Pull request #17 : Duplicates

Merge in RED/redaction-service from duplicates to master * commit '81723ce4022e1e006054eb743f2cc2c0b9faf14f': Use void method type Use EqualsAndHashcode annotation from Lombok Fixed duplicated redaction/RedactionLog entries RED-211, RED-215 Added dictionaries and rules for testing.
Use void method type
2020-08-04 10:59:38 +02:00 · 2020-08-04 10:34:54 +02:00 · 2020-08-04 09:53:58 +02:00 · 2020-07-31 16:25:10 +02:00 · 2020-07-31 16:22:47 +02:00 · 2020-07-28 12:41:35 +02:00
21 changed files with 385 additions and 131 deletions
@@ -39,7 +39,7 @@
        <dependency>
            <groupId>com.iqser.red.service</groupId>
            <artifactId>configuration-service-api-v1</artifactId>
-            <version>1.0.6</version>
+            <version>1.0.12</version>
        </dependency>
        <dependency>
            <groupId>org.drools</groupId>
@@ -4,7 +4,6 @@ import java.util.ArrayList;
 import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
-import java.util.Set;

 import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
@@ -18,7 +17,7 @@ public class Document {

    private List<Page> pages = new ArrayList<>();
    private List<Paragraph> paragraphs = new ArrayList<>();
-    private Map<Integer, Set<Entity>> entities = new HashMap<>();
+    private Map<Integer, List<Entity>> entities = new HashMap<>();
    private FloatFrequencyCounter textHeightCounter = new FloatFrequencyCounter();
    private FloatFrequencyCounter fontSizeCounter= new FloatFrequencyCounter();
    private StringFrequencyCounter fontCounter= new StringFrequencyCounter();
@@ -39,6 +39,12 @@ public class TextPositionSequence implements CharSequence {
        return text.charAt(0);
    }

+    public char charAt(int index, boolean caseInSensitive) {
+        TextPosition textPosition = textPositionAt(index);
+        String text = textPosition.getUnicode();
+        return caseInSensitive ? text.toLowerCase().charAt(0) : text.charAt(0);
+    }
+
    @Override
    public TextPositionSequence subSequence(int start, int end) {
        return new TextPositionSequence(textPositions.subList(start, end), page);
@@ -5,21 +5,38 @@ import java.util.ArrayList;
 import java.util.List;

 import lombok.Data;
+import lombok.EqualsAndHashCode;

@Data
+@EqualsAndHashCode
 public class Entity {

    private final String word;
    private final String type;
+
+    @EqualsAndHashCode.Exclude
    private boolean redaction;
+
+    @EqualsAndHashCode.Exclude
    private String redactionReason;
+
+    @EqualsAndHashCode.Exclude
    private List<EntityPositionSequence> positionSequences = new ArrayList<>();
+
+    @EqualsAndHashCode.Exclude
    private Integer start;
+
+    @EqualsAndHashCode.Exclude
    private Integer end;
+
    private String headline;
+
+    @EqualsAndHashCode.Exclude
    private int matchedRule;

-    public Entity(String word, String type, boolean redaction, String redactionReason, List<EntityPositionSequence> positionSequences, String headline, int matchedRule) {
+    private int sectionNumber;
+
+    public Entity(String word, String type, boolean redaction, String redactionReason, List<EntityPositionSequence> positionSequences, String headline, int matchedRule, int sectionNumber) {
        this.word = word;
        this.type = type;
        this.redaction = redaction;
@@ -27,13 +44,15 @@ public class Entity {
        this.positionSequences = positionSequences;
        this.headline = headline;
        this.matchedRule = matchedRule;
+        this.sectionNumber = sectionNumber;
    }

-    public Entity(String word, String type, Integer start, Integer end, String headline) {
+    public Entity(String word, String type, Integer start, Integer end, String headline, int sectionNumber) {
        this.word = word;
        this.type = type;
        this.start = start;
        this.end = end;
        this.headline = headline;
+        this.sectionNumber = sectionNumber;
    }
 }
@@ -6,15 +6,20 @@ import java.util.UUID;

 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;

+import lombok.AllArgsConstructor;
 import lombok.Data;
+import lombok.EqualsAndHashCode;
 import lombok.RequiredArgsConstructor;

-
@Data
@RequiredArgsConstructor
+@AllArgsConstructor
+@EqualsAndHashCode
 public class EntityPositionSequence {

+    @EqualsAndHashCode.Exclude
    private List<TextPositionSequence> sequences = new ArrayList<>();
    private int pageNumber;
    private final UUID id;
+
 }
@@ -13,45 +13,65 @@ public class SearchableText {

    private List<TextPositionSequence> sequences = new ArrayList<>();

+
    public void add(TextPositionSequence textPositionSequence) {
+
        sequences.add(textPositionSequence);
    }

+
    public void addAll(List<TextPositionSequence> textPositionSequences) {
+
        sequences.addAll(textPositionSequences);
    }


-    public List<EntityPositionSequence> getSequences(String searchString) {
+    public List<EntityPositionSequence> getSequences(String searchString, boolean caseInsensitive) {

-        char[] searchChars = searchString.replaceAll("\\n", " ").toCharArray();
+        String normalizedSearchString;
+        if (caseInsensitive) {
+            normalizedSearchString = searchString.toLowerCase();
+        } else {
+            normalizedSearchString = searchString;
+        }
+
+        char[] searchChars = normalizedSearchString.replaceAll("\\n", " ").toCharArray();
        int counter = 0;

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

-                if(i > 0 && j == 0 && sequences.get(i).charAt(0) == ' ' && sequences.get(i - 1).charAt(sequences.get(i - 1).length() - 1) == ' '
-                || j > 0 && sequences.get(i).charAt(j) == ' ' && sequences.get(i).charAt(j - 1) == ' '){
-                    if(j == sequences.get(i).length() -1  && counter != 0 && !partMatch.getTextPositions().isEmpty()){
+                if (i > 0 && j == 0 && sequences.get(i).charAt(0, caseInsensitive) == ' ' && sequences.get(i - 1)
+                        .charAt(sequences.get(i - 1).length() - 1, caseInsensitive) == ' ' || j > 0 && sequences.get(i)
+                        .charAt(j, caseInsensitive) == ' ' && sequences.get(i).charAt(j - 1, caseInsensitive) == ' ') {
+                    if (j == sequences.get(i).length() - 1 && counter != 0 && !partMatch.getTextPositions().isEmpty()) {
                        crossSequenceParts.add(partMatch);
                    }
                    continue;
                }

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

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

-                    if(counter != 0 || i == 0 && j == 0 || j != 0 && isSeparator(sequences.get(i).charAt(j - 1)) || j == 0 && i != 0 && isSeparator(sequences.get(i - 1).charAt(sequences.get(i - 1).length() -1))
-                            || j == 0 && i != 0 && sequences.get(i - 1).charAt(sequences.get(i - 1).length() -1) != ' ' && sequences.get(i).charAt(j) != ' ') {
+                    if (counter != 0 || i == 0 && j == 0 || j != 0 && isSeparator(sequences.get(i)
+                            .charAt(j - 1, caseInsensitive)) || j == 0 && i != 0 && isSeparator(sequences.get(i - 1)
+                            .charAt(sequences.get(i - 1)
+                                    .length() - 1, caseInsensitive)) || j == 0 && i != 0 && sequences.get(i - 1)
+                            .charAt(sequences.get(i - 1).length() - 1, caseInsensitive) != ' ' && sequences.get(i)
+                            .charAt(j, caseInsensitive) != ' ') {
                        partMatch.add(sequences.get(i).textPositionAt(j));
-                        if (!(j == sequences.get(i).length() -1 && sequences.get(i).charAt(j) == '-' && searchChars[counter] != '-')) {
+                        if (!(j == sequences.get(i).length() - 1 && sequences.get(i)
+                                .charAt(j, caseInsensitive) == '-' && searchChars[counter] != '-')) {
                            counter++;
                        }
                    }
@@ -59,10 +79,13 @@ public class SearchableText {
                    if (counter == searchString.length()) {
                        crossSequenceParts.add(partMatch);

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

@@ -72,14 +95,14 @@ public class SearchableText {
                    }
                } else {
                    counter = 0;
-                    if(!crossSequenceParts.isEmpty()){
+                    if (!crossSequenceParts.isEmpty()) {
                        j--;
                    }
                    crossSequenceParts = new ArrayList<>();
                    partMatch = new TextPositionSequence(sequences.get(i).getPage());
                }

-                if(j == sequences.get(i).length() -1  && counter != 0){
+                if (j == sequences.get(i).length() - 1 && counter != 0) {
                    crossSequenceParts.add(partMatch);
                }
            }
@@ -89,18 +112,18 @@ public class SearchableText {
    }


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

        UUID id = UUID.randomUUID();
        List<EntityPositionSequence> result = new ArrayList<>();
        int currentPage = -1;
        EntityPositionSequence entityPositionSequence = new EntityPositionSequence(id);
-        for (TextPositionSequence textPositionSequence :crossSequenceParts){
-            if(currentPage == -1){
+        for (TextPositionSequence textPositionSequence : crossSequenceParts) {
+            if (currentPage == -1) {
                currentPage = textPositionSequence.getPage();
                entityPositionSequence.setPageNumber(currentPage);
                entityPositionSequence.getSequences().add(textPositionSequence);
-            } else if(currentPage == textPositionSequence.getPage()){
+            } else if (currentPage == textPositionSequence.getPage()) {
                entityPositionSequence.getSequences().add(textPositionSequence);
            } else {
                result.add(entityPositionSequence);
@@ -114,13 +137,14 @@ public class SearchableText {


    private boolean isSeparator(char c) {
+
        return Character.isWhitespace(c) || Pattern.matches("\\p{Punct}", String.valueOf(c)) || c == '\"' || c == '‘' || c == '’';
    }


-
    @Override
    public String toString() {
+
        StringBuilder sb = new StringBuilder();

        TextPositionSequence previous = null;
@@ -137,10 +161,14 @@ public class SearchableText {
            previous = word;
        }

-        return TextNormalizationUtilities.removeHyphenLineBreaks(sb.toString()).replaceAll("\n", " ").replaceAll("  ", " ");
+        return TextNormalizationUtilities.removeHyphenLineBreaks(sb.toString())
+                .replaceAll("\n", " ")
+                .replaceAll("  ", " ");
    }

-    public String getAsStringWithLinebreaks(){
+
+    public String getAsStringWithLinebreaks() {
+
        StringBuilder sb = new StringBuilder();

        TextPositionSequence previous = null;
@@ -25,6 +25,8 @@ public class Section {

    private String headline;

+    private int sectionNumber;
+

    public boolean contains(String type) {

@@ -32,6 +34,11 @@ public class Section {
    }


+    public boolean headlineContainsWord(String word){
+        return StringUtils.containsIgnoreCase(headline, word);
+    }
+
+
    public void redact(String type, int ruleNumber, String reason) {

        entities.forEach(entity -> {
@@ -109,13 +116,11 @@ public class Section {

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

-        removeEntitiesContainedInLarger(found);
-
-        return found;
+        return removeEntitiesContainedInLarger(found);
    }


@@ -125,7 +130,7 @@ public class Section {
    }


-    public void removeEntitiesContainedInLarger(Set<Entity> entities) {
+    public Set<Entity> removeEntitiesContainedInLarger(Set<Entity> entities) {

        List<Entity> wordsToRemove = new ArrayList<>();
        for (Entity word : entities) {
@@ -137,6 +142,7 @@ public class Section {
            }
        }
        entities.removeAll(wordsToRemove);
+        return entities;
    }

 }
@@ -41,6 +41,9 @@ public class DictionaryService {
    @Getter
    private List<String> caseInsensitiveTypes = new ArrayList<>();

+    @Getter
+    private float[] defaultColor;
+

    public void updateDictionary() {

@@ -60,10 +63,6 @@ public class DictionaryService {
                entryColors = typeResponse.getTypes()
                        .stream()
                        .collect(Collectors.toMap(TypeResult::getType, TypeResult::getColor));
-                dictionary = entryColors.keySet()
-                        .stream()
-                        .collect(Collectors.toMap(type -> type, s -> new HashSet<>(dictionaryClient.getDictionaryForType(s)
-                                .getEntries())));
                hintTypes = typeResponse.getTypes()
                        .stream()
                        .filter(TypeResult::isHint)
@@ -74,6 +73,8 @@ public class DictionaryService {
                        .filter(TypeResult::isCaseInsensitive)
                        .map(TypeResult::getType)
                        .collect(Collectors.toList());
+                dictionary = entryColors.keySet().stream().collect(Collectors.toMap(type -> type, this::convertEntries));
+                defaultColor = dictionaryClient.getDefaultColor().getColor();
            }
        } catch (FeignException e) {
            log.warn("Got some unknown feignException", e);
@@ -81,4 +82,16 @@ public class DictionaryService {
        }
    }

+
+    private Set<String> convertEntries(String s) {
+        if (caseInsensitiveTypes.contains(s)) {
+            return dictionaryClient.getDictionaryForType(s)
+                    .getEntries()
+                    .stream()
+                    .map(String::toLowerCase)
+                    .collect(Collectors.toSet());
+        }
+        return new HashSet<>(dictionaryClient.getDictionaryForType(s).getEntries());
+    }
+
 }
@@ -1,6 +1,7 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;

 import java.util.ArrayList;
+import java.util.HashMap;
 import java.util.HashSet;
 import java.util.List;
 import java.util.Map;
@@ -13,6 +14,7 @@ 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.Entity;
+import com.iqser.red.service.redaction.v1.server.redaction.model.EntityPositionSequence;
 import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Section;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
@@ -27,19 +29,20 @@ public class EntityRedactionService {
    private final DictionaryService dictionaryService;
    private final DroolsExecutionService droolsExecutionService;

+
    public void processDocument(Document classifiedDoc) {

        dictionaryService.updateDictionary();
        droolsExecutionService.updateRules();

        Set<Entity> documentEntities = new HashSet<>();
+        int sectionNumber = 1;
        for (Paragraph paragraph : classifiedDoc.getParagraphs()) {

            SearchableText searchableText = paragraph.getSearchableText();

            List<Table> tables = paragraph.getTables();

-            List<SearchableText> searchableRows = new ArrayList<>();
            for (Table table : tables) {
                for (List<Cell> row : table.getRows()) {
                    SearchableText searchableRow = new SearchableText();
@@ -51,92 +54,129 @@ public class EntityRedactionService {
                            searchableRow.addAll(textBlock.getSequences());
                        }
                    }
-                    searchableRows.add(searchableRow);
+                    Set<Entity> rowEntities = findEntities(searchableRow, table.getHeadline(), sectionNumber);
+
+                    Section analysedRowSection = droolsExecutionService.executeRules(Section.builder()
+                            .entities(rowEntities)
+                            .text(searchableRow.getAsStringWithLinebreaks())
+                            .searchText(searchableRow.toString())
+                            .headline(table.getHeadline())
+                            .sectionNumber(sectionNumber)
+                            .build());
+
+                    documentEntities.addAll(clearAndFindPositions(analysedRowSection.getEntities(), searchableRow));
+                    sectionNumber++;
                }
+                sectionNumber++;
            }

-            Set<Entity> entities = findEntities(searchableText, paragraph.getHeadline());
-            Section analysedSection = droolsExecutionService.executeRules(Section
-                    .builder()
+            Set<Entity> entities = findEntities(searchableText, paragraph.getHeadline(), sectionNumber);
+            Section analysedSection = droolsExecutionService.executeRules(Section.builder()
                    .entities(entities)
                    .text(searchableText.getAsStringWithLinebreaks())
                    .searchText(searchableText.toString())
                    .headline(paragraph.getHeadline())
+                    .sectionNumber(sectionNumber)
                    .build());

-            for (Entity entity : analysedSection.getEntities()) {
-                entity.setPositionSequences(searchableText.getSequences(entity.getWord()));
+            documentEntities.addAll(clearAndFindPositions(analysedSection.getEntities(), searchableText));
+            sectionNumber++;
+        }
+
+        for (Entity entity : documentEntities) {
+            Map<Integer, List<EntityPositionSequence>> sequenceOnPage = new HashMap<>();
+            for (EntityPositionSequence entityPositionSequence : entity.getPositionSequences()) {
+                sequenceOnPage.computeIfAbsent(entityPositionSequence.getPageNumber(), (x) -> new ArrayList<>())
+                        .add(entityPositionSequence);
            }

-            documentEntities.addAll(analysedSection.getEntities());
-
-            for (SearchableText searchableRow : searchableRows) {
-                Set<Entity> rowEntities = findEntities(searchableRow, "//TODO TableHeader");
-
-                Section analysedRowSection = droolsExecutionService.executeRules(Section
-                        .builder()
-                        .entities(rowEntities)
-                        .text(searchableRow.getAsStringWithLinebreaks())
-                        .searchText(searchableRow.toString())
-                        .headline("//TODO TableHeader")
-                        .build());
-
-                for (Entity entity : analysedRowSection.getEntities()) {
-                    entity.setPositionSequences(searchableRow.getSequences(entity.getWord()));
-                }
-                documentEntities.addAll(analysedRowSection.getEntities());
+            for (Map.Entry<Integer, List<EntityPositionSequence>> entry : sequenceOnPage.entrySet()) {
+                classifiedDoc.getEntities()
+                        .computeIfAbsent(entry.getKey(), (x) -> new ArrayList<>())
+                        .add(new Entity(entity.getWord(), entity.getType(), entity.isRedaction(), entity.getRedactionReason(), entry
+                                .getValue(), entity.getHeadline(), entity.getMatchedRule(), entity.getSectionNumber()));
            }
        }

-        documentEntities.forEach(entity -> {
-            entity.getPositionSequences().forEach(sequence -> {
-                classifiedDoc.getEntities().computeIfAbsent(sequence.getPageNumber(), (x) -> new HashSet<>()).add(
-                        new Entity(entity.getWord(), entity.getType(), entity.isRedaction(), entity.getRedactionReason(), List.of(sequence), entity.getHeadline(), entity.getMatchedRule())
-                );
-            });
-        });
    }

-    private Set<Entity> findEntities(SearchableText searchableText, String headline) {

-        String normalizedInputString = searchableText.toString();
+    private Set<Entity> clearAndFindPositions(Set<Entity> entities, SearchableText text) {
+
+        removeEntitiesContainedInLarger(entities);
+
+        for (Entity entity : entities) {
+            if (dictionaryService.getCaseInsensitiveTypes().contains(entity.getType())) {
+                entity.setPositionSequences(text.getSequences(entity.getWord(), true));
+            } else {
+                entity.setPositionSequences(text.getSequences(entity.getWord(), false));
+            }
+        }
+
+        return entities;
+    }
+
+
+    private Set<Entity> findEntities(SearchableText searchableText, String headline, int sectionNumber) {
+
+        String inputString = searchableText.toString();
+        String lowercaseInputString = inputString.toLowerCase();

        Set<Entity> found = new HashSet<>();
        for (Map.Entry<String, Set<String>> entry : dictionaryService.getDictionary().entrySet()) {
-            for (String value : entry.getValue()) {
-                int startIndex;
-                int stopIndex = 0;
-                do {
-                    startIndex = normalizedInputString.indexOf(value, stopIndex);
-                    stopIndex = startIndex + value.length();

-                    if (startIndex > -1 &&
-                            (startIndex == 0 || Character.isWhitespace(normalizedInputString.charAt(startIndex - 1)) || isSeparator(normalizedInputString.charAt(startIndex - 1))) &&
-                            (stopIndex == normalizedInputString.length() || isSeparator(normalizedInputString.charAt(stopIndex)))) {
-                        found.add(new Entity(normalizedInputString.substring(startIndex, stopIndex), entry.getKey(), startIndex, stopIndex, headline));
-                    }
-                } while (startIndex > -1);
+            if (dictionaryService.getCaseInsensitiveTypes().contains(entry.getKey())) {
+                found.addAll(find(lowercaseInputString, entry.getValue(), entry.getKey(), headline, sectionNumber));
+            } else {
+                found.addAll(find(inputString, entry.getValue(), entry.getKey(), headline, sectionNumber));
            }
        }

        removeEntitiesContainedInLarger(found);

        return found;
+
    }

+
+    private Set<Entity> find(String inputString, Set<String> values, String type, String headline, int sectionNumber) {
+
+        Set<Entity> found = new HashSet<>();
+        for (String value : values) {
+            int startIndex;
+            int stopIndex = 0;
+            do {
+                startIndex = inputString.indexOf(value, stopIndex);
+                stopIndex = startIndex + value.length();
+
+                if (startIndex > -1 && (startIndex == 0 || Character.isWhitespace(inputString.charAt(startIndex - 1)) || isSeparator(inputString
+                        .charAt(startIndex - 1))) && (stopIndex == inputString.length() || isSeparator(inputString.charAt(stopIndex)))) {
+                    found.add(new Entity(inputString.substring(startIndex, stopIndex), type, startIndex, stopIndex, headline, sectionNumber));
+                }
+            } while (startIndex > -1);
+        }
+        return found;
+    }
+
+
    private boolean isSeparator(char c) {
+
        return Character.isWhitespace(c) || Pattern.matches("\\p{Punct}", String.valueOf(c)) || c == '\"' || c == '‘' || c == '’';
    }

+
    public void removeEntitiesContainedInLarger(Set<Entity> entities) {
+
        List<Entity> wordsToRemove = new ArrayList<>();
        for (Entity word : entities) {
            for (Entity inner : entities) {
-                if (inner.getWord().length() < word.getWord().length() && inner.getStart() >= word.getStart() && inner.getEnd() <= word.getEnd() && word != inner) {
+                if (inner.getWord().length() < word.getWord()
+                        .length() && inner.getStart() >= word.getStart() && inner.getEnd() <= word.getEnd() && word != inner) {
                    wordsToRemove.add(inner);
                }
            }
        }
        entities.removeAll(wordsToRemove);
    }
+
 }
@@ -17,7 +17,6 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
@SuppressWarnings("all")
 public class SectionsBuilderService {

-
    public void buildSections(Document document) {

        List<AbstractTextContainer> chunkWords = new ArrayList<>();
@@ -29,7 +28,8 @@ public class SectionsBuilderService {
        for (Page page : document.getPages()) {
            for (AbstractTextContainer current : page.getTextBlocks()) {

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

@@ -37,7 +37,7 @@ public class SectionsBuilderService {

                if (prev != null && current.getClassification().startsWith("H ") || !document.isHeadlines()) {

-                    Paragraph chunkBlock = buildTextBlock(chunkWords);
+                    Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
                    chunkBlock.setHeadline(lastHeadline);
                    lastHeadline = current.getText();
                    chunkBlockList.add(chunkBlock);
@@ -51,7 +51,7 @@ public class SectionsBuilderService {
            }
        }

-        Paragraph chunkBlock = buildTextBlock(chunkWords);
+        Paragraph chunkBlock = buildTextBlock(chunkWords, lastHeadline);
        if (chunkBlock != null) {
            chunkBlockList.add(chunkBlock);
            chunkBlock.setHeadline(lastHeadline);
@@ -61,7 +61,7 @@ public class SectionsBuilderService {
    }


-    private Paragraph buildTextBlock(List<AbstractTextContainer> wordBlockList) {
+    private Paragraph buildTextBlock(List<AbstractTextContainer> wordBlockList, String lastHeadline) {

        Paragraph paragraph = new Paragraph();
        TextBlock textBlock = null;
@@ -70,17 +70,23 @@ public class SectionsBuilderService {
        boolean splitByTable = false;

        Iterator<AbstractTextContainer> itty = wordBlockList.iterator();
-        boolean alreadyAdded= false;
+        boolean alreadyAdded = false;
+        AbstractTextContainer previous = null;
        while (itty.hasNext()) {
            AbstractTextContainer container = itty.next();

            if (container instanceof Table) {
                splitByTable = true;

+                if (previous != null && previous instanceof TextBlock && previous.getText().startsWith("Table ")) {
+                    ((Table) container).setHeadline(previous.getText());
+                } else {
+                    ((Table) container).setHeadline("Table in: " + lastHeadline);
+                }

                if (textBlock != null && !alreadyAdded) {
                    paragraph.getPageBlocks().add(textBlock);
-                    alreadyAdded =true;
+                    alreadyAdded = true;
                }
                paragraph.getPageBlocks().add(container);
                continue;
@@ -89,24 +95,28 @@ public class SectionsBuilderService {
            TextBlock wordBlock = (TextBlock) container;

            if (textBlock == null) {
-                textBlock = new TextBlock(wordBlock.getMinX(), wordBlock.getMaxX(), wordBlock.getMinY(), wordBlock.getMaxY(), wordBlock.getSequences(), wordBlock.getRotation());
+                textBlock = new TextBlock(wordBlock.getMinX(), wordBlock.getMaxX(), wordBlock.getMinY(), wordBlock.getMaxY(), wordBlock
+                        .getSequences(), wordBlock.getRotation());
                textBlock.setPage(wordBlock.getPage());
            } else if (splitByTable) {
-                textBlock = new TextBlock(wordBlock.getMinX(), wordBlock.getMaxX(), wordBlock.getMinY(), wordBlock.getMaxY(), wordBlock.getSequences(), wordBlock.getRotation());
+                textBlock = new TextBlock(wordBlock.getMinX(), wordBlock.getMaxX(), wordBlock.getMinY(), wordBlock.getMaxY(), wordBlock
+                        .getSequences(), wordBlock.getRotation());
                textBlock.setPage(wordBlock.getPage());
                alreadyAdded = false;
            } else if (pageBefore != -1 && wordBlock.getPage() != pageBefore) {
                textBlock.setPage(pageBefore);
                paragraph.getPageBlocks().add(textBlock);
-                textBlock = new TextBlock(wordBlock.getMinX(), wordBlock.getMaxX(), wordBlock.getMinY(), wordBlock.getMaxY(), wordBlock.getSequences(), wordBlock.getRotation());
+                textBlock = new TextBlock(wordBlock.getMinX(), wordBlock.getMaxX(), wordBlock.getMinY(), wordBlock.getMaxY(), wordBlock
+                        .getSequences(), wordBlock.getRotation());
                textBlock.setPage(wordBlock.getPage());
            } else {
                TextBlock spatialEntity = textBlock.union(wordBlock);
-                textBlock.resize(spatialEntity.getMinX(), spatialEntity.getMinY(),
-                        spatialEntity.getWidth(), spatialEntity.getHeight());
+                textBlock.resize(spatialEntity.getMinX(), spatialEntity.getMinY(), spatialEntity.getWidth(), spatialEntity
+                        .getHeight());
            }
            pageBefore = wordBlock.getPage();
            splitByTable = false;
+            previous = container;
        }

        if (textBlock != null && !alreadyAdded) {
@@ -115,5 +125,4 @@ public class SectionsBuilderService {
        return paragraph;
    }

-
 }
@@ -13,6 +13,7 @@ import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
 import com.iqser.red.service.redaction.v1.server.tableextraction.utils.Utils;

 import lombok.Getter;
+import lombok.Setter;

@SuppressWarnings("all")
 public class Table extends AbstractTextContainer {
@@ -21,6 +22,10 @@ public class Table extends AbstractTextContainer {

    private RectangleSpatialIndex<Cell> si = new RectangleSpatialIndex<>();

+    @Getter
+    @Setter
+    private String headline;
+
    @Getter
    private int rowCount = 0;
    @Getter
@@ -203,7 +203,7 @@ public class AnnotationHighlightService {
        }

        if (!dictionaryService.getEntryColors().containsKey(entity.getType())) {
-            return dictionaryService.getEntryColors().get("default");
+            return dictionaryService.getDefaultColor();
        }

        return dictionaryService.getEntryColors().get(entity.getType());
@@ -35,6 +35,7 @@ import org.springframework.context.annotation.Bean;
 import org.springframework.core.io.ClassPathResource;
 import org.springframework.test.context.junit4.SpringRunner;

+import com.iqser.red.service.configuration.v1.api.model.DefaultColor;
 import com.iqser.red.service.configuration.v1.api.model.DictionaryResponse;
 import com.iqser.red.service.configuration.v1.api.model.RulesResponse;
 import com.iqser.red.service.configuration.v1.api.model.TypeResponse;
@@ -57,7 +58,8 @@ public class RedactionIntegrationTest {
    private static final String ADDRESS_CODE = "address";
    private static final String NAME_CODE = "name";
    private static final String NO_REDACTION_INDICATOR = "no_redaction_indicator";
-    private static final String DEFAULT = "default";
+    private static final String REDACTION_INDICATOR = "redaction_indicator";
+    private static final String HINT_ONLY = "hint_only";

    @Autowired
    private RedactionController redactionController;
@@ -70,6 +72,8 @@ public class RedactionIntegrationTest {

    private final Map<String, List<String>> dictionary = new HashMap<>();
    private final Map<String, float[]> typeColorMap = new HashMap<>();
+    private final Map<String, Boolean> hintTypeMap = new HashMap<>();
+    private final Map<String, Boolean> caseInSensitiveMap = new HashMap<>();

    @TestConfiguration
    public static class RedactionIntegrationTestConfiguration {
@@ -81,7 +85,8 @@ public class RedactionIntegrationTest {

            KieFileSystem kieFileSystem = kieServices.newKieFileSystem();
            InputStream input = new ByteArrayInputStream(RULES.getBytes(StandardCharsets.UTF_8));
-            kieFileSystem.write("src/test/resources/drools/rules.drl", kieServices.getResources().newInputStreamResource(input));
+            kieFileSystem.write("src/test/resources/drools/rules.drl", kieServices.getResources()
+                    .newInputStreamResource(input));
            KieBuilder kieBuilder = kieServices.newKieBuilder(kieFileSystem);
            kieBuilder.buildAll();
            KieModule kieModule = kieBuilder.getKieModule();
@@ -106,7 +111,9 @@ public class RedactionIntegrationTest {
        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE)).thenReturn(getDictionaryResponse(ADDRESS_CODE));
        when(dictionaryClient.getDictionaryForType(NAME_CODE)).thenReturn(getDictionaryResponse(NAME_CODE));
        when(dictionaryClient.getDictionaryForType(NO_REDACTION_INDICATOR)).thenReturn(getDictionaryResponse(NO_REDACTION_INDICATOR));
-        when(dictionaryClient.getDictionaryForType(DEFAULT)).thenReturn(getDictionaryResponse(DEFAULT));
+        when(dictionaryClient.getDictionaryForType(REDACTION_INDICATOR)).thenReturn(getDictionaryResponse(REDACTION_INDICATOR));
+        when(dictionaryClient.getDictionaryForType(HINT_ONLY)).thenReturn(getDictionaryResponse(HINT_ONLY));
+        when(dictionaryClient.getDefaultColor()).thenReturn(new DefaultColor(new float[]{1f, 0.502f, 0f}));
    }


@@ -128,11 +135,20 @@ public class RedactionIntegrationTest {
                        .map(this::cleanDictionaryEntry)
                        .collect(Collectors.toSet()));
        dictionary.computeIfAbsent(NO_REDACTION_INDICATOR, v -> new ArrayList<>())
-                .addAll(ResourceLoader.load("dictionaries/NoRedactionIndicator.txt")
+                .addAll(ResourceLoader.load("dictionaries/no_redaction_indicator.txt")
+                        .stream()
+                        .map(this::cleanDictionaryEntry)
+                        .collect(Collectors.toSet()));
+        dictionary.computeIfAbsent(REDACTION_INDICATOR, v -> new ArrayList<>())
+                .addAll(ResourceLoader.load("dictionaries/redaction_indicator.txt")
+                        .stream()
+                        .map(this::cleanDictionaryEntry)
+                        .collect(Collectors.toSet()));
+        dictionary.computeIfAbsent(HINT_ONLY, v -> new ArrayList<>())
+                .addAll(ResourceLoader.load("dictionaries/hint_only.txt")
                        .stream()
                        .map(this::cleanDictionaryEntry)
                        .collect(Collectors.toSet()));
-        dictionary.put(DEFAULT, new ArrayList<>());
    }


@@ -147,8 +163,23 @@ public class RedactionIntegrationTest {
        typeColorMap.put(VERTEBRATES_CODE, new float[]{0, 1, 0});
        typeColorMap.put(ADDRESS_CODE, new float[]{0, 1, 1});
        typeColorMap.put(NAME_CODE, new float[]{1, 1, 0});
-        typeColorMap.put(NO_REDACTION_INDICATOR, new float[]{1, 0.502f, 0});
-        typeColorMap.put(DEFAULT, new float[]{1, 0.502f, 0});
+        typeColorMap.put(NO_REDACTION_INDICATOR, new float[]{0.8f, 0, 0.8f});
+        typeColorMap.put(REDACTION_INDICATOR, new float[]{1, 0.502f, 0.1f});
+        typeColorMap.put(HINT_ONLY, new float[]{0.8f, 1, 0.8f});
+
+        hintTypeMap.put(VERTEBRATES_CODE, true);
+        hintTypeMap.put(ADDRESS_CODE, false);
+        hintTypeMap.put(NAME_CODE, false);
+        hintTypeMap.put(NO_REDACTION_INDICATOR, true);
+        hintTypeMap.put(REDACTION_INDICATOR, true);
+        hintTypeMap.put(HINT_ONLY, true);
+
+        caseInSensitiveMap.put(VERTEBRATES_CODE, true);
+        caseInSensitiveMap.put(ADDRESS_CODE, false);
+        caseInSensitiveMap.put(NAME_CODE, false);
+        caseInSensitiveMap.put(NO_REDACTION_INDICATOR, true);
+        caseInSensitiveMap.put(REDACTION_INDICATOR, true);
+        caseInSensitiveMap.put(HINT_ONLY, true);
    }


@@ -156,14 +187,25 @@ public class RedactionIntegrationTest {

        return typeColorMap.entrySet()
                .stream()
-                .map(typeColor -> TypeResult.builder().type(typeColor.getKey()).color(typeColor.getValue()).build())
+                .map(typeColor -> TypeResult.builder()
+                        .type(typeColor.getKey())
+                        .color(typeColor.getValue())
+                        .isHint(hintTypeMap.get(typeColor.getKey()))
+                        .isCaseInsensitive(caseInSensitiveMap.get(typeColor.getKey()))
+                        .build())
+
                .collect(Collectors.toList());
    }


    private DictionaryResponse getDictionaryResponse(String type) {

-        return DictionaryResponse.builder().color(typeColorMap.get(type)).entries(dictionary.get(type)).build();
+        return DictionaryResponse.builder()
+                .color(typeColorMap.get(type))
+                .entries(dictionary.get(type))
+                .isHint(hintTypeMap.get(type))
+                .isCaseInsensitive(caseInSensitiveMap.get(type))
+                .build();
    }


@@ -0,0 +1,50 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.util.HashSet;
+import java.util.Set;
+
+import org.junit.Test;
+import org.junit.runner.RunWith;
+import org.kie.api.runtime.KieContainer;
+import org.springframework.beans.factory.annotation.Autowired;
+import org.springframework.boot.test.context.SpringBootTest;
+import org.springframework.boot.test.mock.mockito.MockBean;
+import org.springframework.test.context.junit4.SpringRunner;
+
+import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
+
+@RunWith(SpringRunner.class)
+@SpringBootTest
+public class EntityRedactionServiceTest {
+
+    @MockBean
+    private KieContainer kieContainer;
+
+    @MockBean
+    private DroolsExecutionService droolsExecutionService;
+
+    @MockBean
+    private DictionaryService dictionaryService;
+
+    @Autowired
+    private EntityRedactionService entityRedactionService;
+
+
+    @Test
+    public void testNestedEntitiesRemoval() {
+
+        Set<Entity> entities = new HashSet<>();
+        Entity nested = new Entity("nested", "fake type", 10, 16, "fake headline", 0);
+        Entity nesting = new Entity("nesting nested", "fake type", 2, 16, "fake headline", 0);
+        entities.add(nested);
+        entities.add(nesting);
+        entityRedactionService.removeEntitiesContainedInLarger(entities);
+
+        assertThat(entities.size()).isEqualTo(1);
+        assertThat(entities).contains(nesting);
+
+    }
+
+}
@@ -1,3 +0,0 @@
-In Vitro
-In vitro
-in vitro
@@ -0,0 +1,2 @@
+guideline
+unpublished
@@ -674,6 +674,7 @@ Barnes E
 Blunt H
 Bohnenberger S
 Broich K
+Broich
 Cords SM
 Cowie D
 Davies S
@@ -710,6 +711,8 @@ Randall R
 Manton J
 Nagy K
 Petus–Árpásy M
+Petus–Árpásy
+Petus-Árpásy
 Roth M
 Shearer J
 Sieber M
@@ -2380,6 +2383,7 @@ Bowles
 Dollenmeier
 Myhr
 Chang et al
+Chang
 Beck
 Orton
 Medjakovic
@@ -0,0 +1,3 @@
+published paper
+in vitro
+in-vitro
@@ -0,0 +1,9 @@
+in vivo
+in-vivo
+dermal penetration
+oral toxicity
+oral-toxicity
+acute toxicity
+acute-toxicity
+eco toxicity
+eco-toxicity
@@ -100,15 +100,11 @@ Pseudacris triseriata
 poecilia reticulata
 poultry
 quail
-rabbit
-rabbits
 rainbow trout
 Rana limnocharis
 rana
 limnocharis
 rana pipiens
-rat
-rats
 reptile
 reptiles
 ricefish
@@ -230,4 +226,5 @@ Poultry
 Guinea-pigs
 White rabbits
 Birds
-Wood mice
+Wood mice
+Mallard
@@ -32,27 +32,47 @@ rule "3: Do not redact Names and Addresses if no redaction Indicator is containe
 	end


-rule "4: Redact contact information, if applicant is found"
+rule "4: Redact Names and Addresses if no_redaction_indicator and redaction_indicator is contained"
 	when
-		eval(section.getText().toLowerCase().contains("applicant"));
+		eval(section.contains("vertebrate")==true && section.contains("no_redaction_indicator")==true && section.contains("redaction_indicator")==true);
 	then
-		section.redactLineAfter("Name:", "address", 4, "Redacted because of Rule 4");
-		section.redactBetween("Address:", "Contact", "address", 4, "Redacted because of Rule 4");
-		section.redactLineAfter("Contact point:",  "address", 4, "Redacted because of Rule 4");
-		section.redactLineAfter("Phone:", "address", 4, "Redacted because of Rule 4");
-		section.redactLineAfter("Fax:", "address", 4, "Redacted because of Rule 4");
-		section.redactLineAfter("E-mail:", "address", 4, "Redacted because of Rule 4");
+		section.redact("name", 4, "Vertebrate was found and no_redaction_indicator and redaction_indicator");
+		section.redact("address", 4, "Vertebrate was found and no_redaction_indicator and redaction_indicator");
 	end


-rule "5: Redact contact information, if 'Producer of the plant protection product' is found"
+rule "5: Do not redact in guideline sections"
+	when
+		eval(section.headlineContainsWord("guideline") || section.headlineContainsWord("Guidance"));
+	then
+		section.redactNot("name", 5, "Section is a guideline section.");
+		section.redactNot("address", 5, "Section is a guideline section.");
+	end
+
+
+rule "6: Redact contact information, if applicant is found"
+	when
+		eval(section.getText().toLowerCase().contains("applicant") == true);
+	then
+		section.redactLineAfter("Name:", "address", 6, "contact information was found");
+		section.redactBetween("Address:", "Contact", "address", 6, "contact information was found");
+		section.redactLineAfter("Contact point:",  "address", 6, "contact information was found");
+		section.redactLineAfter("Phone:", "address", 6, "contact information was found");
+		section.redactLineAfter("Fax:", "address", 6, "contact information was found");
+		section.redactLineAfter("E-mail:", "address", 6, "contact information was found");
+		section.redactLineAfter("Contact:", "address", 6, "contact information was found");
+		section.redactLineAfter("Telephone number:", "address", 6, "contact information was found");
+	end
+
+
+rule "7: Redact contact information, if 'Producer of the plant protection product' is found"
 	when
 		eval(section.getText().contains("Producer of the plant protection product"));
 	then
-		section.redactLineAfter("Name:", "address", 5, "xxxx");
-		section.redactBetween("Address:", "Contact", "address", 5, "xxxx");
-		section.redactBetween("Contact:", "Phone", "address", 5, "xxxx");
-		section.redactLineAfter("Phone:", "address", 5, "xxxx");
-		section.redactLineAfter("Fax:", "address", 5, "xxxx");
-		section.redactLineAfter("E-mail:", "address", 5, "xxxx");
+		section.redactLineAfter("Name:", "address", 7, "Producer of the plant protection product was found");
+		section.redactBetween("Address:", "Contact", "address", 7, "Producer of the plant protection product was found");
+		section.redactBetween("Contact:", "Phone", "address", 7, "Producer of the plant protection product was found");
+		section.redactLineAfter("Phone:", "address", 7, "Producer of the plant protection product was found");
+		section.redactLineAfter("Fax:", "address", 7, "Producer of the plant protection product was found");
+		section.redactLineAfter("E-mail:", "address", 7, "Producer of the plant protection product was found");
 	end
Author	SHA1	Message	Date
Thierry Goeckel	70804f111d	Pull request #17 : Duplicates Merge in RED/redaction-service from duplicates to master * commit '81723ce4022e1e006054eb743f2cc2c0b9faf14f': Use void method type Use EqualsAndHashcode annotation from Lombok Fixed duplicated redaction/RedactionLog entries RED-211, RED-215 Added dictionaries and rules for testing.	2020-08-04 10:59:38 +02:00
Thierry Göckel	81723ce402	Use void method type	2020-08-04 10:34:54 +02:00
deiflaender	b1a266d4d4	Use EqualsAndHashcode annotation from Lombok	2020-08-04 09:53:58 +02:00
deiflaender	d2d7f8c50c	Fixed duplicated redaction/RedactionLog entries	2020-07-31 16:25:10 +02:00
deiflaender	e2895a1c7a	RED-211, RED-215 Added dictionaries and rules for testing.	2020-07-31 16:22:47 +02:00
Lena Maldacker	cd07dc6a44	Pull request #16 : Use default color from configuration-service on unknown type Merge in RED/redaction-service from defaultColor to master * commit '872c384dc6da60f421e7aa21f58b574c49414c81': Use default color from configuration-service on unknown type	2020-07-28 12:41:35 +02:00
deiflaender	872c384dc6	Use default color from configuration-service on unknown type	2020-07-28 12:38:59 +02:00
Dominique Eiflaender	2aca35e5a0	Pull request #15 : Let Tables know its headlines Merge in RED/redaction-service from DEV5 to master * commit '88e1c5c58ea44dfab15f086ef58f23df897777d6': Let Tables know its headlines	2020-07-27 15:45:43 +02:00
deiflaender	88e1c5c58e	Let Tables know its headlines	2020-07-27 15:33:25 +02:00
Dominique Eiflaender	b7ee62f44d	Pull request #14 : RED-207: Match caseInsensitive dictionaries caseInSensitive Merge in RED/redaction-service from RED-207 to master * commit '135a715e22e6c2536b268db29161552cfd7a6c1c': Fixed style in EnityRedactionService Fixed wrong naming of caseInsensitive RED-207: Match caseInsensitive dictionaries caseInSensitive	2020-07-27 13:52:40 +02:00
deiflaender	135a715e22	Fixed style in EnityRedactionService	2020-07-27 13:39:31 +02:00
deiflaender	c953f161b2	Fixed wrong naming of caseInsensitive	2020-07-27 13:38:13 +02:00
deiflaender	f0e48087ff	RED-207: Match caseInsensitive dictionaries caseInSensitive	2020-07-27 13:20:00 +02:00
Cheng Zhu	d282680cc8	Pull request #13 : Use Hint in IntegrationTest Merge in RED/redaction-service from DEV3 to master * commit 'b57a4a2db3fb8090004dd5f3babd0d7786312730': Use Hint in IntegrationTest	2020-07-24 11:03:58 +02:00
deiflaender	b57a4a2db3	Use Hint in IntegrationTest	2020-07-24 10:45:04 +02:00
Thierry Goeckel	ca439d821d	Pull request #12 : Bugfix/RED-183 Merge in RED/redaction-service from bugfix/RED-183 to master * commit '7dbe03483b7a8a3c391e9d0791717b9fab5fc5db': RED-183: Fix catching validation errors Fix style.	2020-07-23 14:48:11 +02:00