vertebrate study\");\n" +
                 "        section.highlightCell(\"Vertebrate study Y/N\", 9, \"must_redact\");\n" +
                 "    end";
-        when(rulesClient.getVersion()).thenReturn(1L);
+        when(rulesClient.getVersion()).thenReturn(RULES_VERSION.incrementAndGet());
         when(rulesClient.getRules()).thenReturn(new RulesResponse(tableRules));
         TypeResponse typeResponse = TypeResponse.builder()
                 .types(Arrays.asList(

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

@@ -3,6 +3,7 @@ package com.iqser.red.service.redaction.v1.server.segmentation;
 import static org.assertj.core.api.Assertions.assertThat;
 
 import java.io.IOException;
+import java.util.Collections;
 import java.util.List;
 import java.util.stream.Collectors;
 
@@ -94,6 +95,46 @@ public class PdfSegmentationServiceTest {
             List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
                     .get(0)
                     .stream()
+                    .map(Collections::singletonList)
+                    .collect(Collectors.toList());
+            assertThat(secondTable.getRows().stream()
+                    .allMatch(row -> row.stream()
+                            .map(Cell::getHeaderCells)
+                            .collect(Collectors.toList())
+                            .equals(firstTableHeaderCells)))
+                    .isTrue();
+        }
+    }
+
+
+    @Test
+    public void testMultiPageMetadataPropagation() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Merge Multi Page Table.pdf");
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document document = pdfSegmentationService.parseDocument(pdDocument);
+            assertThat(document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())).isNotEmpty();
+            Table firstTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(0);
+            assertThat(firstTable.getColCount()).isEqualTo(9);
+            assertThat(firstTable.getRowCount()).isEqualTo(5);
+            Table secondTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(1);
+            assertThat(secondTable.getColCount()).isEqualTo(9);
+            assertThat(secondTable.getRowCount()).isEqualTo(6);
+            List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
+                    .get(firstTable.getRowCount() - 1)
+                    .stream()
                     .map(Cell::getHeaderCells)
                     .collect(Collectors.toList());
             assertThat(secondTable.getRows().stream()
@@ -105,4 +146,44 @@ public class PdfSegmentationServiceTest {
         }
     }
 
+
+    @Test
+    public void testHeaderCellsForRotatedTable() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Rotated Table Headers.pdf");
+
+        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
+            Document document = pdfSegmentationService.parseDocument(pdDocument);
+            assertThat(document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())).isNotEmpty();
+            Table firstTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(0);
+            assertThat(firstTable.getColCount()).isEqualTo(8);
+            assertThat(firstTable.getRowCount()).isEqualTo(1);
+            Table secondTable = document.getParagraphs()
+                    .stream()
+                    .flatMap(paragraph -> paragraph.getTables().stream())
+                    .collect(Collectors.toList())
+                    .get(1);
+            assertThat(secondTable.getColCount()).isEqualTo(8);
+            assertThat(secondTable.getRowCount()).isEqualTo(6);
+            List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
+                    .get(0)
+                    .stream()
+                    .map(Collections::singletonList)
+                    .collect(Collectors.toList());
+            assertThat(secondTable.getRows().stream()
+                    .allMatch(row -> row.stream()
+                            .map(Cell::getHeaderCells)
+                            .collect(Collectors.toList())
+                            .equals(firstTableHeaderCells)))
+                    .isTrue();
+        }
+    }
+
 }

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

@@ -227,6 +227,7 @@ CIBA-GEIGY Limited, Toxicology Services, Short-term Toxicology, 4332 Stein, Swit
 CIBA-GEIGY Ltd. CH-4002 Basle, Switzerland
 CIBA-GEIGY Ltd., Product Safety, Ecotoxicology, CH-4002 Basel, Switzerland
 CIBA-GEIGY Ltd., Switzerland
+CibaGeigy Ltd., CH-4002, Basel, Switzerland
 CIP Chemisches Institut Pforzheim GmbH, Pforzheim, Germany
 CL PHARMA AG
 CL PHARMA AG, Linz, Austria
@@ -1563,4 +1564,36 @@ Zyma SA
 Zyma SA, Nyon, Switzerland
 “Bayer CropScience AG” Monheim
 “Biologische Bundesanstalt (BBA)”, Berlin-Dahlem
-“W. Neudorff GmbH KG”, An der Mühle 3, D- 31860 Emmertal
+“W. Neudorff GmbH KG”, An der Mühle 3, D- 31860 Emmertal
+Syngenta Ltd Jealott’s Hill Int. Research Centre Bracknell, Berkshire, RG42 6EY, United Kingdom
+MITOX Consultants Science Park 408, 1098XH Amsterdam, NL
+Eurofins Agrosciences Services Eco-Chem GmbH, Eutinger Str. 24, 75233 Niefern-Öschelbronn, Germany
+Fraunhofer Institute for Molecular Biology and Applied Ecology (IME), 57392 Schmallenberg, Germany
+Syngenta Crop Protection AGRCC Cytotest Cell Research GmbH, Germany
+BASF SE, Ludwigshafen, Germany
+AGRCC Ltd., Switzerland
+AGWIL Research Lab. Inc., USA
+Charles River Laboratories, UK
+Ciba-Geigy Corporation Environmental Health Centre, Farmington
+Syngenta Crop Protection, Inc., Greensboro, United States
+Experimental Toxicology and Ecology, BASF SE
+Wildlife International, Ltd. 8598 Commerce Drive Easton, MD 21601 USA
+Küberich, Wiesentheid / Geesdorf, Germany
+CiToxLAB Hungary Ltd. H8200 Veszprém, Szabadságpuszta, Hungary
+Forellenzucht Trostadt GbR, Trostadt, Germany
+Osage Catfisheries Inc., Osage Beach, MO, USA
+Fischzucht Rhönforelle, Gersfeld, Germany
+Aquatic Biosystems, Ft. Collins, Colorado, USA
+West Aquarium GmbH, 37431 Bad Lauterberg, Germany
+Aquatic BioSystems, Fort Collins, Colorado, USA
+“U.S. Environmental Research Laboratory”, Duluth, Minnesota, USA
+Continuous laboratory cultures
+Wildlife International, Ltd. 8598 Commerce Drive Easton, MD 21601 USA
+Osage Catfisheries, Inc., Osage Beach, MO, USA
+Osage catfisheries, Osage Beach (MO), USA
+Ciba-Geigy Corporation, Environmental Health Centre, Farmington
+Novartis Crop Protection Inc., Greensboro, United States
+Syngenta, Jealott’s Hill International Research Station, UK
+Hillandale Farms, Kenansville, Florida, USA
+Jealott’s Hill International Research Station, UK
+L & C Dairy, Avon Park, Florida, UK

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

@@ -1,3 +1,3 @@
 Batches Produced at
 CTL
-for determination of residues
+determination of residues

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

@@ -5290,6 +5290,7 @@ Naeb O.
 Nagai K.
 Nagashima Yoshikazu
 Nagata H.|Tanaka T.
+Nagra, B
 Nagra B.S
 Nagra B.|Kingdom N.
 Nagy K
@@ -5640,6 +5641,7 @@ Pedersen Carol
 Pedersen Carol|DuCharme Darlene
 Pedersen C|Ducharme D.
 Pederson C.A.
+Pederson
 Peffer
 Peffer R
 Peffer R.
@@ -6249,6 +6251,7 @@ Roth M.
 Roth Markus
 Roth RN|Heilman RD
 Roth, M.
+Roth
 Rotroff
 Rouaud, J. L.
 Roussel C
@@ -6848,7 +6851,7 @@ Sokolowshi Andrea
 Sokolowski A
 Sokolowski A.
 Sokolowski Andrea
-Sokolowski,
+Sokolowski
 Sole
 Sole C
 Sole, C.
@@ -7248,6 +7251,7 @@ Thanei, P.
 Tharp BE
 Thede B.
 Theophilidis G
+Thevenaz
 Thevenaz P
 Thevenaz P.
 Thevenaz Ph.
@@ -8193,4 +8197,113 @@ Zoriki Hosomi R.
 Zoriki Hosomi Rosana
 Zuberer D
 Zubrod J
-Zwicker R.E.
+Zwicker R.E.
+Ott
+Argus et al
+Mihail
+Clapp
+Murfitt R
+Yang
+Swarbrick
+Hertl
+Breitwieser
+Seville, A
+Trumper, C.
+Gasso-Brown, D.
+Váliczkó, É
+Váliczkó, É.
+Durando, J.
+Tarcai, Z.
+Pooles, A
+Roig, J.
+Schäfers, C.
+Nixon, W.
+Zok, S.
+Durjava M.K. et al
+Jatzek J
+Salinas E
+Sousa J. V
+Caferella, M.A.
+Salina, E.
+Walraven J
+Kaspers et al
+Schneider et al
+Lowrie C
+Macdougall J
+Roberts S
+Madrid, S.O.
+Maynard, M.S.
+Ray, W.
+Ray, W.J.
+Váliczkó É
+Tarcai Z
+Orovecz B
+Kaspers
+Tarcai Z.
+Gizler, A
+Whitlow
+Johnson et al
+Bryden et al
+Váliczkó
+Höger
+Dorgerloh
+Schäfers C
+Caferella M.A
+Salinas, E
+Zok S
+Tummon OJ
+Eckstein CL
+Francis PD
+Ott M
+Richeux F
+Rajsekhar PV
+Webbley
+O’Hagan
+MacDonald
+Jewkes
+Harris
+Tomlinson
+Tomlinson et al
+Nagy
+Török-Bathó
+Hargitai
+Shearer
+Robertson
+Strepka
+Hackford
+Blunt
+Sieber
+Elcombe
+Haines
+Kang
+Sayers
+Harris J
+Baxter A
+Macdonald M
+PetusArpasy M
+TorokBatho M.
+van Ravenzwaa y B
+Cords S
+Dymarkows ka K.
+Bahig M
+Pekari K
+Bercz J
+Russell L
+Kitchen K
+Miyagawa M.
+Sasaki Y
+Yin D
+Huff J
+Blackburn K
+Carlson G
+Pereira M
+Stoner G
+Zelenak V
+Valiczko E
+Tomlinso n J
+Trumper C
+Senciuc M
+Vance C.
+Przeorska J.
+Melville S.
+Wicksted G.

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

@@ -49,9 +49,9 @@ rule "5: Do not redact in guideline sections"
 		section.redactNot("address", 5, "Section is a guideline section.");
 	end
 
-rule "6: Redact contact information, if applicant is found"
+rule "6: Redact contact information if applicant is found"
 	when
-		eval(section.headlineContainsWord("applicant") || section.getText().contains("Applicant"));
+		eval(section.headlineContainsWord("applicant") || section.getText().contains("Applicant") || section.headlineContainsWord("Primary contact") || section.headlineContainsWord("Alternative contact"));
 	then
 		section.redactLineAfter("Name:", "address", 6, "Applicant information was found");
 		section.redactBetween("Address:", "Contact", "address", 6, "Applicant information was found");
@@ -62,15 +62,21 @@ rule "6: Redact contact information, if applicant is found"
 		section.redactLineAfter("Tel:", "address", 6, "Applicant information was found");
 		section.redactLineAfter("E-mail:", "address", 6, "Applicant information was found");
 		section.redactLineAfter("Email:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("e-mail:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("E-mail address:", "address", 6, "Applicant information was found");
 		section.redactLineAfter("Contact:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Alternative contact:", "address", 6, "Applicant information was found");
 		section.redactLineAfter("Telephone number:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Telephone No:", "address", 6, "Applicant information was found");
 		section.redactLineAfter("Fax number:", "address", 6, "Applicant information was found");
 		section.redactLineAfter("Telephone:", "address", 6, "Applicant information was found");
+		section.redactLineAfter("Company:", "address", 6, "Applicant information was found");
 		section.redactBetween("No:", "Fax", "address", 6, "Applicant information was found");
 		section.redactBetween("Contact:", "Tel.:", "address", 6, "Applicant information was found");
 	end
 
-rule "7: Redact contact information, if Producer is found"
+
+rule "7: Redact contact information if Producer is found"
 	when
 		eval(section.getText().toLowerCase().contains("producer of the plant protection") || section.getText().toLowerCase().contains("producer of the active substance") || section.getText().contains("Manufacturer of the active substance") || section.getText().contains("Manufacturer:") || section.getText().contains("Producer or producers of the active substance"));
 	then
@@ -93,12 +99,11 @@ rule "7: Redact contact information, if Producer is found"
 
 rule "8: Not redacted because Vertebrate Study = N"
     when
-        Section(isNotVertebrateStudy())
+        Section(rowEquals("Vertebrate study Y/N", "N") || rowEquals("Vertebrate study Y/N", "No"))
     then
         section.redactNot("name", 8, "Not redacted because row is not a vertebrate study");
         section.redactNot("address", 8, "Not redacted because row is not a vertebrate study");
         section.highlightCell("Vertebrate study Y/N", 8, "hint_only");
-        section.highlightCell("Verte brate study Y/N", 8, "hint_only");
     end
 
 
@@ -110,12 +115,12 @@ rule "9: Redact if must redact entry is found"
 		section.redact("address", 9, "must_redact entry was found.");
 	end
 
-rule "10: Redact Authors and Addresses in Reference Table, if it is a Vertebrate study"
+
+rule "10: Redact Authors and Addresses in Reference Table if it is a Vertebrate study"
     when
-        Section(isVertebrateStudy())
+        Section(rowEquals("Vertebrate study Y/N", "Y") || rowEquals("Vertebrate study Y/N", "Yes"))
     then
         section.redact("name", 10, "Redacted because row is a vertebrate study");
         section.redact("address", 10, "Redacted because row is a vertebrate study");
         section.highlightCell("Vertebrate study Y/N", 10, "must_redact");
-        section.highlightCell("Verte brate study Y/N", 10, "must_redact");
-    end
+    end