Marmelator/redaction-service
1
0
Fork 0
Code Issues Pull Requests 14 Actions Packages Projects Releases 622 Wiki Activity

Fixed duplicated redaction/RedactionLog entries

Browse Source
This commit is contained in:
deiflaender 2020-07-31 16:22:30 +02:00
parent e2895a1c7a
commit d2d7f8c50c
6 changed files with 118 additions and 44 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/classification/model/Document.java
View File

@ -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();

31
redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/model/Entity.java
View File

@ -3,6 +3,7 @@ package com.iqser.red.service.redaction.v1.server.redaction.model;
import java.util.ArrayList;
import java.util.List;
import java.util.Objects;
import lombok.Data;
@ -18,8 +19,9 @@ public class Entity {
private Integer end;
private String headline;
private int matchedRule;
private int sectionNumber;
public Entity(String word, String type, boolean redaction, String redactionReason, List<EntityPositionSequence> positionSequences, String headline, int matchedRule) {
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 +29,38 @@ 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;
}
@Override
public boolean equals(Object o) {
if (this == o) {
return true;
}
if (o == null || getClass() != o.getClass()) {
return false;
}
Entity entity = (Entity) o;
return sectionNumber == entity.sectionNumber && Objects.equals(word, entity.word) && Objects.equals(type, entity.type) && Objects
.equals(headline, entity.headline);
}
@Override
public int hashCode() {
return Objects.hash(word, type, headline, sectionNumber);
}
}

25
redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/model/EntityPositionSequence.java
View File

@ -2,19 +2,44 @@ package com.iqser.red.service.redaction.v1.server.redaction.model;
import java.util.ArrayList;
import java.util.List;
import java.util.Objects;
import java.util.UUID;
import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
import lombok.AllArgsConstructor;
import lombok.Data;
import lombok.RequiredArgsConstructor;
@Data
@RequiredArgsConstructor
@AllArgsConstructor
public class EntityPositionSequence {
private List<TextPositionSequence> sequences = new ArrayList<>();
private int pageNumber;
private final UUID id;
@Override
public boolean equals(Object o) {
if (this == o) {
return true;
}
if (o == null || getClass() != o.getClass()) {
return false;
}
EntityPositionSequence that = (EntityPositionSequence) o;
return pageNumber == that.pageNumber && Objects.equals(id, that.id);
}
@Override
public int hashCode() {
return Objects.hash(pageNumber, id);
}
}

16
redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/model/Section.java
View File

@ -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;
}
}

85
redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/service/EntityRedactionService.java
View File

@ -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) {
Set<Entity> cleanEntities = removeEntitiesContainedInLarger(entities);
for (Entity entity : cleanEntities) {
if (dictionaryService.getCaseInsensitiveTypes().contains(entity.getType())) {
entity.setPositionSequences(text.getSequences(entity.getWord(), true));
} else {
entity.setPositionSequences(text.getSequences(entity.getWord(), false));
}
}
return cleanEntities;
}
private Set<Entity> findEntities(SearchableText searchableText, String headline, int sectionNumber) {
String inputString = searchableText.toString();
String lowercaseInputString = inputString.toLowerCase();
@ -110,19 +126,17 @@ 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;
return removeEntitiesContainedInLarger(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 +148,7 @@ 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);
}
@ -148,7 +162,7 @@ public class EntityRedactionService {
}
public void removeEntitiesContainedInLarger(Set<Entity> entities) {
public Set<Entity> removeEntitiesContainedInLarger(Set<Entity> entities) {
List<Entity> wordsToRemove = new ArrayList<>();
for (Entity word : entities) {
@ -160,6 +174,7 @@ public class EntityRedactionService {
}
}
entities.removeAll(wordsToRemove);
return entities;
}
}

2
redaction-service-v1/redaction-service-server-v1/src/test/java/com/iqser/red/service/redaction/v1/server/RedactionIntegrationTest.java
View File

@ -19,6 +19,7 @@ import java.util.stream.Collectors;
import org.apache.commons.io.IOUtils;
import org.junit.Before;
import org.junit.Ignore;
import org.junit.Test;
import org.junit.runner.RunWith;
import org.kie.api.KieServices;
@ -47,6 +48,7 @@ import com.iqser.red.service.redaction.v1.server.controller.RedactionController;
import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
@Ignore
@RunWith(SpringRunner.class)
@SpringBootTest(webEnvironment = DEFINED_PORT)
public class RedactionIntegrationTest {