Pull request #21 : Cleaned up dictionaries

Merge in RED/redaction-service from cleanDictionaries to master * commit 'b5412dc9590d15dcae9f87f7fc9b7ea50e4c63ae': Cleaned up dictionaries
Cleaned up dictionaries
2020-08-10 15:46:37 +02:00 · 2020-08-10 15:16:05 +02:00 · 2020-08-07 12:31:26 +02:00 · 2020-08-07 12:28:21 +02:00 · 2020-08-07 12:09:37 +02:00 · 2020-08-05 13:21:14 +02:00
17 changed files with 9525 additions and 3625 deletions
@@ -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();
@@ -1,14 +1,16 @@
 package com.iqser.red.service.redaction.v1.server.redaction.model;

-
 import java.util.ArrayList;
 import java.util.List;

 import lombok.Data;
+import lombok.EqualsAndHashCode;

@Data
+@EqualsAndHashCode(onlyExplicitlyIncluded = true)
 public class Entity {

+    @EqualsAndHashCode.Include
    private final String word;
    private final String type;
    private boolean redaction;
@@ -16,10 +18,17 @@ public class Entity {
    private List<EntityPositionSequence> positionSequences = new ArrayList<>();
    private Integer start;
    private Integer end;
+
+    @EqualsAndHashCode.Include
    private String headline;
    private int matchedRule;

-    public Entity(String word, String type, boolean redaction, String redactionReason, List<EntityPositionSequence> positionSequences, String headline, int matchedRule) {
+    @EqualsAndHashCode.Include
+    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 +36,18 @@ 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;
+
 }
@@ -25,6 +25,8 @@ public class Section {

    private String headline;

+    private int sectionNumber;
+

    public boolean contains(String type) {

@@ -32,6 +34,12 @@ 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 -> {
@@ -58,11 +66,15 @@ public class Section {

    public void redactLineAfter(String start, String asType, int ruleNumber, String reason) {

-        String value = StringUtils.substringBetween(text, start, "\n");
+        String[] values = StringUtils.substringsBetween(text, start, "\n");

-        if (value != null) {
-            Set<Entity> found = findEntity(value.trim(), asType);
-            entities.addAll(found);
+        if (values != null) {
+            for (String value : values) {
+                if (StringUtils.isNotBlank(value)) {
+                    Set<Entity> found = findEntity(value.trim(), asType);
+                    entities.addAll(found);
+                }
+            }
        }

        // TODO No need to iterate
@@ -79,11 +91,15 @@ public class Section {

    public void redactBetween(String start, String stop, String asType, int ruleNumber, String reason) {

-        String value = StringUtils.substringBetween(searchText, start, stop);
+        String[] values = StringUtils.substringsBetween(searchText, start, stop);

-        if (value != null) {
-            Set<Entity> found = findEntity(value.trim(), asType);
-            entities.addAll(found);
+        if (values != null) {
+            for (String value : values) {
+                if (value != null && StringUtils.isNotBlank(value)) {
+                    Set<Entity> found = findEntity(value.trim(), asType);
+                    entities.addAll(found);
+                }
+            }
        }

        // TODO No need to iterate
@@ -109,13 +125,11 @@ 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 +139,7 @@ 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 +151,7 @@ public class Section {
            }
        }
        entities.removeAll(wordsToRemove);
+        return entities;
    }

 }
@@ -73,7 +73,7 @@ public class DictionaryService {
                        .filter(TypeResult::isCaseInsensitive)
                        .map(TypeResult::getType)
                        .collect(Collectors.toList());
-                dictionary = entryColors.keySet().stream().collect(Collectors.toMap(type -> type, s -> convertEntries(s)));
+                dictionary = entryColors.keySet().stream().collect(Collectors.toMap(type -> type, this::convertEntries));
                defaultColor = dictionaryClient.getDefaultColor().getColor();
            }
        } catch (FeignException e) {
@@ -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;
@@ -34,6 +36,7 @@ public class EntityRedactionService {
        droolsExecutionService.updateRules();

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

            SearchableText searchableText = paragraph.getSearchableText();
@@ -51,57 +54,70 @@ public class EntityRedactionService {
                            searchableRow.addAll(textBlock.getSequences());
                        }
                    }
-                    Set<Entity> rowEntities = findEntities(searchableRow, table.getHeadline());
+                    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());

-                    for (Entity entity : analysedRowSection.getEntities()) {
-                        if (dictionaryService.getCaseInsensitiveTypes().contains(entity.getType())) {
-                            entity.setPositionSequences(searchableRow.getSequences(entity.getWord(), true));
-                        } else {
-                            entity.setPositionSequences(searchableRow.getSequences(entity.getWord(), false));
-                        }
-                    }
-                    documentEntities.addAll(analysedRowSection.getEntities());
+                    documentEntities.addAll(clearAndFindPositions(analysedRowSection.getEntities(), searchableRow));
+                    sectionNumber++;
                }
+                sectionNumber++;
            }

-            Set<Entity> entities = findEntities(searchableText, paragraph.getHeadline());
+            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()) {
-                if (dictionaryService.getCaseInsensitiveTypes().contains(entity.getType())) {
-                    entity.setPositionSequences(searchableText.getSequences(entity.getWord(), true));
-                } else {
-                    entity.setPositionSequences(searchableText.getSequences(entity.getWord(), false));
-                }
-            }
-
-            documentEntities.addAll(analysedSection.getEntities());
+            documentEntities.addAll(clearAndFindPositions(analysedSection.getEntities(), searchableText));
+            sectionNumber++;
        }

-        documentEntities.forEach(entity -> {
-            entity.getPositionSequences().forEach(sequence -> {
+        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);
+            }
+
+            for (Map.Entry<Integer, List<EntityPositionSequence>> entry : sequenceOnPage.entrySet()) {
                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()));
-            });
-        });
+                        .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()));
+            }
+        }
+
    }


-    private Set<Entity> findEntities(SearchableText searchableText, String headline) {
+    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();
@@ -110,19 +126,20 @@ public class EntityRedactionService {
        for (Map.Entry<String, Set<String>> entry : dictionaryService.getDictionary().entrySet()) {

            if (dictionaryService.getCaseInsensitiveTypes().contains(entry.getKey())) {
-                found.addAll(find(lowercaseInputString, entry.getValue(), entry.getKey(), headline));
+                found.addAll(find(lowercaseInputString, entry.getValue(), entry.getKey(), headline, sectionNumber));
            } else {
-                found.addAll(find(inputString, entry.getValue(), entry.getKey(), headline));
+                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) {
+    private Set<Entity> find(String inputString, Set<String> values, String type, String headline, int sectionNumber) {

        Set<Entity> found = new HashSet<>();
        for (String value : values) {
@@ -134,7 +151,7 @@ public class EntityRedactionService {

                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));
+                    found.add(new Entity(inputString.substring(startIndex, stopIndex), type, startIndex, stopIndex, headline, sectionNumber));
                }
            } while (startIndex > -1);
        }
@@ -58,6 +58,9 @@ 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 REDACTION_INDICATOR = "redaction_indicator";
+    private static final String HINT_ONLY = "hint_only";
+    private static final String MUST_REDACT = "must_redact";

    @Autowired
    private RedactionController redactionController;
@@ -109,6 +112,9 @@ 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(REDACTION_INDICATOR)).thenReturn(getDictionaryResponse(REDACTION_INDICATOR));
+        when(dictionaryClient.getDictionaryForType(HINT_ONLY)).thenReturn(getDictionaryResponse(HINT_ONLY));
+        when(dictionaryClient.getDictionaryForType(MUST_REDACT)).thenReturn(getDictionaryResponse(MUST_REDACT));
        when(dictionaryClient.getDefaultColor()).thenReturn(new DefaultColor(new float[]{1f, 0.502f, 0f}));
    }

@@ -131,7 +137,22 @@ 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.computeIfAbsent(MUST_REDACT, v -> new ArrayList<>())
+                .addAll(ResourceLoader.load("dictionaries/must_redact.txt")
                        .stream()
                        .map(this::cleanDictionaryEntry)
                        .collect(Collectors.toSet()));
@@ -149,17 +170,26 @@ 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(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});
+        typeColorMap.put(MUST_REDACT, new float[]{1, 0, 0});

        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);
+        hintTypeMap.put(MUST_REDACT, 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);
+        caseInSensitiveMap.put(MUST_REDACT, true);
    }


@@ -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 +0,0 @@
-In Vitro
@@ -0,0 +1,2 @@
+guideline
+unpublished
@@ -0,0 +1,3 @@
+Batches Produced at
+CTL
+for determination of residues
@@ -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
@@ -1,48 +1,63 @@
-Vulpes vulpes
+a. sylvaticus
 african clawed frog
 agalychnis callidryas
+albino rat
+american bullfrog tadpole
+american toad
 amphibian
 amphibians
-American bullfrog tadpole
-american toad
 anad platyrhynchos
-Anas platyrhynchos
+anas platyrhynchos
 anuran
 anurans
+apodemus
+apodemus flavicollis
 apodemus syl vaticus
+apodemus sylvaticus
+arvicola terrestris
 avian
+bank vole
 bird
 birds
 bluegill
 bluegill sunfish
 bobwhite
 bobwhite quail
-bullfrog
-Bufo americanus
 brachydanio rerio
+brown hare
+bufo americanus
+bullfrog
 canary
 carassius carassius
 carp
+catesbeiana
 catfish
 cattle
 cattles
 channel catfish
-Chinook
 chicken
-Colinus virginianus
+chinese hamster
+chinese hamsters
+chinook
+coho salmon
 colinus virginianus
-Common carp
+columba palumbus
+columbidae
+common carp
+common vole
 coturnix japonica
-Coturnix japonica
 cow
 cows
-Crucian carp
+crocidura russula
+crucian carp
 cyprinodon variegatus
 cyprinus carpio
 dog
 dogs
 duck
 ducks
+european brown hare
+european rabbit
 fathead minnow
 fish
 fishes
@@ -56,174 +71,121 @@ galaxias truttaceus
 gasterosteus aculeatus
 goat
 goats
+greater white-toothed shrew
 guinea
 guinea pig
 guinea pigs
-Guppy
+guinea-pigs
+guppy
 hamster
 hamsters
 hen
 hens
-Hyla versicolor
+house mouse
+hyla versicolor
 ictalurus melas
 ictalurus punctatus
 japanese quail
 japonica
+kisutch
+lagomorph
 lebistes reticulatus
 leiostomus xanthurus
 leisostomus xanthurus
 lepomis macrochirus
+lepus europaeus
+limnocharis
+limnodynastes
+limnodynastes tasmaniensis
 livestock
 livestocks
+mallard
 mallard duck
 mammal
+mammalian
 mammals
-Mammalian
+marten
+martes
 mice
+microtus
+microtus agrestis
+microtus arvalis
+microtus subterraneus
 midwestern anurans
+minnow
+minnows
 monkey
 mouse
+mus musculus
+myodes glareolus
 northern bobwhite
+o. cuniculus
 o. mykiss
-Oncorhynchus mykiss
-Oncorhynchus
-O. mykiss
+o. tshawytscha
+oncorhynchus
+oncorhynchus mykiss
+oncorhynchus tshawytscha
+oryctolagus cuniculus
 oryzias melastigma
 oryzias melastigma larvae
 p. promelas
 pagrus major
+palumbus
 pig
 pigeon
 pigeons
+pigs
+pimephales promela
 pimephales promelas
-Pseudacris triseriata
 poecilia reticulata
 poultry
+pseudacris
+pseudacris triseriata
 quail
+r. catesbeiana
+rabbit
+rabbits
 rainbow trout
-Rana limnocharis
 rana
-limnocharis
+rana catesbeiana
+rana limnocharis
 rana pipiens
+rat
+rats
 reptile
 reptiles
 ricefish
 ruminant
 ruminants
+salmo gairdneri
+salmon
+serinus canaria
 sheepshead minnow
 sheepshead minnows
-spea multiplicata
-Salmo gairdneri
-salmon
-spotted march frog
-tadpoles
-treefrog
-toad
-terrestrial vertrebrates
-Limnodynastes tasmaniensis
-trout
-Vulpes vulpes
-wistar
-xenopus laevis
-xenpous leavis
-zebra fish
-zebrafish
-Salmo gairdneri
-minnow
-minnows
-Pimephales promela
-Cyprinodon variegatus
-limnodynastes
-Rana catesbeiana
-R. catesbeiana
-coho salmon
-Oncorhynchus tshawytscha
-O. tshawytscha
-tshawytscha
-catesbeiana
-kisutch
-Pseudacris triseriata
-Pseudacris
-triseriata
-Wood pigeon
-Columba palumbus
-palumbus
-Columbidae
 shrew
 shrews
-bank vole
-common vole
+sorex araneus
+spea multiplicata
+spotted march frog
+tadpoles
+terrestrial vertrebrates
+toad
+treefrog
+triseriata
+trout
+tshawytscha
 vole
 voles
-lagomorph
-Wood mouse
-Apodemus sylvaticus
-A. sylvaticus
-Apodemus flavicollis
-Apodemus
-mus musculus
-Microtus arvalis
-Microtus agrestis
-Microtus
-Arvicola terrestris
-Sorex araneus
-Myodes glareolus
-yellow-necked mouse
-house mouse
-Oryctolagus cuniculus
-marten
-martes
+vulpes vulpes
+white rabbits
 white-toothed shrew
-greater white-toothed shrew
-Lepus europaeus
-brown hare
-European brown hare
-European rabbit
-O. cuniculus
-Crocidura russula
-Chinese Hamster
-Rat
-Rats
-Dog
-Chinese hamsters
-Chinese hamster
-Mouse
-Guinea pig
-Wistar rats
-Rabbit
-mammalian
-Japanese quail
-Microtus subterraneus
-Lepomis macrochirus
-P. promelas
-Cyprinus carpio
-Fish
-Ictalurus punctatus
-Carassius carassius
-Lepomis macrochirus
-Poecilia reticulata
-Lebistes reticulatus
-Lepomis macrochirus
-Leiostomus xanthurus
-Pimephales promelas
-Lepomis macrochirus
-Albino rat
-Hen
-Goat
-Livestock
-Guinea Pigs
-Hamster
+wistar
+wistar rats
+wood mice
 wood mouse
-Rabbits
-Mice
-Rainbow trout
-Canary
-Serinus canaria
-Guinea Pig
-Cow
-Pigs
-Poultry
-Guinea-pigs
-White rabbits
-Birds
-Wood mice
+wood pigeon
+xenopus laevis
+xenpous leavis
+yellow-necked mouse
+zebra fish
+zebrafish
@@ -32,27 +32,70 @@ 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.getText().contains("Producer of the plant protection product"));
+		eval(section.headlineContainsWord("guideline") || section.headlineContainsWord("Guidance"));
 	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");
-	end
+		section.redactNot("name", 5, "Section is a guideline section.");
+		section.redactNot("address", 5, "Section is a guideline section.");
+	end
+
+rule "6: Redact if must redact entry is found"
+	when
+		eval(section.contains("must_redact")==true);
+	then
+		section.redact("name", 6, "must_redact entry was found.");
+		section.redact("address", 6, "must_redact entry was found.");
+	end
+
+
+rule "7: Redact contact information, if applicant is found"
+	when
+		eval(section.headlineContainsWord("applicant") || section.getText().contains("Applicant"));
+	then
+		section.redactLineAfter("Name:", "address", 7, "Applicant information was found");
+		section.redactBetween("Address:", "Contact", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Contact point:",  "address", 7, "Applicant information was found");
+		section.redactLineAfter("Phone:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Fax:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Tel.:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Tel:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("E-mail:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Email:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Contact:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Telephone number:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Fax number:", "address", 7, "Applicant information was found");
+		section.redactLineAfter("Telephone:", "address", 7, "Applicant information was found");
+		section.redactBetween("No:", "Fax", "address", 7, "Applicant information was found");
+		section.redactBetween("Contact:", "Tel.:", "address", 7, "Applicant information was found");
+	end
+
+rule "8: 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
+		section.redactLineAfter("Name:", "address", 8, "Producer was found");
+		section.redactBetween("Address:", "Contact", "address", 8, "Producer was found");
+		section.redactBetween("Contact:", "Phone", "address", 8, "Producer was found");
+		section.redactBetween("Contact:", "Telephone number:", "address", 8, "Producer was found");
+		section.redactBetween("Address:", "Manufacturing", "address", 8, "Producer was found");
+		section.redactLineAfter("Telephone:", "address", 8, "Producer was found");
+		section.redactLineAfter("Phone:", "address", 8, "Producer was found");
+		section.redactLineAfter("Fax:", "address", 8, "Producer was found");
+		section.redactLineAfter("E-mail:", "address", 8, "Producer was found");
+		section.redactLineAfter("Contact:", "address", 8, "Producer was found");
+		section.redactLineAfter("Fax number:", "address", 8, "Producer was found");
+		section.redactLineAfter("Telephone number:", "address", 8, "Producer was found");
+		section.redactLineAfter("Tel:", "address", 8, "Producer was found");
+		section.redactBetween("No:", "Fax", "address", 8, "Producer was found");
+	end
+
Author	SHA1	Message	Date
Cheng Zhu	81048dcc9f	Pull request #21 : Cleaned up dictionaries Merge in RED/redaction-service from cleanDictionaries to master * commit 'b5412dc9590d15dcae9f87f7fc9b7ea50e4c63ae': Cleaned up dictionaries	2020-08-10 15:46:37 +02:00
deiflaender	b5412dc959	Cleaned up dictionaries	2020-08-10 15:16:05 +02:00
Dominique Eiflaender	40e40a01ad	Pull request #19 : Avoid duplicate redaction if type have same entries, made Applicant and Producer rules more specific Merge in RED/redaction-service from ApplicantRule to master * commit '99bac4550a9cade36de313735916687ea26d7b4d': Use @EqualsAndHashCode(onlyExplicitlyIncluded = true) in Entity.java Avoid duplicate redaction if type have same entries, made Applicant and Producer rules more specific	2020-08-07 12:31:26 +02:00
deiflaender	99bac4550a	Use @EqualsAndHashCode(onlyExplicitlyIncluded = true) in Entity.java	2020-08-07 12:28:21 +02:00
deiflaender	7d0b0ed3d0	Avoid duplicate redaction if type have same entries, made Applicant and Producer rules more specific	2020-08-07 12:09:37 +02:00
Cheng Zhu	d465a4ba5b	Pull request #18 : RED-149: Added must_redact dictionary and Rule, Adjusted rules for applicant and producer to work on all documents. Merge in RED/redaction-service from RED-149 to master * commit 'cce8200d433ec89160af3af32f40be57c0b67678': redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/model/Section.java online editiert mit Bitbucket RED-149: Added must_redact dictionary and Rule, Adjusted rules for applicant and producer to work on all documents. Fixed endless loop in rules. Detect multiple occurences in rules	2020-08-05 13:21:14 +02:00
Dominique Eiflaender	cce8200d43	redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/model/Section.java online editiert mit Bitbucket	2020-08-05 13:15:02 +02:00
deiflaender	1f8d371a82	RED-149: Added must_redact dictionary and Rule, Adjusted rules for applicant and producer to work on all documents. Fixed endless loop in rules. Detect multiple occurences in rules	2020-08-05 13:10:31 +02:00
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