Merge branch 'master' of ssh://git.iqser.com:2222/red/redaction-service

bamboo-specs/pom.xml

@@ -5,7 +5,7 @@
   <parent>
     <groupId>com.atlassian.bamboo</groupId>
     <artifactId>bamboo-specs-parent</artifactId>
-    <version>7.1.2</version>
+    <version>7.2.2</version>
     <relativePath/>
   </parent>
 

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

@@ -13,6 +13,6 @@ RUN apt-get update \
     wget cabextract xfonts-utils fonts-liberation \
  && rm -rf /var/lib/apt/lists/*
 
-RUN curl http://ftp.br.debian.org/debian/pool/contrib/m/msttcorefonts/ttf-mscorefonts-installer_3.7_all.deb -o /tmp/ttf-mscorefonts-installer_3.7_all.deb \
+RUN curl http://ftp.br.debian.org/debian/pool/contrib/m/msttcorefonts/ttf-mscorefonts-installer_3.8_all.deb -o /tmp/ttf-mscorefonts-installer_3.8_all.deb \
- && dpkg -i /tmp/ttf-mscorefonts-installer_3.7_all.deb \
+ && dpkg -i /tmp/ttf-mscorefonts-installer_3.8_all.deb \
- && rm /tmp/ttf-mscorefonts-installer_3.7_all.deb \
+ && rm /tmp/ttf-mscorefonts-installer_3.8_all.deb \

redaction-service-v1/pom.xml

@@ -5,7 +5,7 @@
     <parent>
         <artifactId>platform-dependency</artifactId>
         <groupId>com.iqser.red</groupId>
-        <version>1.1.2</version>
+        <version>1.1.3</version>
     </parent>
     <modelVersion>4.0.0</modelVersion>
 
@@ -32,7 +32,7 @@
             <dependency>
                 <groupId>com.iqser.red</groupId>
                 <artifactId>platform-commons-dependency</artifactId>
-                <version>1.3.1</version>
+                <version>1.3.6</version>
                 <scope>import</scope>
                 <type>pom</type>
             </dependency>

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

@@ -19,14 +19,8 @@
         </dependency>
         <dependency>
             <groupId>com.iqser.red.service</groupId>
-            <artifactId>configuration-service-api-v1</artifactId>
+            <artifactId>persistence-service-api-v1</artifactId>
-            <version>2.7.0</version>
+            <version>0.4.0</version>
-            <exclusions>
-                <exclusion>
-                    <groupId>com.iqser.red.service</groupId>
-                    <artifactId>file-management-service-api-v1</artifactId>
-                </exclusion>
-            </exclusions>
         </dependency>
     </dependencies>
 </project>

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

@@ -6,6 +6,10 @@ import lombok.Data;
 import lombok.NoArgsConstructor;
 
 import java.time.OffsetDateTime;
+import java.util.ArrayList;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Set;
 
 @Data
 @Builder
@@ -13,12 +17,20 @@ import java.time.OffsetDateTime;
 @AllArgsConstructor
 public class AnalyzeRequest {
 
-    private String projectId;
+    private String dossierId;
     private String fileId;
-    private String ruleSetId;
+    private String dossierTemplateId;
     private boolean reanalyseOnlyIfPossible;
     private ManualRedactions manualRedactions;
     private OffsetDateTime lastProcessed;
 
+    @Builder.Default
+    private Set<Integer> excludedPages = new HashSet<>();
+    @Builder.Default
+    private Set<Integer> sectionsToReanalyse = new HashSet<>();
+
+    @Builder.Default
+    private List<FileAttribute> fileAttributes = new ArrayList<>();
+
 }
 

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

@@ -11,20 +11,20 @@ import lombok.NoArgsConstructor;
 @AllArgsConstructor
 public class AnalyzeResult {
 
-    private String projectId;
+    private String dossierId;
     private String fileId;
     private long duration;
     private int numberOfPages;
-    private boolean hasHints;
-    private boolean hasRequests;
-    private boolean hasRedactions;
-    private boolean hasImages;
     private boolean hasUpdates;
     private long dictionaryVersion;
     private long dossierDictionaryVersion;
     private long rulesVersion;
     private long legalBasisVersion;
 
+    private boolean wasReanalyzed;
+
+    private int analysisVersion;
+
 
 }
 

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

@@ -11,6 +11,7 @@ import lombok.NoArgsConstructor;
 @AllArgsConstructor
 public class AnnotateRequest {
 
-    private String projectId;
+    private String dossierId;
+    private String dossierTemplateId;
     private String fileId;
 }

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

@@ -0,0 +1,15 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@NoArgsConstructor
+@AllArgsConstructor
+public class Argument {
+
+    private String name;
+    private ArgumentType type;
+
+}

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

@@ -0,0 +1,7 @@
+package com.iqser.red.service.redaction.v1.model;
+
+public enum ArgumentType {
+
+    INTEGER, BOOLEAN, STRING, FILE_ATTRIBUTE, REGEX, TYPE, RULE_NUMBER, LEGAL_BASIS, REFERENCE_TYPE
+
+}

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

@@ -0,0 +1,18 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import java.time.OffsetDateTime;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class Change {
+
+    private ChangeType type;
+    private OffsetDateTime dateTime;
+}

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

@@ -1,5 +1,5 @@
 package com.iqser.red.service.redaction.v1.model;
 
 public enum ChangeType {
-    ADDED, REMOVED
+    ADDED, REMOVED, CHANGED
 }

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

@@ -0,0 +1,5 @@
+package com.iqser.red.service.redaction.v1.model;
+
+public enum Engine {
+    DICTIONARY, NER, RULE
+}

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

@@ -0,0 +1,19 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class FileAttribute {
+
+    private String id;
+    private String label;
+    private String placeholder;
+    private String value;
+
+}

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

@@ -5,6 +5,8 @@ import lombok.Builder;
 import lombok.Data;
 import lombok.NoArgsConstructor;
 
+import java.time.OffsetDateTime;
+
 @Data
 @Builder
 @AllArgsConstructor
@@ -16,4 +18,8 @@ public class IdRemoval {
     private Status status;
     private boolean removeFromDictionary;
 
+    private OffsetDateTime requestDate;
+    private OffsetDateTime processedDate;
+    private OffsetDateTime softDeletedTime;
+
 }

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

@@ -5,6 +5,8 @@ import lombok.Builder;
 import lombok.Data;
 import lombok.NoArgsConstructor;
 
+import java.time.OffsetDateTime;
+
 @Data
 @Builder
 @AllArgsConstructor
@@ -16,4 +18,8 @@ public class ManualForceRedact {
     private Status status;
     private String legalBasis;
 
+    private OffsetDateTime requestDate;
+    private OffsetDateTime processedDate;
+    private OffsetDateTime softDeletedTime;
+
 }

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

@@ -0,0 +1,25 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+import java.time.OffsetDateTime;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class ManualImageRecategorization {
+
+    private String id;
+    private String user;
+    private Status status;
+    private String type;
+
+    private OffsetDateTime requestDate;
+    private OffsetDateTime processedDate;
+    private OffsetDateTime softDeletedTime;
+
+}

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

@@ -0,0 +1,25 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+import java.time.OffsetDateTime;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class ManualLegalBasisChange {
+
+    private String id;
+    private String user;
+    private Status status;
+    private String legalBasis;
+
+    private OffsetDateTime requestDate;
+    private OffsetDateTime processedDate;
+    private OffsetDateTime softDeletedTime;
+
+}

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

@@ -5,6 +5,7 @@ import lombok.Builder;
 import lombok.Data;
 import lombok.NoArgsConstructor;
 
+import java.time.OffsetDateTime;
 import java.util.ArrayList;
 import java.util.List;
 
@@ -22,11 +23,12 @@ public class ManualRedactionEntry {
     private String legalBasis;
     private List<Rectangle> positions = new ArrayList<>();
     private Status status;
+
     private boolean addToDictionary;
-
-    private String section;
-    private int sectionNumber;
-
     private boolean addToDossierDictionary;
 
+    private OffsetDateTime requestDate;
+    private OffsetDateTime processedDate;
+    private OffsetDateTime softDeletedTime;
+
 }

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

@@ -1,5 +1,5 @@
 package com.iqser.red.service.redaction.v1.model;
 
 public enum ManualRedactionType {
-    ADD, REMOVE, FORCE_REDACT
+    ADD, REMOVE, FORCE_REDACT, RECATEGORIZE, LEGAL_BASIS_CHANGE
 }

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

@@ -26,6 +26,12 @@ public class ManualRedactions {
     @Builder.Default
     private Set<ManualRedactionEntry> entriesToAdd = new HashSet<>();
 
+    @Builder.Default
+    private Set<ManualImageRecategorization> imageRecategorizations = new HashSet<>();
+
+    @Builder.Default
+    private Set<ManualLegalBasisChange> manualLegalBasisChanges = new HashSet<>();
+
     @Builder.Default
     private Map<String, List<Comment>> comments = new HashMap<>();
 

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

@@ -1,47 +0,0 @@
-package com.iqser.red.service.redaction.v1.model;
-
-import lombok.AllArgsConstructor;
-import lombok.Builder;
-import lombok.Data;
-import lombok.NoArgsConstructor;
-
-import java.util.ArrayList;
-import java.util.List;
-
-@Data
-@Builder
-@NoArgsConstructor
-@AllArgsConstructor
-public class RedactionChangeLogEntry {
-
-    private String id;
-    private String type;
-    private String value;
-    private String reason;
-    private int matchedRule;
-    private String legalBasis;
-    private boolean redacted;
-    private boolean isHint;
-    private boolean isRecommendation;
-    private String section;
-    private float[] color;
-
-    @Builder.Default
-    private List<Rectangle> positions = new ArrayList<>();
-    private int sectionNumber;
-    private boolean manual;
-    private Status status;
-    private ManualRedactionType manualRedactionType;
-    private boolean isDictionaryEntry;
-
-    private String textBefore;
-    private String textAfter;
-
-    @Builder.Default
-    private List<Comment> comments = new ArrayList<>();
-
-    private ChangeType changeType;
-
-    private boolean isDossierDictionaryEntry;
-
-}

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

@@ -1,27 +1,29 @@
 package com.iqser.red.service.redaction.v1.model;
 
-import com.iqser.red.service.configuration.v1.api.model.LegalBasisMapping;
 import lombok.AllArgsConstructor;
 import lombok.Data;
 
 import java.util.List;
 
+import com.iqser.red.service.persistence.service.v1.api.model.data.configuration.LegalBasis;
+
 @Data
 @AllArgsConstructor
 public class RedactionLog {
 
-    private List<RedactionLogEntry> redactionLogEntry;
-    private List<LegalBasisMapping> legalBasis;
 
-    private String ruleSetId;
+    /**
+     * Version 0 Redaction Logs have manual redactions merged inside them
+     * Version 1 Redaction Logs only contain system ( rule/dictionary ) redactions. Manual Redactions are merged in at runtime.
+     */
+    private long analysisVersion;
+
+    private List<RedactionLogEntry> redactionLogEntry;
+    private List<LegalBasis> legalBasis;
 
     private long dictionaryVersion = -1;
-    private long rulesVersion = -1;
     private long dossierDictionaryVersion = -1;
+    private long rulesVersion = -1;
     private long legalBasisVersion = -1;
 
-
-
-
-
 }

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

@@ -0,0 +1,17 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class RedactionLogChanges {
+
+    private RedactionLog redactionLog;
+    private boolean hasChanges;
+
+}

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

@@ -7,13 +7,15 @@ import lombok.EqualsAndHashCode;
 import lombok.NoArgsConstructor;
 
 import java.util.ArrayList;
+import java.util.HashSet;
 import java.util.List;
+import java.util.Set;
 
 @Data
 @Builder
 @NoArgsConstructor
 @AllArgsConstructor
-@EqualsAndHashCode(of = "id")
+@EqualsAndHashCode
 public class RedactionLogEntry {
 
     private String id;
@@ -34,6 +36,7 @@ public class RedactionLogEntry {
     private boolean manual;
     private Status status;
     private ManualRedactionType manualRedactionType;
+    private String manualRedactionUserId;
     private boolean isDictionaryEntry;
 
     private String textBefore;
@@ -46,7 +49,23 @@ public class RedactionLogEntry {
     private int endOffset;
 
     private boolean isImage;
+    private boolean imageHasTransparency;
 
     private boolean isDossierDictionaryEntry;
 
+    private boolean excluded;
+
+    private String recategorizationType;
+    private String legalBasisChangeValue;
+
+    @EqualsAndHashCode.Exclude
+    @Builder.Default
+    private List<Change> changes = new ArrayList<>();
+
+    private Set<Engine> engines= new HashSet<>();
+
+    private Set<String> reference = new HashSet<>();
+
+
+
 }

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

@@ -1,5 +1,8 @@
 package com.iqser.red.service.redaction.v1.model;
 
+import java.util.HashSet;
+import java.util.Set;
+
 import lombok.AllArgsConstructor;
 import lombok.Builder;
 import lombok.Data;
@@ -11,8 +14,10 @@ import lombok.NoArgsConstructor;
 @AllArgsConstructor
 public class RedactionRequest {
 
-    private String projectId;
+    private String dossierId;
     private String fileId;
-    private String ruleSetId;
+    private String dossierTemplateId;
     private ManualRedactions manualRedactions;
+    @Builder.Default
+    private Set<Integer> excludedPages = new HashSet<>();
 }

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

@@ -0,0 +1,14 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.Data;
+
+import java.util.ArrayList;
+import java.util.List;
+
+@Data
+public class RuleBuilderModel {
+
+    private List<RuleElement> whenClauses = new ArrayList<>();
+    private List<RuleElement> thenConditions = new ArrayList<>();
+
+}

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

@@ -8,15 +8,11 @@ import java.util.ArrayList;
 import java.util.List;
 
 @Data
-@AllArgsConstructor
 @NoArgsConstructor
-public class RedactionChangeLog {
+@AllArgsConstructor
+public class RuleElement {
 
-    private List<RedactionChangeLogEntry> redactionLogEntry = new ArrayList<>();
+    private String conditionName;
-
+    private List<Argument> arguments = new ArrayList<>();
-    private long dictionaryVersion = -1;
-    private long rulesVersion = -1;
-
-    private String ruleSetId;
 
 }

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

@@ -30,4 +30,5 @@ public class SectionRectangle {
     private int numberOfParts;
 
     private List<CellRectangle> tableCells;
+
 }

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

@@ -0,0 +1,18 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class StructureAnalyzeRequest {
+
+    private String dossierId;
+    private String fileId;
+
+}
+

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/resources/RedactionResource.java

@@ -8,12 +8,6 @@ import org.springframework.web.bind.annotation.RequestBody;
 
 public interface RedactionResource {
 
-    String SERVICE_NAME = "redaction-service-v1";
-
-    String RULE_SET_PARAMETER_NAME = "ruleSetId";
-    String RULE_SET_PATH_VARIABLE = "/{" + RULE_SET_PARAMETER_NAME + "}";
-
-
     @PostMapping(value = "/annotate", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
     AnnotateResponse annotate(@RequestBody AnnotateRequest annotateRequest);
 
@@ -26,10 +20,10 @@ public interface RedactionResource {
     @PostMapping(value = "/debug/htmlTables", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
     RedactionResult htmlTables(@RequestBody RedactionRequest redactionRequest);
 
-    @PostMapping(value = "/rules/update" + RULE_SET_PATH_VARIABLE, consumes = MediaType.APPLICATION_JSON_VALUE)
-    void updateRules(@PathVariable(RULE_SET_PARAMETER_NAME) String ruleSetId);
-
     @PostMapping(value = "/rules/test", consumes = MediaType.APPLICATION_JSON_VALUE)
     void testRules(@RequestBody String rules);
 
+    @PostMapping(value = "/redaction-log/preview", consumes = MediaType.APPLICATION_JSON_VALUE)
+    RedactionLog getRedactionLog(@RequestBody RedactionRequest redactionRequest);
+
 }

redaction-service-v1/redaction-service-api-v1/src/main/java/com/iqser/red/service/redaction/v1/resources/RuleBuilderResource.java

@@ -0,0 +1,12 @@
+package com.iqser.red.service.redaction.v1.resources;
+
+import com.iqser.red.service.redaction.v1.model.RuleBuilderModel;
+import org.springframework.http.MediaType;
+import org.springframework.web.bind.annotation.PostMapping;
+
+public interface RuleBuilderResource {
+
+    @PostMapping(value = "/rule-builder-model", produces = MediaType.APPLICATION_JSON_VALUE)
+    RuleBuilderModel getRuleBuilderModel();
+
+}

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

@@ -21,21 +21,6 @@
             <artifactId>redaction-service-api-v1</artifactId>
             <version>${project.version}</version>
         </dependency>
-        <dependency>
-            <groupId>com.iqser.red.service</groupId>
-            <artifactId>file-management-service-api-v1</artifactId>
-            <version>2.7.4</version>
-            <exclusions>
-                <exclusion>
-                    <groupId>com.iqser.red.service</groupId>
-                    <artifactId>redaction-service-api-v1</artifactId>
-                </exclusion>
-                <exclusion>
-                    <groupId>com.iqser.red.service</groupId>
-                    <artifactId>configuration-service-api-v1</artifactId>
-                </exclusion>
-            </exclusions>
-        </dependency>
         <dependency>
             <groupId>org.drools</groupId>
             <artifactId>drools-core</artifactId>

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

@@ -1,19 +1,14 @@
 package com.iqser.red.service.redaction.v1.server.classification.model;
 
-import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import java.util.ArrayList;
+import java.util.List;
+
 import com.iqser.red.service.redaction.v1.model.SectionGrid;
 import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryVersion;
-import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
+
-import com.iqser.red.service.redaction.v1.server.redaction.model.Image;
 import lombok.Data;
 import lombok.NoArgsConstructor;
 
-import java.util.ArrayList;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
-import java.util.Set;
-
 @Data
 @NoArgsConstructor
 public class Document {
@@ -23,20 +18,14 @@ public class Document {
     private List<Header> headers = new ArrayList<>();
     private List<Footer> footers = new ArrayList<>();
     private List<UnclassifiedText> unclassifiedTexts = new ArrayList<>();
-    private Map<Integer, List<Entity>> entities = new HashMap<>();
     private FloatFrequencyCounter textHeightCounter = new FloatFrequencyCounter();
     private FloatFrequencyCounter fontSizeCounter = new FloatFrequencyCounter();
     private StringFrequencyCounter fontCounter = new StringFrequencyCounter();
     private StringFrequencyCounter fontStyleCounter = new StringFrequencyCounter();
     private boolean headlines;
 
-    private List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
     private SectionGrid sectionGrid = new SectionGrid();
     private DictionaryVersion dictionaryVersion;
     private long rulesVersion;
 
-    private List<SectionText> sectionText = new ArrayList<>();
-
-    private Map<Integer, Set<Image>> images = new HashMap<>();
-
 }

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

@@ -0,0 +1,6 @@
+package com.iqser.red.service.redaction.v1.server.classification.model;
+
+public enum Orientation {
+
+    NONE, LEFT, RIGHT
+}

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

@@ -32,6 +32,7 @@ public class TextBlock extends AbstractTextContainer {
 
     private String classification;
 
+
     public TextBlock(float minX, float maxX, float minY, float maxY, List<TextPositionSequence> sequences, int rotation) {
         this.minX = minX;
         this.maxX = maxX;

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

@@ -1,6 +1,7 @@
 package com.iqser.red.service.redaction.v1.server.classification.service;
 
 import com.iqser.red.service.redaction.v1.server.classification.model.FloatFrequencyCounter;
+import com.iqser.red.service.redaction.v1.server.classification.model.Orientation;
 import com.iqser.red.service.redaction.v1.server.classification.model.Page;
 import com.iqser.red.service.redaction.v1.server.classification.model.StringFrequencyCounter;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
@@ -11,16 +12,21 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Rectangle;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Ruling;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
+
 import org.springframework.stereotype.Service;
 
 import java.util.ArrayList;
+import java.util.Iterator;
 import java.util.List;
 
 @Service
 @SuppressWarnings("all")
 public class BlockificationService {
 
-    public Page blockify(List<TextPositionSequence> textPositions, List<Ruling> horizontalRulingLines, List<Ruling> verticalRulingLines) {
+    static final float THRESHOLD = 1f;
+
+    public Page blockify(List<TextPositionSequence> textPositions, List<Ruling> horizontalRulingLines,
+                         List<Ruling> verticalRulingLines) {
 
         List<TextPositionSequence> chunkWords = new ArrayList<>();
         List<AbstractTextContainer> chunkBlockList1 = new ArrayList<>();
@@ -28,21 +34,46 @@ public class BlockificationService {
         float minX = 1000, maxX = 0, minY = 1000, maxY = 0;
         TextPositionSequence prev = null;
 
+        boolean wasSplitted = false;
+        Float splitX1 = null;
         for (TextPositionSequence word : textPositions) {
 
             boolean lineSeparation = minY - word.getY2() > word.getHeight() * 1.25;
             boolean startFromTop = word.getY1() > maxY + word.getHeight();
+            boolean splitByX = prev != null && maxX + 50 < word.getX1() && prev.getY1() == word.getY1();
+            boolean newLineAfterSplit = prev != null && word.getY1() != prev.getY1() && wasSplitted && splitX1 != word.getX1();
+            boolean splittedByRuling = word.getRotation() == 0 && isSplittedByRuling(maxX, minY, word.getX1(), word.getY1(), verticalRulingLines) || word
+                    .getRotation() == 0 && isSplittedByRuling(minX, minY, word.getX1(), word.getY2(), horizontalRulingLines) || word
+                    .getRotation() == 90 && isSplittedByRuling(maxX, minY, word.getX1(), word.getY1(), horizontalRulingLines) || word
+                    .getRotation() == 90 && isSplittedByRuling(minX, minY, word.getX1(), word.getY2(), verticalRulingLines);
 
-            if (prev != null && (lineSeparation || startFromTop || word.getRotation() == 0 && isSplittedByRuling(maxX, minY, word
+            if (prev != null && (lineSeparation || startFromTop || splitByX || newLineAfterSplit || splittedByRuling)) {
-                    .getX1(), word.getY1(), verticalRulingLines) || word.getRotation() == 0 && isSplittedByRuling(minX, minY, word
+
-                    .getX1(), word.getY2(), horizontalRulingLines) || word.getRotation() == 90 && isSplittedByRuling(maxX, minY, word
+                Orientation prevOrientation = null;
-                    .getX1(), word.getY1(), horizontalRulingLines) || word.getRotation() == 90 && isSplittedByRuling(minX, minY, word
+                if(!chunkBlockList1.isEmpty()) {
-                    .getX1(), word.getY2(), verticalRulingLines))) {
+                    prevOrientation = chunkBlockList1.get(chunkBlockList1.size() - 1).getOrientation();
+                }
 
                 TextBlock cb1 = buildTextBlock(chunkWords);
                 chunkBlockList1.add(cb1);
                 chunkWords = new ArrayList<>();
 
+                if (splitByX && !splittedByRuling) {
+                    wasSplitted = true;
+                    cb1.setOrientation(Orientation.LEFT);
+                    splitX1 = word.getX1();
+                } else
+
+                if (newLineAfterSplit && !splittedByRuling) {
+                    wasSplitted = false;
+                    cb1.setOrientation(Orientation.RIGHT);
+                    splitX1 = null;
+                } else
+
+                if(prevOrientation != null && prevOrientation.equals(Orientation.RIGHT) && (lineSeparation || !startFromTop || !splitByX || !newLineAfterSplit || !splittedByRuling)){
+                    cb1.setOrientation(Orientation.LEFT);
+                }
+
                 minX = 1000;
                 maxX = 0;
                 minY = 1000;
@@ -72,9 +103,62 @@ public class BlockificationService {
             chunkBlockList1.add(cb1);
         }
 
+        Iterator<AbstractTextContainer> itty = chunkBlockList1.iterator();
+
+        TextBlock previousLeft = null;
+        TextBlock previousRight = null;
+        while (itty.hasNext()) {
+            TextBlock block = (TextBlock) itty.next();
+
+            if(previousLeft != null && block.getOrientation().equals(Orientation.LEFT)){
+                if (previousLeft.getMinY() > block.getMinY() && block.getMaxY() + block.getMostPopularWordHeight() > previousLeft.getMinY()){
+                    previousLeft.add(block);
+                    itty.remove();
+                    continue;
+                }
+            }
+
+            if(previousRight != null && block.getOrientation().equals(Orientation.RIGHT)){
+                if (previousRight.getMinY() > block.getMinY() && block.getMaxY() + block.getMostPopularWordHeight() > previousRight.getMinY()){
+                    previousRight.add(block);
+                    itty.remove();
+                    continue;
+                }
+            }
+
+            if (block.getOrientation().equals(Orientation.LEFT)) {
+                previousLeft = block;
+            } else if (block.getOrientation().equals(Orientation.RIGHT)) {
+                previousRight = block;
+            }
+        }
+
+
+        itty = chunkBlockList1.iterator();
+        TextBlock previous = null;
+        while (itty.hasNext()) {
+            TextBlock block = (TextBlock) itty.next();
+
+            if(previous != null && previous.getOrientation().equals(Orientation.LEFT) && block.getOrientation().equals(Orientation.LEFT) && equalsWithThreshold(block.getMaxY(), previous
+                    .getMaxY())||
+                    previous != null && previous.getOrientation().equals(Orientation.LEFT) && block.getOrientation().equals(Orientation.RIGHT) && equalsWithThreshold(block.getMaxY(), previous
+                    .getMaxY())){
+                previous.add(block);
+                itty.remove();
+                continue;
+            }
+
+            previous = block;
+        }
+
+
         return new Page(chunkBlockList1);
     }
 
+    private boolean equalsWithThreshold(float f1, float f2){
+        return Math.abs(f1 - f2) < THRESHOLD;
+    }
+
 
     private TextBlock buildTextBlock(List<TextPositionSequence> wordBlockList) {
 
@@ -117,7 +201,8 @@ public class BlockificationService {
     }
 
 
-    private boolean isSplittedByRuling(float previousX2, float previousY1, float currentX1, float currentY1, List<Ruling> rulingLines) {
+    private boolean isSplittedByRuling(float previousX2, float previousY1, float currentX1, float currentY1,
+                                       List<Ruling> rulingLines) {
 
         for (Ruling ruling : rulingLines) {
             if (ruling.intersectsLine(previousX2, previousY1, currentX1, currentY1)) {
@@ -128,7 +213,8 @@ public class BlockificationService {
     }
 
 
-    public Rectangle calculateBodyTextFrame(List<Page> pages, FloatFrequencyCounter documentFontSizeCounter, boolean landscape) {
+    public Rectangle calculateBodyTextFrame(List<Page> pages, FloatFrequencyCounter documentFontSizeCounter,
+                                            boolean landscape) {
 
         float minX = 10000;
         float maxX = -100;

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

@@ -1,8 +1,9 @@
 package com.iqser.red.service.redaction.v1.server.client;
 
-import com.iqser.red.service.configuration.v1.api.resource.DictionaryResource;
 import org.springframework.cloud.openfeign.FeignClient;
 
-@FeignClient(name = "DictionaryResource", url = "${configuration-service.url}")
+import com.iqser.red.service.persistence.service.v1.api.resources.DictionaryResource;
+
+@FeignClient(name = "DictionaryResource", url = "${persistence-service.url}")
 public interface DictionaryClient extends DictionaryResource {
 }

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

@@ -0,0 +1,19 @@
+package com.iqser.red.service.redaction.v1.server.client;
+
+import java.util.List;
+import java.util.Map;
+
+import org.springframework.cloud.openfeign.FeignClient;
+import org.springframework.http.MediaType;
+import org.springframework.web.bind.annotation.PostMapping;
+
+import com.iqser.red.service.redaction.v1.server.client.model.EntityRecogintionEntity;
+import com.iqser.red.service.redaction.v1.server.client.model.EntityRecognitionRequest;
+import com.iqser.red.service.redaction.v1.server.client.model.NerEntities;
+
+@FeignClient(name = "EntityRecognitionClient", url = "${entity-recognition-service.url}")
+public interface EntityRecognitionClient {
+
+    @PostMapping(value = "/find_authors", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
+    NerEntities findAuthors(EntityRecognitionRequest entityRecognitionRequest);
+}

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

@@ -1,9 +1,10 @@
 package com.iqser.red.service.redaction.v1.server.client;
 
 
-import com.iqser.red.service.file.management.v1.api.resources.FileStatusProcessingUpdateResource;
 import org.springframework.cloud.openfeign.FeignClient;
 
-@FeignClient(name = "FileStatusProcessingUpdateResource", url = "${file-management-service.url}")
+import com.iqser.red.service.persistence.service.v1.api.resources.FileStatusProcessingUpdateResource;
+
+@FeignClient(name = "FileStatusProcessingUpdateResource", url = "${persistence-service.url}")
 public interface FileStatusProcessingUpdateClient extends FileStatusProcessingUpdateResource {
 }

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

@@ -1,8 +1,9 @@
 package com.iqser.red.service.redaction.v1.server.client;
 
-import com.iqser.red.service.configuration.v1.api.resource.LegalBasisMappingResource;
 import org.springframework.cloud.openfeign.FeignClient;
 
-@FeignClient(name = "LegalBasisMappingResource", url = "${configuration-service.url}")
+import com.iqser.red.service.persistence.service.v1.api.resources.LegalBasisMappingResource;
+
+@FeignClient(name = "LegalBasisMappingResource", url = "${persistence-service.url}")
 public interface LegalBasisClient extends LegalBasisMappingResource {
 }

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

@@ -1,8 +1,9 @@
 package com.iqser.red.service.redaction.v1.server.client;
 
-import com.iqser.red.service.configuration.v1.api.resource.RulesResource;
 import org.springframework.cloud.openfeign.FeignClient;
 
-@FeignClient(name = "RulesResource", url = "${configuration-service.url}")
+import com.iqser.red.service.persistence.service.v1.api.resources.RulesResource;
+
+@FeignClient(name = "RulesResource", url = "${persistence-service.url}")
 public interface RulesClient extends RulesResource {
 }

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

@@ -0,0 +1,19 @@
+package com.iqser.red.service.redaction.v1.server.client.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class EntityRecogintionEntity {
+
+      private String value;
+      private int startOffset;
+      private int endOffset;
+      private String type;
+
+}

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

@@ -0,0 +1,18 @@
+package com.iqser.red.service.redaction.v1.server.client.model;
+
+import java.util.List;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class EntityRecognitionRequest {
+
+   private List<EntityRecognitionSection> data;
+
+}

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

@@ -0,0 +1,20 @@
+package com.iqser.red.service.redaction.v1.server.client.model;
+
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class EntityRecognitionResult {
+
+    @Builder.Default
+    private Map<Integer, List<EntityRecogintionEntity>> entities = new HashMap<>();
+}

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

@@ -0,0 +1,16 @@
+package com.iqser.red.service.redaction.v1.server.client.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class EntityRecognitionSection {
+
+    private int sectionNumber;
+    private String text;
+}

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

@@ -0,0 +1,21 @@
+package com.iqser.red.service.redaction.v1.server.client.model;
+
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@AllArgsConstructor
+@NoArgsConstructor
+public class NerEntities {
+
+   @Builder.Default
+   private Map<Integer, List<EntityRecogintionEntity>> result = new HashMap<>();
+
+}

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

@@ -1,10 +1,7 @@
 package com.iqser.red.service.redaction.v1.server.controller;
 
-import com.iqser.red.service.file.management.v1.api.model.FileType;
+import com.iqser.red.service.persistence.service.v1.api.model.FileType;
-import com.iqser.red.service.redaction.v1.model.AnnotateRequest;
+import com.iqser.red.service.redaction.v1.model.*;
-import com.iqser.red.service.redaction.v1.model.AnnotateResponse;
-import com.iqser.red.service.redaction.v1.model.RedactionRequest;
-import com.iqser.red.service.redaction.v1.model.RedactionResult;
 import com.iqser.red.service.redaction.v1.resources.RedactionResource;
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
 import com.iqser.red.service.redaction.v1.server.classification.model.Page;
@@ -12,6 +9,7 @@ import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
 import com.iqser.red.service.redaction.v1.server.redaction.service.AnnotationService;
 import com.iqser.red.service.redaction.v1.server.redaction.service.DictionaryService;
 import com.iqser.red.service.redaction.v1.server.redaction.service.DroolsExecutionService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.RedactionLogMergeService;
 import com.iqser.red.service.redaction.v1.server.segmentation.PdfSegmentationService;
 import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
@@ -21,7 +19,6 @@ import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
 import org.apache.pdfbox.io.MemoryUsageSetting;
 import org.apache.pdfbox.pdmodel.PDDocument;
-import org.springframework.web.bind.annotation.PathVariable;
 import org.springframework.web.bind.annotation.RequestBody;
 import org.springframework.web.bind.annotation.RestController;
 
@@ -39,18 +36,18 @@ public class RedactionController implements RedactionResource {
     private final AnnotationService annotationService;
     private final PdfSegmentationService pdfSegmentationService;
     private final RedactionStorageService redactionStorageService;
-
+    private final RedactionLogMergeService redactionLogMergeService;
 
     public AnnotateResponse annotate(@RequestBody AnnotateRequest annotateRequest) {
 
-        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(annotateRequest.getProjectId(), annotateRequest.getFileId(), FileType.ORIGIN));
+        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(annotateRequest.getDossierId(), annotateRequest.getFileId(), FileType.ORIGIN));
-        var redactionLog = redactionStorageService.getRedactionLog(annotateRequest.getProjectId(), annotateRequest.getFileId());
+        var redactionLog = redactionStorageService.getRedactionLog(annotateRequest.getDossierId(), annotateRequest.getFileId());
-        var sectionsGrid = redactionStorageService.getSectionGrid(annotateRequest.getProjectId(), annotateRequest.getFileId());
+        var sectionsGrid = redactionStorageService.getSectionGrid(annotateRequest.getDossierId(), annotateRequest.getFileId());
 
         try (PDDocument pdDocument = PDDocument.load(storedObjectStream, MemoryUsageSetting.setupTempFileOnly())) {
             pdDocument.setAllSecurityToBeRemoved(true);
 
-            dictionaryService.updateDictionary(redactionLog.getRuleSetId(), annotateRequest.getProjectId());
+            dictionaryService.updateDictionary(annotateRequest.getDossierTemplateId(), annotateRequest.getDossierId());
             annotationService.annotate(pdDocument, redactionLog, sectionsGrid);
 
             try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream()) {
@@ -66,11 +63,11 @@ public class RedactionController implements RedactionResource {
 
     @Override
     public RedactionResult classify(@RequestBody RedactionRequest redactionRequest) {
-        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getProjectId(), redactionRequest.getFileId(), FileType.ORIGIN));
+        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getDossierId(), redactionRequest.getFileId(), FileType.ORIGIN));
         try {
             Document classifiedDoc = pdfSegmentationService.parseDocument(storedObjectStream);
 
-            storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getProjectId(), redactionRequest.getFileId(), FileType.ORIGIN));
+            storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getDossierId(), redactionRequest.getFileId(), FileType.ORIGIN));
             try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
                 pdDocument.setAllSecurityToBeRemoved(true);
 
@@ -91,11 +88,11 @@ public class RedactionController implements RedactionResource {
 
     @Override
     public RedactionResult sections(@RequestBody RedactionRequest redactionRequest) {
-        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getProjectId(), redactionRequest.getFileId(), FileType.ORIGIN));
+        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getDossierId(), redactionRequest.getFileId(), FileType.ORIGIN));
         try {
             Document classifiedDoc = pdfSegmentationService.parseDocument(storedObjectStream);
 
-            storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getProjectId(), redactionRequest.getFileId(), FileType.ORIGIN));
+            storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getDossierId(), redactionRequest.getFileId(), FileType.ORIGIN));
             try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
                 pdDocument.setAllSecurityToBeRemoved(true);
 
@@ -120,7 +117,7 @@ public class RedactionController implements RedactionResource {
         Document classifiedDoc;
 
         try {
-            var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getProjectId(), redactionRequest.getFileId(), FileType.ORIGIN));
+            var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getDossierId(), redactionRequest.getFileId(), FileType.ORIGIN));
             classifiedDoc = pdfSegmentationService.parseDocument(storedObjectStream, true);
         } catch (Exception e) {
             throw new RedactionException(e);
@@ -141,20 +138,28 @@ public class RedactionController implements RedactionResource {
 
     }
 
-
-    @Override
-    public void updateRules(@PathVariable(RULE_SET_PARAMETER_NAME) String ruleSetId) {
-
-        droolsExecutionService.updateRules(ruleSetId);
-    }
-
-
     @Override
     public void testRules(@RequestBody String rules) {
 
         droolsExecutionService.testRules(rules);
     }
 
+    @Override
+    public RedactionLog getRedactionLog(RedactionRequest redactionRequest) {
+
+        log.info("Requested preview for: {}", redactionRequest);
+        dictionaryService.updateDictionary(redactionRequest.getDossierTemplateId(), redactionRequest.getDossierId());
+
+        var redactionLog = redactionStorageService.getRedactionLog(redactionRequest.getDossierId(), redactionRequest.getFileId());
+
+        log.info("Loaded redaction log with computationalVersion: {}", redactionLog.getAnalysisVersion());
+        if (redactionLog.getAnalysisVersion() == 0) {
+            // old redaction logs are returned directly
+            return redactionLog;
+        } else {
+            return redactionLogMergeService.mergeRedactionLogData(redactionLog, redactionRequest.getDossierTemplateId(), redactionRequest.getManualRedactions(), redactionRequest.getExcludedPages());
+        }
+    }
 
     private RedactionResult convert(PDDocument document, int numberOfPages) throws IOException {
 

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

@@ -0,0 +1,21 @@
+package com.iqser.red.service.redaction.v1.server.controller;
+
+
+import com.iqser.red.service.redaction.v1.model.RuleBuilderModel;
+import com.iqser.red.service.redaction.v1.resources.RuleBuilderResource;
+import com.iqser.red.service.redaction.v1.server.redaction.rulebuilder.RuleBuilderModelService;
+import lombok.RequiredArgsConstructor;
+import org.springframework.web.bind.annotation.RestController;
+
+@RestController
+@RequiredArgsConstructor
+public class RuleBuilderController implements RuleBuilderResource {
+
+    private final RuleBuilderModelService ruleBuilderModelService;
+
+    @Override
+    public RuleBuilderModel getRuleBuilderModel() {
+        return ruleBuilderModelService.getRuleBuilderModel();
+    }
+
+}

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

@@ -46,6 +46,17 @@ public class PDFAreaTextStripper extends PDFTextStripperByArea {
                 startIndex = i;
             }
 
+
+            if (textPositions.get(i).getRotation() == 0 && i > 0 && textPositions.get(i).getX() > textPositions.get(i - 1).getEndX() + 1) {
+                List<TextPosition> sublist = textPositions.subList(startIndex, i);
+                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0)
+                        .getUnicode()
+                        .equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+                    textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
+                }
+                startIndex = i;
+            }
+
             if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
                     .getUnicode()
                     .equals("\u00A0")) && i <= textPositions.size() - 2) {

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

@@ -189,18 +189,16 @@ public class PDFLinesTextStripper extends PDFTextStripper {
             COSName objectName = (COSName) arguments.get(0);
             PDXObject xobject = getResources().getXObject(objectName);
             if (xobject instanceof PDImageXObject) {
-                PDImageXObject pdfImage = (PDImageXObject) xobject;
+                PDImageXObject image = (PDImageXObject)xobject;
+                Matrix ctmNew = getGraphicsState().getCurrentTransformationMatrix();
 
-                Rectangle2D imageBounds = calculateImagePosition(pdfImage);
+                Rectangle2D rect = new Rectangle2D.Float(ctmNew.getTranslateX(), ctmNew.getTranslateY(), ctmNew.getScaleX(), ctmNew.getScaleY());
-
-                Rectangle2D rect = new Rectangle2D.Float((float) imageBounds.getX(), (float) imageBounds.getY(), (float) imageBounds
-                        .getWidth(), (float) imageBounds.getHeight());
 
                 // Memory Hack - sofReference kills me
-                FieldUtils.writeField(pdfImage, "cachedImageSubsampling", -1, true);
+                FieldUtils.writeField(image, "cachedImageSubsampling", -1, true);
 
                 if (rect.getHeight() > 2 && rect.getWidth() > 2) {
-                    this.images.add(new PdfImage(pdfImage.getImage(), rect, pageNumber));
+                    this.images.add(new PdfImage(image.getImage(), rect, pageNumber, image.getImage().getColorModel().hasAlpha()));
                 }
             }
         } catch (Exception e) {
@@ -209,21 +207,6 @@ public class PDFLinesTextStripper extends PDFTextStripper {
     }
 
 
-    private Rectangle2D calculateImagePosition(PDImageXObject pdfImage) throws IOException {
-
-        Matrix ctm = getGraphicsState().getCurrentTransformationMatrix();
-
-        Rectangle2D imageBounds = pdfImage.getImage().getRaster().getBounds();
-
-        AffineTransform imageTransform = new AffineTransform(ctm.createAffineTransform());
-        imageTransform.scale(1.0 / pdfImage.getWidth(), -1.0 / pdfImage.getHeight());
-        imageTransform.translate(0, -pdfImage.getHeight());
-
-        AffineTransform pageTransform = new AffineTransform();
-        pageTransform.concatenate(imageTransform);
-
-        return pageTransform.createTransformedShape(imageBounds).getBounds2D();
-    }
 
 
     private float floatValue(COSBase value) {
@@ -300,6 +283,18 @@ public class PDFLinesTextStripper extends PDFTextStripper {
                 startIndex = i;
             }
 
+
+            if (textPositions.get(i).getRotation() == 0 && i > 0 && textPositions.get(i).getX() > textPositions.get(i - 1).getEndX() + 1) {
+                List<TextPosition> sublist = textPositions.subList(startIndex, i);
+                if (!(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0)
+                        .getUnicode()
+                        .equals(" ") || sublist.get(0).getUnicode().equals("\u00A0")))) {
+                    textPositionSequences.add(new TextPositionSequence(sublist, pageNumber));
+                }
+                startIndex = i;
+            }
+
+
             if (i > 0 && (textPositions.get(i).getUnicode().equals(" ") || textPositions.get(i)
                     .getUnicode()
                     .equals("\u00A0")) && i <= textPositions.size() - 2) {

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

@@ -21,7 +21,7 @@ public class MessagingConfiguration {
 
         return QueueBuilder.durable(REDACTION_QUEUE)
                 .withArgument("x-dead-letter-exchange", "")
-                .withArgument("x-dead-letter-routing-key", REDACTION_QUEUE)
+                .withArgument("x-dead-letter-routing-key", REDACTION_DQL)
                 .maxPriority(2)
                 .build();
     }

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

@@ -4,10 +4,14 @@ import com.fasterxml.jackson.core.JsonProcessingException;
 import com.fasterxml.jackson.databind.ObjectMapper;
 import com.iqser.red.service.redaction.v1.model.AnalyzeRequest;
 import com.iqser.red.service.redaction.v1.model.AnalyzeResult;
+import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
 import com.iqser.red.service.redaction.v1.server.client.FileStatusProcessingUpdateClient;
-import com.iqser.red.service.redaction.v1.server.redaction.service.ReanalyzeService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.AnalyzeService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.NerAnalyserService;
+
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
+
 import org.springframework.amqp.rabbit.annotation.RabbitHandler;
 import org.springframework.amqp.rabbit.annotation.RabbitListener;
 import org.springframework.stereotype.Service;
@@ -21,8 +25,10 @@ import static com.iqser.red.service.redaction.v1.server.queue.MessagingConfigura
 public class RedactionMessageReceiver {
 
     private final ObjectMapper objectMapper;
-    private final ReanalyzeService reanalyzeService;
+    private final AnalyzeService analyzeService;
     private final FileStatusProcessingUpdateClient fileStatusProcessingUpdateClient;
+    private final NerAnalyserService nerAnalyserService;
+
 
     @RabbitHandler
     @RabbitListener(queues = REDACTION_QUEUE)
@@ -32,15 +38,25 @@ public class RedactionMessageReceiver {
         log.info("Processing analyze request: {}", analyzeRequest);
         AnalyzeResult result;
         if (analyzeRequest.isReanalyseOnlyIfPossible()) {
-            result = reanalyzeService.reanalyze(analyzeRequest);
+            result = analyzeService.reanalyze(analyzeRequest);
+            log.info("Successfully reanalyzed dossier {} file {} took: {}", analyzeRequest.getDossierId(), analyzeRequest
+                    .getFileId(), result.getDuration());
         } else {
-            result = reanalyzeService.analyze(analyzeRequest);
+            // TODO Seperate stucture analysis by other queue
-        }
+            analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(analyzeRequest.getDossierId(), analyzeRequest.getFileId()));
-        log.info("Successfully analyzed {}", analyzeRequest);
 
-        fileStatusProcessingUpdateClient.analysisSuccessful(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), result);
+            // TODO NerEntities should be computed and stored in entity-recognition-service, should be triggered by a seperate queue after structure analysis
+            nerAnalyserService.computeNerEntities(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+
+            result = analyzeService.analyze(analyzeRequest);
+            log.info("Successfully analyzed dossier {} file {} took: {}", analyzeRequest.getDossierId(), analyzeRequest.getFileId(), result
+                    .getDuration());
         }
 
+        fileStatusProcessingUpdateClient.analysisSuccessful(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), result);
+    }
+
+
     @RabbitHandler
     @RabbitListener(queues = REDACTION_DQL)
     public void receiveAnalyzeRequestDQL(String in) throws JsonProcessingException {
@@ -48,7 +64,7 @@ public class RedactionMessageReceiver {
         var analyzeRequest = objectMapper.readValue(in, AnalyzeRequest.class);
         log.info("Failed to process analyze request: {}", analyzeRequest);
 
-        fileStatusProcessingUpdateClient.analysisFailed(analyzeRequest.getProjectId(), analyzeRequest.getFileId());
+        fileStatusProcessingUpdateClient.analysisFailed(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
     }
 
 }

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

@@ -1,7 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.redaction.model;
 
 
-import com.iqser.red.service.configuration.v1.api.model.DictionaryEntry;
 import lombok.AllArgsConstructor;
 import lombok.Data;
 
@@ -9,6 +8,8 @@ import java.io.Serializable;
 import java.util.Set;
 import java.util.stream.Collectors;
 
+import com.iqser.red.service.persistence.service.v1.api.model.data.configuration.DictionaryEntry;
+
 @Data
 @AllArgsConstructor
 public class DictionaryModel implements Serializable {

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

@@ -10,7 +10,7 @@ import java.util.Map;
 @Data
 public class DictionaryRepresentation {
 
-    private String ruleSetId;
+    private String dossierTemplateId;
     private long dictionaryVersion = -1;
     private List<DictionaryModel> dictionary = new ArrayList<>();
     private float[] defaultColor;

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

@@ -11,6 +11,6 @@ import lombok.NoArgsConstructor;
 @AllArgsConstructor
 public class DictionaryVersion {
 
-    long rulesetVersion;
+    long dossierTemplateVersion;
     long dossierVersion;
 }

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

@@ -1,16 +1,19 @@
 package com.iqser.red.service.redaction.v1.server.redaction.model;
 
+import com.iqser.red.service.redaction.v1.model.Engine;
 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
+
 import lombok.Data;
 import lombok.EqualsAndHashCode;
 
 import java.util.ArrayList;
+import java.util.HashSet;
 import java.util.List;
+import java.util.Set;
 
 @Data
 @EqualsAndHashCode(onlyExplicitlyIncluded = true)
-public class Entity {
+public class Entity implements ReasonHolder {
-
 
     private final String word;
     private final String type;
@@ -39,8 +42,15 @@ public class Entity {
 
     private boolean isDossierDictionaryEntry;
 
+    private Set<Engine> engines = new HashSet<>();
 
-    public Entity(String word, String type, boolean redaction, String redactionReason, List<EntityPositionSequence> positionSequences, String headline, int matchedRule, int sectionNumber, String legalBasis, boolean isDictionaryEntry, String textBefore, String textAfter, Integer start, Integer end, boolean isDossierDictionaryEntry) {
+    private Set<Entity> references = new HashSet<>();
+
+
+    public Entity(String word, String type, boolean redaction, String redactionReason,
+                  List<EntityPositionSequence> positionSequences, String headline, int matchedRule, int sectionNumber,
+                  String legalBasis, boolean isDictionaryEntry, String textBefore, String textAfter, Integer start,
+                  Integer end, boolean isDossierDictionaryEntry, Set<Engine> engines, Set<Entity> references) {
 
         this.word = word;
         this.type = type;
@@ -57,10 +67,13 @@ public class Entity {
         this.start = start;
         this.end = end;
         this.isDossierDictionaryEntry = isDossierDictionaryEntry;
+        this.engines = engines;
+        this.references = references;
     }
 
 
-    public Entity(String word, String type, Integer start, Integer end, String headline, int sectionNumber, boolean isDictionaryEntry, boolean isDossierDictionaryEntry) {
+    public Entity(String word, String type, Integer start, Integer end, String headline, int sectionNumber,
+                  boolean isDictionaryEntry, boolean isDossierDictionaryEntry, Engine engine) {
 
         this.word = word;
         this.type = type;
@@ -70,6 +83,8 @@ public class Entity {
         this.sectionNumber = sectionNumber;
         this.isDictionaryEntry = isDictionaryEntry;
         this.isDossierDictionaryEntry = isDossierDictionaryEntry;
+        this.engines.add(engine);
     }
 
+
 }

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

@@ -9,7 +9,7 @@ import lombok.NoArgsConstructor;
 @Builder
 @NoArgsConstructor
 @AllArgsConstructor
-public class Image {
+public class Image implements ReasonHolder {
 
     private String type;
     private RedRectangle2D position;
@@ -20,5 +20,6 @@ public class Image {
     private int sectionNumber;
     private String section;
     private int page;
+    private boolean hasTransparency;
 
 }

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

@@ -0,0 +1,23 @@
+package com.iqser.red.service.redaction.v1.server.redaction.model;
+
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+
+@Data
+@AllArgsConstructor
+public class PageEntities {
+
+    @Builder.Default
+    private Map<Integer, List<Entity>> entitiesPerPage = new HashMap<>();
+
+    @Builder.Default
+    private Map<Integer, Set<Image>> imagesPerPage = new HashMap<>();
+
+
+}

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

@@ -18,14 +18,16 @@ public class PdfImage {
     private RedRectangle2D position;
     private ImageType imageType;
     private boolean isAppendedToParagraph;
+    private boolean hasTransparency;
 
     @NonNull
     private int page;
 
-    public PdfImage(BufferedImage image, Rectangle2D position, int page) {
+    public PdfImage(BufferedImage image, Rectangle2D position, int page, boolean hasTransparency) {
         this.image = image;
         this.position = new RedRectangle2D(position.getX(), position.getY(), position.getWidth(), position.getHeight());
         this.page = page;
+        this.hasTransparency = hasTransparency;
     }
 
 }

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

@@ -0,0 +1,14 @@
+package com.iqser.red.service.redaction.v1.server.redaction.model;
+
+public interface ReasonHolder {
+
+    String getRedactionReason();
+
+    void setRedactionReason(String reason);
+
+    boolean isRedaction();
+
+    void setRedaction(boolean value);
+
+
+}

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

@@ -1,5 +1,8 @@
 package com.iqser.red.service.redaction.v1.server.redaction.model;
 
+import com.iqser.red.service.redaction.v1.model.ArgumentType;
+import com.iqser.red.service.redaction.v1.model.Engine;
+import com.iqser.red.service.redaction.v1.model.FileAttribute;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.EntitySearchUtils;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.Patterns;
@@ -8,11 +11,11 @@ import lombok.Data;
 import lombok.extern.slf4j.Slf4j;
 import org.apache.commons.lang3.StringUtils;
 
-import java.util.Collection;
+import java.lang.annotation.ElementType;
-import java.util.HashMap;
+import java.lang.annotation.Retention;
-import java.util.HashSet;
+import java.lang.annotation.RetentionPolicy;
-import java.util.Map;
+import java.lang.annotation.Target;
-import java.util.Set;
+import java.util.*;
 import java.util.regex.Matcher;
 import java.util.regex.Pattern;
 import java.util.stream.Collectors;
@@ -52,8 +55,49 @@ public class Section {
     @Builder.Default
     private Set<Image> images = new HashSet<>();
 
+    @Builder.Default
+    private List<FileAttribute> fileAttributes = new ArrayList<>();
 
-    public boolean rowEquals(String headerName, String value) {
+
+    @WhenCondition
+    public boolean fileAttributeByIdEquals(@Argument(ArgumentType.FILE_ATTRIBUTE) String id,
+                                           @Argument(ArgumentType.STRING) String value) {
+        return fileAttributes != null && fileAttributes.stream().filter(attribute -> id.equals(attribute.getId()) && value.equals(attribute.getValue())).findFirst().isPresent();
+    }
+
+    @WhenCondition
+    public boolean fileAttributeByPlaceholderEquals(@Argument(ArgumentType.FILE_ATTRIBUTE) String placeholder,
+                                                    @Argument(ArgumentType.STRING) String value) {
+        return fileAttributes != null && fileAttributes.stream().filter(attribute -> placeholder.equals(attribute.getPlaceholder()) && value.equals(attribute.getValue())).findFirst().isPresent();
+    }
+
+    @WhenCondition
+    public boolean fileAttributeByLabelEquals(@Argument(ArgumentType.FILE_ATTRIBUTE) String label,
+                                              @Argument(ArgumentType.STRING) String value) {
+        return fileAttributes != null && fileAttributes.stream().filter(attribute -> label.equals(attribute.getLabel()) && value.equals(attribute.getValue())).findFirst().isPresent();
+    }
+
+    @WhenCondition
+    public boolean fileAttributeByIdEqualsIgnoreCase(@Argument(ArgumentType.FILE_ATTRIBUTE) String id,
+                                                     @Argument(ArgumentType.STRING) String value) {
+        return fileAttributes != null && fileAttributes.stream().filter(attribute -> id.equals(attribute.getId()) && value.equalsIgnoreCase(attribute.getValue())).findFirst().isPresent();
+    }
+
+    @WhenCondition
+    public boolean fileAttributeByPlaceholderEqualsIgnoreCase(@Argument(ArgumentType.FILE_ATTRIBUTE) String placeholder,
+                                                              @Argument(ArgumentType.STRING) String value) {
+        return fileAttributes != null && fileAttributes.stream().filter(attribute -> placeholder.equals(attribute.getPlaceholder()) && value.equalsIgnoreCase(attribute.getValue())).findFirst().isPresent();
+    }
+
+    @WhenCondition
+    public boolean fileAttributeByLabelEqualsIgnoreCase(@Argument(ArgumentType.FILE_ATTRIBUTE) String label,
+                                                        @Argument(ArgumentType.STRING) String value) {
+        return fileAttributes != null && fileAttributes.stream().filter(attribute -> label.equals(attribute.getLabel()) && value.equalsIgnoreCase(attribute.getValue())).findFirst().isPresent();
+    }
+
+    @WhenCondition
+    public boolean rowEquals(@Argument(ArgumentType.STRING) String headerName,
+                             @Argument(ArgumentType.STRING) String value) {
 
         String cleanHeaderName = headerName.replaceAll("\n", "").replaceAll(" ", "").replaceAll("-", "");
 
@@ -62,33 +106,36 @@ public class Section {
                 .equals(value);
     }
 
-
+    @WhenCondition
-    public boolean hasTableHeader(String headerName) {
+    public boolean hasTableHeader(@Argument(ArgumentType.STRING) String headerName) {
 
         String cleanHeaderName = headerName.replaceAll("\n", "").replaceAll(" ", "").replaceAll("-", "");
         return tabularData != null && tabularData.containsKey(cleanHeaderName);
     }
 
-
+    @WhenCondition
-    public boolean matchesType(String type) {
+    public boolean matchesType(@Argument(ArgumentType.TYPE) String type) {
 
         return entities.stream().anyMatch(entity -> entity.getType().equals(type));
     }
 
-
+    @WhenCondition
-    public boolean matchesImageType(String type) {
+    public boolean matchesImageType(@Argument(ArgumentType.TYPE) String type) {
 
         return images.stream().anyMatch(image -> image.getType().equals(type));
     }
 
-
+    @WhenCondition
-    public boolean headlineContainsWord(String word) {
+    public boolean headlineContainsWord(@Argument(ArgumentType.STRING) String word) {
 
         return StringUtils.containsIgnoreCase(headline, word);
     }
 
-
+    @ThenAction
-    public void expandByRegEx(String type, String pattern, boolean patternCaseInsensitive, int group) {
+    public void expandByRegEx(@Argument(ArgumentType.TYPE) String type,
+                              @Argument(ArgumentType.REGEX) String pattern,
+                              @Argument(ArgumentType.BOOLEAN) boolean patternCaseInsensitive,
+                              @Argument(ArgumentType.INTEGER) int group) {
 
         Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
 
@@ -115,8 +162,11 @@ public class Section {
         EntitySearchUtils.removeEntitiesContainedInLarger(entities);
     }
 
-
+    @ThenAction
-    public void redactImage(String type, int ruleNumber, String reason, String legalBasis) {
+    public void redactImage(@Argument(ArgumentType.TYPE) String type,
+                            @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                            @Argument(ArgumentType.STRING) String reason,
+                            @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         images.forEach(image -> {
             if (image.getType().equals(type)) {
@@ -128,8 +178,11 @@ public class Section {
         });
     }
 
-
+    @ThenAction
-    public void redact(String type, int ruleNumber, String reason, String legalBasis) {
+    public void redact(@Argument(ArgumentType.TYPE) String type,
+                       @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                       @Argument(ArgumentType.STRING) String reason,
+                       @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         boolean hasRecommendationDictionary = dictionaryTypes.contains(RECOMMENDATION_PREFIX + type);
 
@@ -144,8 +197,10 @@ public class Section {
         });
     }
 
-
+    @ThenAction
-    public void redactNotImage(String type, int ruleNumber, String reason) {
+    public void redactNotImage(@Argument(ArgumentType.TYPE) String type,
+                               @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                               @Argument(ArgumentType.STRING) String reason) {
 
         images.forEach(image -> {
             if (image.getType().equals(type)) {
@@ -156,8 +211,10 @@ public class Section {
         });
     }
 
-
+    @ThenAction
-    public void redactNot(String type, int ruleNumber, String reason) {
+    public void redactNot(@Argument(ArgumentType.TYPE) String type,
+                          @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                          @Argument(ArgumentType.STRING) String reason) {
 
         boolean hasRecommendationDictionary = dictionaryTypes.contains(RECOMMENDATION_PREFIX + type);
 
@@ -172,8 +229,35 @@ public class Section {
     }
 
 
-    public void expandToHintAnnotationByRegEx(String type, String pattern, boolean patternCaseInsensitive, int group,
+    @ThenAction
-                                              String asType) {
+    public void redactNotAndReference(@Argument(ArgumentType.TYPE) String type,
+                                      @Argument(ArgumentType.REFERENCE_TYPE) String referenceType,
+                          @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                          @Argument(ArgumentType.STRING) String reason) {
+
+        boolean hasRecommendationDictionary = dictionaryTypes.contains(RECOMMENDATION_PREFIX + type);
+
+        Set<Entity> references = entities.stream().filter(entity -> entity.getType().equals(referenceType)).collect(Collectors.toSet());
+
+        entities.forEach(entity -> {
+            if (entity.getType().equals(type) || hasRecommendationDictionary && entity.getType()
+                    .equals(RECOMMENDATION_PREFIX + type)) {
+                entity.setRedaction(false);
+                entity.setMatchedRule(ruleNumber);
+                entity.setRedactionReason(reason);
+                entity.setReferences(references);
+            }
+        });
+    }
+
+
+
+    @ThenAction
+    public void expandToHintAnnotationByRegEx(@Argument(ArgumentType.TYPE) String type,
+                                              @Argument(ArgumentType.STRING) String pattern,
+                                              @Argument(ArgumentType.BOOLEAN) boolean patternCaseInsensitive,
+                                              @Argument(ArgumentType.INTEGER) int group,
+                                              @Argument(ArgumentType.TYPE) String asType) {
 
         Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
 
@@ -198,8 +282,11 @@ public class Section {
         EntitySearchUtils.removeEntitiesContainedInLarger(entities);
     }
 
-
+    @ThenAction
-    public void addHintAnnotationByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType) {
+    public void addHintAnnotationByRegEx(@Argument(ArgumentType.REGEX) String pattern,
+                                         @Argument(ArgumentType.BOOLEAN) boolean patternCaseInsensitive,
+                                         @Argument(ArgumentType.INTEGER) int group,
+                                         @Argument(ArgumentType.TYPE) String asType) {
 
         Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
 
@@ -214,8 +301,12 @@ public class Section {
         }
     }
 
-
+    @ThenAction
-    public void redactIfPrecededBy(String prefix, String type, int ruleNumber, String reason, String legalBasis) {
+    public void redactIfPrecededBy(@Argument(ArgumentType.STRING) String prefix,
+                                   @Argument(ArgumentType.TYPE) String type,
+                                   @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                                   @Argument(ArgumentType.STRING) String reason,
+                                   @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         entities.forEach(entity -> {
             if (entity.getType().equals(type) && searchText.indexOf(prefix + entity.getWord()) != 1) {
@@ -227,23 +318,32 @@ public class Section {
         });
     }
 
-
+    @ThenAction
-    public void addHintAnnotation(String value, String asType) {
+    public void addHintAnnotation(@Argument(ArgumentType.STRING) String value,
+                                  @Argument(ArgumentType.TYPE) String asType) {
 
         Set<Entity> found = findEntities(value.trim(), asType, true, false, 0, null, null);
         EntitySearchUtils.addEntitiesIgnoreRank(entities, found);
     }
 
-
+    @ThenAction
-    public void addRedaction(String value, String asType, int ruleNumber, String reason, String legalBasis) {
+    public void addRedaction(@Argument(ArgumentType.STRING) String value,
+                             @Argument(ArgumentType.TYPE) String asType,
+                             @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                             @Argument(ArgumentType.STRING) String reason,
+                             @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         Set<Entity> found = findEntities(value.trim(), asType, true, true, ruleNumber, reason, legalBasis);
         EntitySearchUtils.addEntitiesIgnoreRank(entities, found);
     }
 
-
+    @ThenAction
-    public void redactLineAfter(String start, String asType, int ruleNumber, boolean redactEverywhere, String reason,
+    public void redactLineAfter(@Argument(ArgumentType.STRING) String start,
-                                String legalBasis) {
+                                @Argument(ArgumentType.TYPE) String asType,
+                                @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                                @Argument(ArgumentType.BOOLEAN) boolean redactEverywhere,
+                                @Argument(ArgumentType.STRING) String reason,
+                                @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         String[] values = StringUtils.substringsBetween(text, start, "\n");
 
@@ -261,8 +361,9 @@ public class Section {
         }
     }
 
-
+    @ThenAction
-    public void recommendLineAfter(String start, String asType) {
+    public void recommendLineAfter(@Argument(ArgumentType.STRING) String start,
+                                   @Argument(ArgumentType.TYPE) String asType) {
 
         String[] values = StringUtils.substringsBetween(text, start, "\n");
 
@@ -285,9 +386,14 @@ public class Section {
         }
     }
 
-
+    @ThenAction
-    public void redactByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType, int ruleNumber,
+    public void redactByRegEx(@Argument(ArgumentType.REGEX) String pattern,
-                              String reason, String legalBasis) {
+                              @Argument(ArgumentType.BOOLEAN) boolean patternCaseInsensitive,
+                              @Argument(ArgumentType.INTEGER) int group,
+                              @Argument(ArgumentType.TYPE) String asType,
+                              @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                              @Argument(ArgumentType.STRING) String reason,
+                              @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
 
@@ -302,8 +408,11 @@ public class Section {
         }
     }
 
-
+    @ThenAction
-    public void addRecommendationByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType) {
+    public void addRecommendationByRegEx(@Argument(ArgumentType.REGEX) String pattern,
+                                         @Argument(ArgumentType.BOOLEAN) boolean patternCaseInsensitive,
+                                         @Argument(ArgumentType.INTEGER) int group,
+                                         @Argument(ArgumentType.TYPE) String asType) {
 
         Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
 
@@ -317,9 +426,14 @@ public class Section {
         }
     }
 
-
+    @ThenAction
-    public void redactAndRecommendByRegEx(String pattern, boolean patternCaseInsensitive, int group, String asType,
+    public void redactAndRecommendByRegEx(@Argument(ArgumentType.REGEX) String pattern,
-                                          int ruleNumber, String reason, String legalBasis) {
+                                          @Argument(ArgumentType.BOOLEAN) boolean patternCaseInsensitive,
+                                          @Argument(ArgumentType.INTEGER) int group,
+                                          @Argument(ArgumentType.TYPE) String asType,
+                                          @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                                          @Argument(ArgumentType.STRING) String reason,
+                                          @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         Pattern compiledPattern = Patterns.getCompiledPattern(pattern, patternCaseInsensitive);
 
@@ -334,9 +448,14 @@ public class Section {
         }
     }
 
-
+    @ThenAction
-    public void redactBetween(String start, String stop, String asType, int ruleNumber, boolean redactEverywhere,
+    public void redactBetween(@Argument(ArgumentType.STRING) String start,
-                              String reason, String legalBasis) {
+                              @Argument(ArgumentType.STRING) String stop,
+                              @Argument(ArgumentType.TYPE) String asType,
+                              @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                              @Argument(ArgumentType.BOOLEAN) boolean redactEverywhere,
+                              @Argument(ArgumentType.STRING) String reason,
+                              @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         String[] values = StringUtils.substringsBetween(searchText, start, stop);
 
@@ -355,9 +474,14 @@ public class Section {
         }
     }
 
-
+    @ThenAction
-    public void redactLinesBetween(String start, String stop, String asType, int ruleNumber, boolean redactEverywhere,
+    public void redactLinesBetween(@Argument(ArgumentType.STRING) String start,
-                                   String reason, String legalBasis) {
+                                   @Argument(ArgumentType.STRING) String stop,
+                                   @Argument(ArgumentType.TYPE) String asType,
+                                   @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                                   @Argument(ArgumentType.BOOLEAN) boolean redactEverywhere,
+                                   @Argument(ArgumentType.STRING) String reason,
+                                   @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         String[] values = StringUtils.substringsBetween(text, start, stop);
 
@@ -384,34 +508,48 @@ public class Section {
         }
     }
 
-
+    @ThenAction
-    public void highlightCell(String cellHeader, int ruleNumber, String type) {
+    public void highlightCell(@Argument(ArgumentType.STRING) String cellHeader,
+                              @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                              @Argument(ArgumentType.TYPE) String type) {
 
         annotateCell(cellHeader, ruleNumber, type, false, false, null, null);
     }
 
-
+    @ThenAction
-    public void redactCell(String cellHeader, int ruleNumber, String type, boolean addAsRecommendations, String reason,
+    public void redactCell(@Argument(ArgumentType.STRING) String cellHeader,
-                           String legalBasis) {
+                           @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                           @Argument(ArgumentType.TYPE) String type,
+                           @Argument(ArgumentType.BOOLEAN) boolean addAsRecommendations,
+                           @Argument(ArgumentType.STRING) String reason,
+                           @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         annotateCell(cellHeader, ruleNumber, type, true, addAsRecommendations, reason, legalBasis);
     }
 
-
+    @ThenAction
-    public void redactNotCell(String cellHeader, int ruleNumber, String type, boolean addAsRecommendations,
+    public void redactNotCell(@Argument(ArgumentType.STRING) String cellHeader,
-                              String reason) {
+                              @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                              @Argument(ArgumentType.TYPE) String type,
+                              @Argument(ArgumentType.BOOLEAN) boolean addAsRecommendations,
+                              @Argument(ArgumentType.STRING) String reason) {
 
         annotateCell(cellHeader, ruleNumber, type, false, addAsRecommendations, reason, null);
     }
 
 
-    private Set<Entity> findEntities(String value, String asType, boolean caseInsensitive, boolean redacted,
+    private Set<Entity> findEntities(@Argument(ArgumentType.STRING) String value,
-                                     int ruleNumber, String reason, String legalBasis) {
+                                     @Argument(ArgumentType.TYPE) String asType,
+                                     @Argument(ArgumentType.BOOLEAN) boolean caseInsensitive,
+                                     @Argument(ArgumentType.BOOLEAN) boolean redacted,
+                                     @Argument(ArgumentType.RULE_NUMBER) int ruleNumber,
+                                     @Argument(ArgumentType.STRING) String reason,
+                                     @Argument(ArgumentType.LEGAL_BASIS) String legalBasis) {
 
         String text = caseInsensitive ? searchText.toLowerCase() : searchText;
         String searchValue = caseInsensitive ? value.toLowerCase() : value;
 
-        Set<Entity> found = EntitySearchUtils.find(text, Set.of(searchValue), asType, headline, sectionNumber, true, false);
+        Set<Entity> found = EntitySearchUtils.find(text, Set.of(searchValue), asType, headline, sectionNumber, false, false, Engine.RULE);
 
         found.forEach(entity -> {
             if (redacted) {
@@ -437,7 +575,7 @@ public class Section {
         } else {
             String word = value.toString();
 
-            Entity entity = new Entity(word, type, value.getRowSpanStart(), value.getRowSpanStart() + word.length(), headline, sectionNumber, false, false);
+            Entity entity = new Entity(word, type, value.getRowSpanStart(), value.getRowSpanStart() + word.length(), headline, sectionNumber, false, false, Engine.RULE);
             entity.setRedaction(redact);
             entity.setMatchedRule(ruleNumber);
             entity.setRedactionReason(reason);
@@ -475,6 +613,25 @@ public class Section {
         }
     }
 
+    @Retention(RetentionPolicy.RUNTIME)
+    @Target(ElementType.METHOD)
+    public @interface WhenCondition {
+
+    }
+
+    @Retention(RetentionPolicy.RUNTIME)
+    @Target(ElementType.METHOD)
+    public @interface ThenAction {
+
+    }
+
+    @Retention(RetentionPolicy.RUNTIME)
+    @Target(ElementType.PARAMETER)
+    public @interface Argument {
+
+        ArgumentType value() default ArgumentType.STRING;
+    }
+
 }
 
 

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

@@ -0,0 +1,36 @@
+package com.iqser.red.service.redaction.v1.server.redaction.rulebuilder;
+
+import com.iqser.red.service.redaction.v1.model.Argument;
+import com.iqser.red.service.redaction.v1.model.RuleBuilderModel;
+import com.iqser.red.service.redaction.v1.model.RuleElement;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Section;
+import org.springframework.stereotype.Service;
+
+import java.lang.reflect.Method;
+import java.util.Arrays;
+import java.util.List;
+import java.util.stream.Collectors;
+
+@Service
+public class RuleBuilderModelService {
+
+    public RuleBuilderModel getRuleBuilderModel() {
+
+        var whenConditions = Arrays.stream(Section.class.getDeclaredMethods()).filter(m -> m.isAnnotationPresent(Section.WhenCondition.class)).collect(Collectors.toList());
+        var thenActions = Arrays.stream(Section.class.getDeclaredMethods()).filter(m -> m.isAnnotationPresent(Section.ThenAction.class)).collect(Collectors.toList());
+
+        RuleBuilderModel ruleBuilderModel = new RuleBuilderModel();
+
+
+        ruleBuilderModel.setWhenClauses(whenConditions.stream().map(c -> new RuleElement(c.getName(), toArguments(c))).collect(Collectors.toList()));
+        ruleBuilderModel.setThenConditions(thenActions.stream().map(c -> new RuleElement(c.getName(), toArguments(c))).collect(Collectors.toList()));
+
+        return ruleBuilderModel;
+    }
+
+    private List<Argument> toArguments(Method c) {
+        return Arrays.stream(c.getParameters())
+                .map(parameter -> new Argument(parameter.getName(), parameter.getAnnotation(Section.Argument.class).value()))
+                .collect(Collectors.toList());
+    }
+}

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

@@ -1,49 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.redaction.service;
-
-import com.iqser.red.service.redaction.v1.model.AnalyzeResult;
-import com.iqser.red.service.redaction.v1.model.RedactionChangeLog;
-import com.iqser.red.service.redaction.v1.model.RedactionLog;
-import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
-import org.springframework.stereotype.Service;
-
-@Service
-public class AnalyzeResponseService {
-
-    public AnalyzeResult createAnalyzeResponse(String projectId, String fileId, long duration, int pageCount, RedactionLog redactionLog, RedactionChangeLog redactionChangeLog) {
-        boolean hasHints = redactionLog.getRedactionLogEntry().stream().anyMatch(RedactionLogEntry::isHint);
-
-        boolean hasRequests = redactionLog.getRedactionLogEntry()
-                .stream()
-                .anyMatch(entry -> entry.isManual() && entry.getStatus()
-                        .equals(com.iqser.red.service.redaction.v1.model.Status.REQUESTED));
-
-        boolean hasRedactions = redactionLog.getRedactionLogEntry()
-                .stream()
-                .anyMatch(entry -> entry.isRedacted() && !entry.isManual() || entry.isManual() && entry.getStatus()
-                        .equals(com.iqser.red.service.redaction.v1.model.Status.APPROVED));
-
-        boolean hasImages = redactionLog.getRedactionLogEntry()
-                .stream()
-                .anyMatch(entry -> entry.isHint() && entry.getType().equals("image"));
-
-        boolean hasUpdates = redactionChangeLog != null && redactionChangeLog.getRedactionLogEntry() != null && !redactionChangeLog
-                .getRedactionLogEntry()
-                .isEmpty() && redactionChangeLog.getRedactionLogEntry().stream().anyMatch(entry -> !entry.getType().equals("false_positive"));
-
-        return AnalyzeResult.builder()
-                .projectId(projectId)
-                .fileId(fileId)
-                .duration(duration)
-                .numberOfPages(pageCount)
-                .hasHints(hasHints)
-                .hasRedactions(hasRedactions)
-                .hasRequests(hasRequests)
-                .hasImages(hasImages)
-                .hasUpdates(hasUpdates)
-                .rulesVersion(redactionLog.getRulesVersion())
-                .dictionaryVersion(redactionLog.getDictionaryVersion())
-                .legalBasisVersion(redactionLog.getLegalBasisVersion())
-                .dossierDictionaryVersion(redactionLog.getDossierDictionaryVersion())
-                .build();
-    }
-}

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

@@ -0,0 +1,283 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+import java.util.stream.Stream;
+
+import org.kie.api.runtime.KieContainer;
+import org.springframework.stereotype.Service;
+import org.springframework.web.bind.annotation.RequestBody;
+
+import com.iqser.red.service.persistence.service.v1.api.model.FileType;
+import com.iqser.red.service.redaction.v1.model.AnalyzeRequest;
+import com.iqser.red.service.redaction.v1.model.AnalyzeResult;
+import com.iqser.red.service.redaction.v1.model.IdRemoval;
+import com.iqser.red.service.redaction.v1.model.ManualForceRedact;
+import com.iqser.red.service.redaction.v1.model.ManualImageRecategorization;
+import com.iqser.red.service.redaction.v1.model.ManualLegalBasisChange;
+import com.iqser.red.service.redaction.v1.model.ManualRedactions;
+import com.iqser.red.service.redaction.v1.model.Rectangle;
+import com.iqser.red.service.redaction.v1.model.RedactionLog;
+import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import com.iqser.red.service.redaction.v1.model.StructureAnalyzeRequest;
+import com.iqser.red.service.redaction.v1.server.classification.model.Document;
+import com.iqser.red.service.redaction.v1.server.classification.model.SectionText;
+import com.iqser.red.service.redaction.v1.server.classification.model.Text;
+import com.iqser.red.service.redaction.v1.server.client.LegalBasisClient;
+import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
+import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryIncrement;
+import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryVersion;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Image;
+import com.iqser.red.service.redaction.v1.server.redaction.model.PageEntities;
+import com.iqser.red.service.redaction.v1.server.redaction.model.RedRectangle2D;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.EntitySearchUtils;
+import com.iqser.red.service.redaction.v1.server.segmentation.PdfSegmentationService;
+import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
+import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
+
+import lombok.RequiredArgsConstructor;
+import lombok.SneakyThrows;
+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
+@Service
+@RequiredArgsConstructor
+public class AnalyzeService {
+
+    private final DictionaryService dictionaryService;
+    private final DroolsExecutionService droolsExecutionService;
+    private final EntityRedactionService entityRedactionService;
+    private final RedactionLogCreatorService redactionLogCreatorService;
+    private final RedactionStorageService redactionStorageService;
+    private final PdfSegmentationService pdfSegmentationService;
+    private final RedactionChangeLogService redactionChangeLogService;
+    private final LegalBasisClient legalBasisClient;
+    private final RedactionServiceSettings redactionServiceSettings;
+    private final SectionTextBuilderService sectionTextBuilderService;
+    private final SectionGridCreatorService sectionGridCreatorService;
+    private final NerAnalyserService nerAnalyserService;
+
+
+    public void analyzeDocumentStructure(StructureAnalyzeRequest analyzeRequest) {
+
+        long startTime = System.currentTimeMillis();
+
+        var pageCount = 0;
+        Document classifiedDoc;
+
+        try {
+            var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(analyzeRequest
+                    .getDossierId(), analyzeRequest.getFileId(), FileType.ORIGIN));
+            classifiedDoc = pdfSegmentationService.parseDocument(storedObjectStream);
+            pageCount = classifiedDoc.getPages().size();
+        } catch (Exception e) {
+            throw new RedactionException(e);
+        }
+
+        List<SectionText> sectionTexts = sectionTextBuilderService.buildSectionText(classifiedDoc);
+        sectionGridCreatorService.createSectionGrid(classifiedDoc, pageCount);
+
+        Text text = new Text(pageCount, sectionTexts);
+        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.TEXT, text);
+        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.SECTION_GRID, classifiedDoc
+                .getSectionGrid());
+
+        log.info("Document structure analysis successful, took: {}", System.currentTimeMillis() - startTime);
+    }
+
+
+    public AnalyzeResult analyze(AnalyzeRequest analyzeRequest) {
+
+        long startTime = System.currentTimeMillis();
+
+        var text = redactionStorageService.getText(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+        var nerEntities = redactionStorageService.getNerEntities(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+        if(redactionServiceSettings.isEnableEntityRecognition() && nerEntities == null){
+            nerAnalyserService.computeNerEntities(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+            nerEntities = redactionStorageService.getNerEntities(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+        }
+
+        dictionaryService.updateDictionary(analyzeRequest.getDossierTemplateId(), analyzeRequest.getDossierId());
+        KieContainer kieContainer = droolsExecutionService.updateRules(analyzeRequest.getDossierTemplateId());
+        long rulesVersion = droolsExecutionService.getRulesVersion(analyzeRequest.getDossierTemplateId());
+        Dictionary dictionary = dictionaryService.getDeepCopyDictionary(analyzeRequest.getDossierTemplateId(), analyzeRequest
+                .getDossierId());
+
+        PageEntities pageEntities = entityRedactionService.findEntities(dictionary, text.getSectionTexts(), kieContainer, analyzeRequest, nerEntities);
+
+        dictionaryService.updateExternalDictionary(dictionary, analyzeRequest.getDossierTemplateId());
+
+        List<RedactionLogEntry> redactionLogEntries = redactionLogCreatorService.createRedactionLog(pageEntities, text.getNumberOfPages(), analyzeRequest
+                .getDossierTemplateId());
+
+        var legalBasis = legalBasisClient.getLegalBasisMapping(analyzeRequest.getDossierTemplateId());
+        var redactionLog = new RedactionLog(redactionServiceSettings.getAnalysisVersion(), redactionLogEntries, legalBasis, dictionary
+                .getVersion()
+                .getDossierTemplateVersion(), dictionary.getVersion()
+                .getDossierVersion(), rulesVersion, legalBasisClient.getVersion(analyzeRequest.getDossierTemplateId()));
+
+        return finalizeAnalysis(analyzeRequest, startTime, redactionLog, text, dictionary.getVersion(), false);
+    }
+
+
+    @SneakyThrows
+    public AnalyzeResult reanalyze(@RequestBody AnalyzeRequest analyzeRequest) {
+
+        long startTime = System.currentTimeMillis();
+
+        var redactionLog = redactionStorageService.getRedactionLog(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+        var text = redactionStorageService.getText(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+
+        // not yet ready for reanalysis
+        if (redactionLog == null || text == null || text.getNumberOfPages() == 0) {
+            return analyze(analyzeRequest);
+        }
+
+        DictionaryIncrement dictionaryIncrement = dictionaryService.getDictionaryIncrements(analyzeRequest.getDossierTemplateId(), new DictionaryVersion(redactionLog
+                .getDictionaryVersion(), redactionLog.getDossierDictionaryVersion()), analyzeRequest.getDossierId());
+
+        Set<Integer> sectionsToReanalyse = !analyzeRequest.getSectionsToReanalyse()
+                .isEmpty() ? analyzeRequest.getSectionsToReanalyse() : findSectionsToReanalyse(dictionaryIncrement, redactionLog, text, analyzeRequest);
+
+        if (sectionsToReanalyse.isEmpty()) {
+            return finalizeAnalysis(analyzeRequest, startTime, redactionLog, text, dictionaryIncrement.getDictionaryVersion(), true);
+        }
+
+        var nerEntities = redactionStorageService.getNerEntities(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+        if(redactionServiceSettings.isEnableEntityRecognition() && nerEntities == null){
+            nerAnalyserService.computeNerEntities(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+            nerEntities = redactionStorageService.getNerEntities(analyzeRequest.getDossierId(), analyzeRequest.getFileId());
+        }
+
+        List<SectionText> reanalysisSections = text.getSectionTexts()
+                .stream()
+                .filter(sectionText -> sectionsToReanalyse.contains(sectionText.getSectionNumber()))
+                .collect(Collectors.toList());
+
+        KieContainer kieContainer = droolsExecutionService.updateRules(analyzeRequest.getDossierTemplateId());
+
+        Dictionary dictionary = dictionaryService.getDeepCopyDictionary(analyzeRequest.getDossierTemplateId(), analyzeRequest
+                .getDossierId());
+
+        PageEntities pageEntities = entityRedactionService.findEntities(dictionary, reanalysisSections, kieContainer, analyzeRequest, nerEntities);
+        var newRedactionLogEntries = redactionLogCreatorService.createRedactionLog(pageEntities, text.getNumberOfPages(), analyzeRequest
+                .getDossierTemplateId());
+
+        redactionLog.getRedactionLogEntry().removeIf(entry -> sectionsToReanalyse.contains(entry.getSectionNumber()));
+        redactionLog.getRedactionLogEntry().addAll(newRedactionLogEntries);
+        return finalizeAnalysis(analyzeRequest, startTime, redactionLog, text, dictionaryIncrement.getDictionaryVersion(), true);
+    }
+
+
+    private Set<Integer> findSectionsToReanalyse(DictionaryIncrement dictionaryIncrement, RedactionLog redactionLog,
+                                                 Text text, AnalyzeRequest analyzeRequest) {
+
+        long start = System.currentTimeMillis();
+        Set<String> relevantManuallyModifiedAnnotationIds = getRelevantManuallyModifiedAnnotationIds(analyzeRequest.getManualRedactions());
+
+        Set<Integer> sectionsToReanalyse = new HashSet<>();
+        Map<Integer, Set<Image>> imageEntries = new HashMap<>();
+        for (RedactionLogEntry entry : redactionLog.getRedactionLogEntry()) {
+            if (entry.isManual() || relevantManuallyModifiedAnnotationIds.contains(entry.getId())) {
+                sectionsToReanalyse.add(entry.getSectionNumber());
+            }
+            if (entry.isImage() || entry.getType().equals("image")) {
+                imageEntries.computeIfAbsent(entry.getSectionNumber(), x -> new HashSet<>()).add(convert(entry));
+            }
+        }
+
+        for (SectionText sectionText : text.getSectionTexts()) {
+
+            if (EntitySearchUtils.sectionContainsAny(sectionText.getText(), dictionaryIncrement.getValues())) {
+                sectionsToReanalyse.add(sectionText.getSectionNumber());
+            }
+
+        }
+
+        log.info("Should reanalyze {} sections for request: {}, took: {}", sectionsToReanalyse.size(), analyzeRequest, System.currentTimeMillis() - start);
+
+        return sectionsToReanalyse;
+    }
+
+
+    private AnalyzeResult finalizeAnalysis(@RequestBody AnalyzeRequest analyzeRequest, long startTime,
+                                           RedactionLog redactionLog, Text text, DictionaryVersion dictionaryVersion,
+                                           boolean isReanalysis) {
+
+        redactionLog.setDictionaryVersion(dictionaryVersion.getDossierTemplateVersion());
+        redactionLog.setDossierDictionaryVersion(dictionaryVersion.getDossierVersion());
+
+        excludeExcludedPages(redactionLog, analyzeRequest.getExcludedPages());
+
+        var redactionLogChange = redactionChangeLogService.computeChanges(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), redactionLog);
+        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.REDACTION_LOG, redactionLogChange
+                .getRedactionLog());
+
+        long duration = System.currentTimeMillis() - startTime;
+
+        return AnalyzeResult.builder()
+                .dossierId(analyzeRequest.getDossierId())
+                .fileId(analyzeRequest.getFileId())
+                .duration(duration)
+                .numberOfPages(text.getNumberOfPages())
+                .hasUpdates(redactionLogChange.isHasChanges())
+                .analysisVersion(redactionServiceSettings.getAnalysisVersion())
+                .rulesVersion(redactionLog.getRulesVersion())
+                .dictionaryVersion(redactionLog.getDictionaryVersion())
+                .legalBasisVersion(redactionLog.getLegalBasisVersion())
+                .dossierDictionaryVersion(redactionLog.getDossierDictionaryVersion())
+                .wasReanalyzed(isReanalysis)
+                .build();
+    }
+
+
+    private Set<String> getRelevantManuallyModifiedAnnotationIds(ManualRedactions manualRedactions) {
+
+        if (manualRedactions == null) {
+            return new HashSet<>();
+        }
+
+        return Stream.concat(manualRedactions.getManualLegalBasisChanges()
+                .stream()
+                .map(ManualLegalBasisChange::getId), Stream.concat(manualRedactions.getImageRecategorizations()
+                .stream()
+                .map(ManualImageRecategorization::getId), Stream.concat(manualRedactions.getIdsToRemove()
+                .stream()
+                .map(IdRemoval::getId), manualRedactions.getForceRedacts().stream().map(ManualForceRedact::getId))))
+                .collect(Collectors.toSet());
+    }
+
+
+    public Image convert(RedactionLogEntry entry) {
+
+        Rectangle position = entry.getPositions().get(0);
+
+        return Image.builder()
+                .type(entry.getType())
+                .position(new RedRectangle2D(position.getTopLeft().getX(), position.getTopLeft()
+                        .getY(), position.getWidth(), position.getHeight()))
+                .sectionNumber(entry.getSectionNumber())
+                .section(entry.getSection())
+                .page(position.getPage())
+                .hasTransparency(entry.isImageHasTransparency())
+                .build();
+    }
+
+
+    private void excludeExcludedPages(RedactionLog redactionLog, Set<Integer> excludedPages) {
+
+        if (excludedPages != null && !excludedPages.isEmpty()) {
+            redactionLog.getRedactionLogEntry().forEach(entry -> entry.getPositions().forEach(pos -> {
+                if (excludedPages.contains(pos.getPage())) {
+                    entry.setExcluded(true);
+                }
+            }));
+        }
+    }
+
+}

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

@@ -1,19 +1,10 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;
 
-import static com.iqser.red.service.configuration.v1.api.resource.DictionaryResource.GLOBAL_DOSSIER;
+import com.iqser.red.service.persistence.service.v1.api.model.data.configuration.Colors;
-
+import com.iqser.red.service.persistence.service.v1.api.model.data.configuration.DictionaryEntry;
-import com.iqser.red.service.configuration.v1.api.model.Colors;
-import com.iqser.red.service.configuration.v1.api.model.DictionaryEntry;
-import com.iqser.red.service.configuration.v1.api.model.TypeResponse;
-import com.iqser.red.service.configuration.v1.api.model.TypeResult;
 import com.iqser.red.service.redaction.v1.server.client.DictionaryClient;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
-import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryIncrement;
+import com.iqser.red.service.redaction.v1.server.redaction.model.*;
-import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryIncrementValue;
-import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryModel;
-import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryRepresentation;
-import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryVersion;
-
 import feign.FeignException;
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
@@ -25,6 +16,7 @@ import java.awt.Color;
 import java.util.*;
 import java.util.stream.Collectors;
 
+
 @Slf4j
 @Service
 @RequiredArgsConstructor
@@ -32,37 +24,38 @@ public class DictionaryService {
 
     private final DictionaryClient dictionaryClient;
 
-    private final Map<String, DictionaryRepresentation> dictionariesByRuleSets = new HashMap<>();
+    private final Map<String, DictionaryRepresentation> dictionariesByDossierTemplate = new HashMap<>();
     private final Map<String, DictionaryRepresentation> dictionariesByDossier = new HashMap<>();
 
 
-    public DictionaryVersion updateDictionary(String ruleSetId, String dossierId) {
+    public DictionaryVersion updateDictionary(String dossierTemplateId, String dossierId) {
 
-        long rulesetDictionaryVersion = dictionaryClient.getVersion(ruleSetId, GLOBAL_DOSSIER);
+        log.info("Updating dictionary data for dossierTemplate {} and dossier {}", dossierTemplateId, dossierId);
-        var rulesetDictionary = dictionariesByRuleSets.get(ruleSetId);
+        long dossierTemplateDictionaryVersion = dictionaryClient.getVersion(dossierTemplateId);
-        if (rulesetDictionary == null || rulesetDictionaryVersion > rulesetDictionary.getDictionaryVersion()) {
+        var dossierTemplateDictionary = dictionariesByDossierTemplate.get(dossierTemplateId);
-            updateDictionaryEntry(ruleSetId, rulesetDictionaryVersion, GLOBAL_DOSSIER);
+        if (dossierTemplateDictionary == null || dossierTemplateDictionaryVersion > dossierTemplateDictionary.getDictionaryVersion()) {
+            updateDictionaryEntry(dossierTemplateId, dossierTemplateDictionaryVersion, null);
         }
 
-        long dossierDictionaryVersion = dictionaryClient.getVersion(ruleSetId, dossierId);
+        long dossierDictionaryVersion = dictionaryClient.getVersionForDossier(dossierId);
         var dossierDictionary = dictionariesByDossier.get(dossierId);
         if (dossierDictionary == null || dossierDictionaryVersion > dossierDictionary.getDictionaryVersion()) {
-            updateDictionaryEntry(ruleSetId, dossierDictionaryVersion, dossierId);
+            updateDictionaryEntry(dossierTemplateId, dossierDictionaryVersion, dossierId);
         }
 
-        return DictionaryVersion.builder().rulesetVersion(rulesetDictionaryVersion).dossierVersion(dossierDictionaryVersion).build();
+        return DictionaryVersion.builder().dossierTemplateVersion(dossierTemplateDictionaryVersion).dossierVersion(dossierDictionaryVersion).build();
     }
 
 
-    public DictionaryIncrement getDictionaryIncrements(String ruleSetId, DictionaryVersion fromVersion, String dossierId) {
+    public DictionaryIncrement getDictionaryIncrements(String dossierTemplateId, DictionaryVersion fromVersion, String dossierId) {
 
-        DictionaryVersion version = updateDictionary(ruleSetId, dossierId);
+        DictionaryVersion version = updateDictionary(dossierTemplateId, dossierId);
 
         Set<DictionaryIncrementValue> newValues = new HashSet<>();
-        List<DictionaryModel> dictionaryModels = dictionariesByRuleSets.get(ruleSetId).getDictionary();
+        List<DictionaryModel> dictionaryModels = dictionariesByDossierTemplate.get(dossierTemplateId).getDictionary();
         dictionaryModels.forEach(dictionaryModel -> {
             dictionaryModel.getEntries().forEach(dictionaryEntry -> {
-                if (dictionaryEntry.getVersion() > fromVersion.getRulesetVersion()) {
+                if (dictionaryEntry.getVersion() > fromVersion.getDossierTemplateVersion()) {
                     newValues.add(new DictionaryIncrementValue(dictionaryEntry.getValue(), dictionaryModel.isCaseInsensitive()));
                 }
             });
@@ -83,35 +76,35 @@ public class DictionaryService {
     }
 
 
-    private void updateDictionaryEntry(String ruleSetId, long version, String dossierId) {
+    private void updateDictionaryEntry(String dossierTemplateId, long version, String dossierId) {
 
         try {
             DictionaryRepresentation dictionaryRepresentation = new DictionaryRepresentation();
 
-            TypeResponse typeResponse = dictionaryClient.getAllTypes(ruleSetId, dossierId);
+            var typeResponse = dossierId == null ? dictionaryClient.getAllTypesForDossierTemplate(dossierTemplateId) : dictionaryClient.getAllTypesForDossier(dossierId);
-            if (typeResponse != null && CollectionUtils.isNotEmpty(typeResponse.getTypes())) {
+            if (typeResponse != null && CollectionUtils.isNotEmpty(typeResponse)) {
 
-                List<DictionaryModel> dictionary = typeResponse.getTypes()
+                List<DictionaryModel> dictionary = typeResponse
                         .stream()
                         .map(t -> new DictionaryModel(t.getType(), t.getRank(), convertColor(t.getHexColor()), t.isCaseInsensitive(), t
-                                .isHint(), t.isRecommendation(), convertEntries(t, dossierId), new HashSet<>(),dossierId.equals(GLOBAL_DOSSIER) ? false : true))
+                                .isHint(), t.isRecommendation(), convertEntries(t.getId()), new HashSet<>(), dossierId != null))
                         .sorted(Comparator.comparingInt(DictionaryModel::getRank).reversed())
                         .collect(Collectors.toList());
 
                 dictionary.forEach(dm -> dictionaryRepresentation.getLocalAccessMap().put(dm.getType(), dm));
 
-                Colors colors = dictionaryClient.getColors(ruleSetId);
+                Colors colors = dictionaryClient.getColors(dossierTemplateId);
 
                 dictionaryRepresentation.setDefaultColor(convertColor(colors.getDefaultColor()));
                 dictionaryRepresentation.setRequestAddColor(convertColor(colors.getRequestAdd()));
                 dictionaryRepresentation.setRequestRemoveColor(convertColor(colors.getRequestRemove()));
                 dictionaryRepresentation.setNotRedactedColor(convertColor(colors.getNotRedacted()));
-                dictionaryRepresentation.setRuleSetId(ruleSetId);
+                dictionaryRepresentation.setDossierTemplateId(dossierTemplateId);
                 dictionaryRepresentation.setDictionaryVersion(version);
                 dictionaryRepresentation.setDictionary(dictionary);
 
-                if(dossierId.equals(GLOBAL_DOSSIER)) {
+                if(dossierId == null) {
-                    dictionariesByRuleSets.put(ruleSetId, dictionaryRepresentation);
+                    dictionariesByDossierTemplate.put(dossierTemplateId, dictionaryRepresentation);
                 } else {
                     dictionariesByDossier.put(dossierId, dictionaryRepresentation);
                 }
@@ -123,26 +116,28 @@ public class DictionaryService {
     }
 
 
-    public void updateExternalDictionary(Dictionary dictionary, String ruleSetId) {
+    public void updateExternalDictionary(Dictionary dictionary, String dossierTemplateId) {
 
         dictionary.getDictionaryModels().forEach(dm -> {
             if (dm.isRecommendation() && !dm.getLocalEntries().isEmpty()) {
-                dictionaryClient.addEntries(dm.getType(), ruleSetId, new ArrayList<>(dm.getLocalEntries()), false, GLOBAL_DOSSIER);
+                dictionaryClient.addEntries(dm.getType(), new ArrayList<>(dm.getLocalEntries()), false);
-                long externalVersion = dictionaryClient.getVersion(ruleSetId, GLOBAL_DOSSIER);
+                long externalVersion = dictionaryClient.getVersion(dossierTemplateId);
-                if (externalVersion == dictionary.getVersion().getRulesetVersion() + 1) {
+                if (externalVersion == dictionary.getVersion().getDossierTemplateVersion() + 1) {
-                    dictionary.getVersion().setRulesetVersion(externalVersion);
+                    dictionary.getVersion().setDossierTemplateVersion(externalVersion);
                 }
             }
         });
     }
 
 
-    private Set<DictionaryEntry> convertEntries(TypeResult t, String dossierId) {
+    private Set<DictionaryEntry> convertEntries(String typeId) {
 
-        Set<DictionaryEntry> entries = new HashSet<>(dictionaryClient.getDictionaryForType(t.getType(), t.getRuleSetId(), dossierId)
+        var type = dictionaryClient.getDictionaryForType(typeId);
+
+        Set<DictionaryEntry> entries = new HashSet<>(type
                 .getEntries());
 
-        if (t.isCaseInsensitive()) {
+        if (type.isCaseInsensitive()) {
             entries.forEach(entry -> entry.setValue(entry.getValue().toLowerCase(Locale.ROOT)));
         }
         return entries;
@@ -156,9 +151,9 @@ public class DictionaryService {
     }
 
 
-    public boolean isCaseInsensitiveDictionary(String type, String ruleSetId) {
+    public boolean isCaseInsensitiveDictionary(String type, String dossierTemplateId) {
 
-        DictionaryModel dictionaryModel = dictionariesByRuleSets.get(ruleSetId).getLocalAccessMap().get(type);
+        DictionaryModel dictionaryModel = dictionariesByDossierTemplate.get(dossierTemplateId).getLocalAccessMap().get(type);
         if (dictionaryModel != null) {
             return dictionaryModel.isCaseInsensitive();
         }
@@ -166,19 +161,19 @@ public class DictionaryService {
     }
 
 
-    public float[] getColor(String type, String ruleSetId) {
+    public float[] getColor(String type, String dossierTemplateId) {
 
-        DictionaryModel model = dictionariesByRuleSets.get(ruleSetId).getLocalAccessMap().get(type);
+        DictionaryModel model = dictionariesByDossierTemplate.get(dossierTemplateId).getLocalAccessMap().get(type);
         if (model != null) {
             return model.getColor();
         }
-        return dictionariesByRuleSets.get(ruleSetId).getDefaultColor();
+        return dictionariesByDossierTemplate.get(dossierTemplateId).getDefaultColor();
     }
 
 
-    public boolean isHint(String type, String ruleSetId) {
+    public boolean isHint(String type, String dossierTemplateId) {
 
-        DictionaryModel model = dictionariesByRuleSets.get(ruleSetId).getLocalAccessMap().get(type);
+        DictionaryModel model = dictionariesByDossierTemplate.get(dossierTemplateId).getLocalAccessMap().get(type);
         if (model != null) {
             return model.isHint();
         }
@@ -186,9 +181,9 @@ public class DictionaryService {
     }
 
 
-    public boolean isRecommendation(String type, String ruleSetId) {
+    public boolean isRecommendation(String type, String dossierTemplateId) {
 
-        DictionaryModel model = dictionariesByRuleSets.get(ruleSetId).getLocalAccessMap().get(type);
+        DictionaryModel model = dictionariesByDossierTemplate.get(dossierTemplateId).getLocalAccessMap().get(type);
         if (model != null) {
             return model.isRecommendation();
         }
@@ -196,12 +191,12 @@ public class DictionaryService {
     }
 
 
-    public Dictionary getDeepCopyDictionary(String ruleSetId, String dossierId) {
+    public Dictionary getDeepCopyDictionary(String dossierTemplateId, String dossierId) {
 
         List<DictionaryModel> copy = new ArrayList<>();
 
-        var rulesetRepresentation = dictionariesByRuleSets.get(ruleSetId);
+        var dossierTemplateRepresentation = dictionariesByDossierTemplate.get(dossierTemplateId);
-        rulesetRepresentation.getDictionary().forEach(dm -> {
+        dossierTemplateRepresentation.getDictionary().forEach(dm -> {
             copy.add(SerializationUtils.clone(dm));
         });
 
@@ -215,25 +210,25 @@ public class DictionaryService {
             dossierDictionaryVersion = dossierRepresentation.getDictionaryVersion();
         }
 
-        return new Dictionary(copy, DictionaryVersion.builder().rulesetVersion(rulesetRepresentation.getDictionaryVersion()).dossierVersion(dossierDictionaryVersion).build());
+        return new Dictionary(copy.stream().sorted(Comparator.comparingInt(DictionaryModel::getRank).reversed()).collect(Collectors.toList()), DictionaryVersion.builder().dossierTemplateVersion(dossierTemplateRepresentation.getDictionaryVersion()).dossierVersion(dossierDictionaryVersion).build());
     }
 
 
-    public float[] getRequestRemoveColor(String ruleSetId) {
+    public float[] getRequestRemoveColor(String dossierTemplateId) {
 
-        return dictionariesByRuleSets.get(ruleSetId).getRequestAddColor();
+        return dictionariesByDossierTemplate.get(dossierTemplateId).getRequestRemoveColor();
     }
 
 
-    public float[] getNotRedactedColor(String ruleSetId) {
+    public float[] getNotRedactedColor(String dossierTemplateId) {
 
-        return dictionariesByRuleSets.get(ruleSetId).getNotRedactedColor();
+        return dictionariesByDossierTemplate.get(dossierTemplateId).getNotRedactedColor();
     }
 
 
-    public float[] getRequestAddColor(String ruleSetId) {
+    public float[] getRequestAddColor(String dossierTemplateId) {
 
-        return dictionariesByRuleSets.get(ruleSetId).getRequestAddColor();
+        return dictionariesByDossierTemplate.get(dossierTemplateId).getRequestAddColor();
     }
 
 }

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

@@ -1,6 +1,5 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;
 
-import com.iqser.red.service.configuration.v1.api.model.RulesResponse;
 import com.iqser.red.service.redaction.v1.server.client.RulesClient;
 import com.iqser.red.service.redaction.v1.server.exception.RulesValidationException;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Section;
@@ -28,14 +27,14 @@ public class DroolsExecutionService {
 
     private final Map<String, KieContainer> kieContainers = new HashMap<>();
 
-    private final Map<String, Long> rulesVersionPerRuleSetId = new HashMap<>();
+    private final Map<String, Long> rulesVersionPerDossierTemplateId = new HashMap<>();
 
 
-    public KieContainer getKieContainer(String ruleSetId) {
+    public KieContainer getKieContainer(String dossierTemplateId) {
 
-        KieContainer container = kieContainers.get(ruleSetId);
+        KieContainer container = kieContainers.get(dossierTemplateId);
         if (container == null) {
-            return createOrUpdateKieContainer(ruleSetId);
+            return createOrUpdateKieContainer(dossierTemplateId);
         } else {
             return container;
         }
@@ -55,43 +54,43 @@ public class DroolsExecutionService {
     }
 
 
-    public KieContainer updateRules(String ruleSetId) {
+    public KieContainer updateRules(String dossierTemplateId) {
 
-        long version = rulesClient.getVersion(ruleSetId);
+        long version = rulesClient.getVersion(dossierTemplateId);
-        Long rulesVersion = rulesVersionPerRuleSetId.get(ruleSetId);
+        Long rulesVersion = rulesVersionPerDossierTemplateId.get(dossierTemplateId);
         if (rulesVersion == null) {
             rulesVersion = -1L;
         }
 
-        if (version > rulesVersion.longValue()) {
+        if (version > rulesVersion) {
-            rulesVersionPerRuleSetId.put(ruleSetId, version);
+            rulesVersionPerDossierTemplateId.put(dossierTemplateId, version);
-            return createOrUpdateKieContainer(ruleSetId);
+            return createOrUpdateKieContainer(dossierTemplateId);
         }
-        return getKieContainer(ruleSetId);
+        return getKieContainer(dossierTemplateId);
 
     }
 
 
-    private KieContainer createOrUpdateKieContainer(String ruleSetId) {
+    private KieContainer createOrUpdateKieContainer(String dossierTemplateId) {
 
         try {
 
-            RulesResponse rules = rulesClient.getRules(ruleSetId);
+            var rules = rulesClient.getRules(dossierTemplateId);
-            if (rules == null || StringUtils.isEmpty(rules.getRules())) {
+            if (rules == null || StringUtils.isEmpty(rules.getValue())) {
                 throw new RuntimeException("Rules cannot be empty.");
             }
 
             KieServices kieServices = KieServices.Factory.get();
-            KieModule kieModule = getKieModule(ruleSetId, rules.getRules(), kieServices);
+            KieModule kieModule = getKieModule(dossierTemplateId, rules.getValue(), kieServices);
 
-            var container = kieContainers.get(ruleSetId);
+            var container = kieContainers.get(dossierTemplateId);
             if (container != null) {
                 container.updateToVersion(kieModule.getReleaseId());
                 return container;
             }
 
             container = kieServices.newKieContainer(kieModule.getReleaseId());
-            kieContainers.put(ruleSetId, container);
+            kieContainers.put(dossierTemplateId, container);
             return container;
         } catch (Exception e) {
             throw new RulesValidationException("Could not update rules: " + e.getMessage(), e);
@@ -100,11 +99,11 @@ public class DroolsExecutionService {
     }
 
 
-    private KieModule getKieModule(String ruleSetId, String rules, KieServices kieServices) {
+    private KieModule getKieModule(String dossierTemplateId, String rules, KieServices kieServices) {
 
         KieFileSystem kieFileSystem = kieServices.newKieFileSystem();
         InputStream input = new ByteArrayInputStream(rules.getBytes(StandardCharsets.UTF_8));
-        kieFileSystem.write("src/main/resources/drools/rules" + ruleSetId + ".drl", kieServices.getResources()
+        kieFileSystem.write("src/main/resources/drools/rules" + dossierTemplateId + ".drl", kieServices.getResources()
                 .newInputStreamResource(input));
         KieBuilder kieBuilder = kieServices.newKieBuilder(kieFileSystem);
         kieBuilder.buildAll();
@@ -122,13 +121,13 @@ public class DroolsExecutionService {
     }
 
 
-    public long getRulesVersion(String ruleSetId) {
+    public long getRulesVersion(String dossierTemplateId) {
 
-        Long rulesVersion = rulesVersionPerRuleSetId.get(ruleSetId);
+        Long rulesVersion = rulesVersionPerDossierTemplateId.get(dossierTemplateId);
         if (rulesVersion == null) {
             return -1;
         }
-        return rulesVersion.longValue();
+        return rulesVersion;
     }
 
 }

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

@@ -1,64 +1,143 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;
 
-import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
+import java.util.ArrayList;
-import com.iqser.red.service.redaction.v1.model.ManualRedactions;
+import java.util.HashMap;
-import com.iqser.red.service.redaction.v1.model.Point;
+import java.util.HashSet;
-import com.iqser.red.service.redaction.v1.model.Rectangle;
+import java.util.List;
-import com.iqser.red.service.redaction.v1.model.SectionArea;
+import java.util.Map;
-import com.iqser.red.service.redaction.v1.server.classification.model.*;
+import java.util.Set;
-import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
+import java.util.stream.Collectors;
-import com.iqser.red.service.redaction.v1.server.redaction.model.*;
+import java.util.stream.Stream;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.EntitySearchUtils;
+
-import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
+import org.apache.commons.codec.binary.Base64;
-import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
-import lombok.RequiredArgsConstructor;
-import lombok.extern.slf4j.Slf4j;
-import org.apache.commons.collections4.CollectionUtils;
 import org.apache.commons.lang3.StringUtils;
 import org.kie.api.runtime.KieContainer;
 import org.springframework.stereotype.Service;
 
-import java.util.*;
+import com.iqser.red.service.redaction.v1.model.AnalyzeRequest;
-import java.util.concurrent.atomic.AtomicInteger;
+import com.iqser.red.service.redaction.v1.model.Engine;
-import java.util.stream.Collectors;
+import com.iqser.red.service.redaction.v1.model.ManualImageRecategorization;
-import java.util.stream.Stream;
+import com.iqser.red.service.redaction.v1.model.Status;
+import com.iqser.red.service.redaction.v1.server.classification.model.SectionText;
+import com.iqser.red.service.redaction.v1.server.client.model.NerEntities;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
+import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryModel;
+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.Image;
+import com.iqser.red.service.redaction.v1.server.redaction.model.PageEntities;
+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.redaction.model.SectionSearchableTextPair;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.EntitySearchUtils;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.IdBuilder;
+import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
+
+import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;
 
 @Slf4j
 @Service
 @RequiredArgsConstructor
 public class EntityRedactionService {
 
-    private final DictionaryService dictionaryService;
+    private final RedactionServiceSettings redactionServiceSettings;
     private final DroolsExecutionService droolsExecutionService;
     private final SurroundingWordsService surroundingWordsService;
 
 
-    public void processDocument(Document classifiedDoc, String ruleSetId, ManualRedactions manualRedactions, String dossierId) {
+    public PageEntities findEntities(Dictionary dictionary, List<SectionText> sectionTexts, KieContainer kieContainer,
+                                     AnalyzeRequest analyzeRequest, NerEntities nerEntities) {
 
-        dictionaryService.updateDictionary(ruleSetId, dossierId);
+        Map<Integer, Set<Image>> imagesPerPage = new HashMap<>();
-        KieContainer container = droolsExecutionService.updateRules(ruleSetId);
+        Set<Entity> entities = findEntities(sectionTexts, dictionary, kieContainer, analyzeRequest, false, null, imagesPerPage, nerEntities);
-        long rulesVersion = droolsExecutionService.getRulesVersion(ruleSetId);
-
-        Dictionary dictionary = dictionaryService.getDeepCopyDictionary(ruleSetId, dossierId);
-
-        Set<Entity> documentEntities = new HashSet<>(findEntities(classifiedDoc, container, manualRedactions, dictionary, false, null));
 
         if (dictionary.hasLocalEntries()) {
+            Map<Integer, Set<Entity>> hintsPerSectionNumber = getHintsPerSection(entities, dictionary);
+            Set<Entity> foundByLocal = findEntities(sectionTexts, dictionary, kieContainer, analyzeRequest, true, hintsPerSectionNumber, imagesPerPage, nerEntities);
+            EntitySearchUtils.addEntitiesWithHigherRank(entities, foundByLocal, dictionary);
+            EntitySearchUtils.removeEntitiesContainedInLarger(entities);
+        }
 
-            Map<Integer, Set<Entity>> hintsPerSectionNumber = new HashMap<>();
+        Map<Integer, List<Entity>> entitiesPerPage = convertToEnititesPerPage(entities);
-            documentEntities.stream().forEach(entity -> {
+        return new PageEntities(entitiesPerPage, imagesPerPage);
-                if (dictionary.isHint(entity.getType()) && entity.isDictionaryEntry()) {
+    }
-                    hintsPerSectionNumber.computeIfAbsent(entity.getSectionNumber(), (x) -> new HashSet<>())
+
-                            .add(entity);
+
+    public Set<Entity> findEntities(List<SectionText> reanalysisSections, Dictionary dictionary,
+                                    KieContainer kieContainer, AnalyzeRequest analyzeRequest, boolean local,
+                                    Map<Integer, Set<Entity>> hintsPerSectionNumber,
+                                    Map<Integer, Set<Image>> imagesPerPage, NerEntities nerEntities) {
+
+        List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();
+        for (SectionText reanalysisSection : reanalysisSections) {
+
+            Set<Entity> entities = findEntities(reanalysisSection.getSearchableText(), reanalysisSection.getHeadline(), reanalysisSection
+                    .getSectionNumber(), dictionary, local, nerEntities, reanalysisSection.getCellStarts());
+
+            if (reanalysisSection.getCellStarts() != null && !reanalysisSection.getCellStarts().isEmpty()) {
+                surroundingWordsService.addSurroundingText(entities, reanalysisSection.getSearchableText(), dictionary, reanalysisSection
+                        .getCellStarts());
+            } else {
+                surroundingWordsService.addSurroundingText(entities, reanalysisSection.getSearchableText(), dictionary);
+            }
+
+            if (!local && reanalysisSection.getImages() != null && !reanalysisSection.getImages()
+                    .isEmpty() && analyzeRequest.getManualRedactions() != null && analyzeRequest.getManualRedactions()
+                    .getImageRecategorizations() != null) {
+                for (Image image : reanalysisSection.getImages()) {
+                    String imageId = IdBuilder.buildId(image.getPosition(), image.getPage());
+                    for (ManualImageRecategorization imageRecategorization : analyzeRequest.getManualRedactions()
+                            .getImageRecategorizations()) {
+                        if (imageRecategorization.getStatus().equals(Status.APPROVED) && imageRecategorization.getId()
+                                .equals(imageId)) {
+                            image.setType(imageRecategorization.getType());
+                        }
+                    }
+                }
+            }
+
+            sectionSearchableTextPairs.add(new SectionSearchableTextPair(Section.builder()
+                    .isLocal(false)
+                    .dictionaryTypes(dictionary.getTypes())
+                    .entities(hintsPerSectionNumber != null && hintsPerSectionNumber.containsKey(reanalysisSection.getSectionNumber()) ? Stream
+                            .concat(entities.stream(), hintsPerSectionNumber.get(reanalysisSection.getSectionNumber())
+                                    .stream())
+                            .collect(Collectors.toSet()) : entities)
+                    .text(reanalysisSection.getSearchableText().getAsStringWithLinebreaks())
+                    .searchText(reanalysisSection.getSearchableText().toString())
+                    .headline(reanalysisSection.getHeadline())
+                    .sectionNumber(reanalysisSection.getSectionNumber())
+                    .tabularData(reanalysisSection.getTabularData())
+                    .searchableText(reanalysisSection.getSearchableText())
+                    .dictionary(dictionary)
+                    .images(reanalysisSection.getImages())
+                    .fileAttributes(analyzeRequest.getFileAttributes())
+                    .build(), reanalysisSection.getSearchableText()));
+        }
+
+        Set<Entity> entities = new HashSet<>();
+        sectionSearchableTextPairs.forEach(sectionSearchableTextPair -> {
+            Section analysedSection = droolsExecutionService.executeRules(kieContainer, sectionSearchableTextPair.getSection());
+            EntitySearchUtils.removeEntitiesContainedInLarger(analysedSection.getEntities());
+            entities.addAll(analysedSection.getEntities());
+
+            if (!local) {
+                for (Image image : analysedSection.getImages()) {
+                    imagesPerPage.computeIfAbsent(image.getPage(), (a) -> new HashSet<>()).add(image);
+                }
+                addLocalValuesToDictionary(analysedSection, dictionary);
             }
         });
 
-            Set<Entity> foundByLocal = findEntities(classifiedDoc, container, manualRedactions, dictionary, true, hintsPerSectionNumber);
+        return entities;
-            EntitySearchUtils.addEntitiesWithHigherRank(documentEntities, foundByLocal, dictionary);
-            EntitySearchUtils.removeEntitiesContainedInLarger(documentEntities);
     }
 
-        for (Entity entity : documentEntities) {
+
+    private Map<Integer, List<Entity>> convertToEnititesPerPage(Set<Entity> entities) {
+
+        Map<Integer, List<Entity>> entitiesPerPage = new HashMap<>();
+        for (Entity entity : entities) {
             Map<Integer, List<EntityPositionSequence>> sequenceOnPage = new HashMap<>();
             for (EntityPositionSequence entityPositionSequence : entity.getPositionSequences()) {
                 sequenceOnPage.computeIfAbsent(entityPositionSequence.getPageNumber(), (x) -> new ArrayList<>())
@@ -66,80 +145,40 @@ public class EntityRedactionService {
             }
 
             for (Map.Entry<Integer, List<EntityPositionSequence>> entry : sequenceOnPage.entrySet()) {
-                classifiedDoc.getEntities()
+                entitiesPerPage.computeIfAbsent(entry.getKey(), (x) -> new ArrayList<>())
-                        .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(), entity
                                 .getLegalBasis(), entity.isDictionaryEntry(), entity.getTextBefore(), entity.getTextAfter(), entity
-                                .getStart(), entity.getEnd(), entity.isDossierDictionaryEntry()));
+                                .getStart(), entity.getEnd(), entity.isDossierDictionaryEntry(), entity.getEngines(), entity.getReferences()));
             }
         }
-
+        return entitiesPerPage;
-        dictionaryService.updateExternalDictionary(dictionary, ruleSetId);
-
-        classifiedDoc.setDictionaryVersion(dictionary.getVersion());
-        classifiedDoc.setRulesVersion(rulesVersion);
     }
 
 
-    private Set<Entity> findEntities(Document classifiedDoc, KieContainer kieContainer,
+    private Map<Integer, Set<Entity>> getHintsPerSection(Set<Entity> entities, Dictionary dictionary) {
-                                     ManualRedactions manualRedactions, Dictionary dictionary, boolean local,
-                                     Map<Integer, Set<Entity>> hintsPerSectionNumber) {
 
-        Set<Entity> documentEntities = new HashSet<>();
+        Map<Integer, Set<Entity>> hintsPerSectionNumber = new HashMap<>();
-
+        entities.stream().forEach(entity -> {
-        AtomicInteger sectionNumber = new AtomicInteger(1);
+            if (dictionary.isHint(entity.getType()) && entity.isDictionaryEntry()) {
-        List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();
+                hintsPerSectionNumber.computeIfAbsent(entity.getSectionNumber(), (x) -> new HashSet<>()).add(entity);
-        for (Paragraph paragraph : classifiedDoc.getParagraphs()) {
-
-            List<Table> tables = paragraph.getTables();
-            for (Table table : tables) {
-                if (table.getColCount() == 2) {
-                    sectionSearchableTextPairs.addAll(processTableAsOneText(classifiedDoc, table, manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
-                } else {
-                    sectionSearchableTextPairs.addAll(processTablePerRow(classifiedDoc, table, manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber));
             }
-                sectionNumber.incrementAndGet();
+        });
-            }
+        return hintsPerSectionNumber;
-            sectionSearchableTextPairs.add(processText(classifiedDoc, paragraph.getSearchableText(), paragraph.getTextBlocks(), paragraph
-                    .getHeadline(), manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber, paragraph
-                    .getImages()));
-            sectionNumber.incrementAndGet();
     }
 
-        for (Header header : classifiedDoc.getHeaders()) {
-            sectionSearchableTextPairs.add(processText(classifiedDoc, header.getSearchableText(), header.getTextBlocks(), "Header", manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber, new ArrayList<>()));
-            sectionNumber.incrementAndGet();
-        }
 
-        for (Footer footer : classifiedDoc.getFooters()) {
+    private void addLocalValuesToDictionary(Section analysedSection, Dictionary dictionary) {
-            sectionSearchableTextPairs.add(processText(classifiedDoc, footer.getSearchableText(), footer.getTextBlocks(), "Footer", manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber, new ArrayList<>()));
-            sectionNumber.incrementAndGet();
-        }
 
-        for (UnclassifiedText unclassifiedText : classifiedDoc.getUnclassifiedTexts()) {
+        analysedSection.getLocalDictionaryAdds().keySet().forEach(key -> {
-            sectionSearchableTextPairs.add(processText(classifiedDoc, unclassifiedText.getSearchableText(), unclassifiedText
-                    .getTextBlocks(), "", manualRedactions, sectionNumber, dictionary, local, hintsPerSectionNumber, new ArrayList<>()));
-            sectionNumber.incrementAndGet();
-        }
-
-        sectionSearchableTextPairs.forEach(sectionSearchableTextPair -> {
-            Section analysedRowSection = droolsExecutionService.executeRules(kieContainer, sectionSearchableTextPair.getSection());
-            documentEntities.addAll(analysedRowSection.getEntities());
-
-            for (Image image : analysedRowSection.getImages()) {
-                classifiedDoc.getImages().computeIfAbsent(image.getPage(), (a) -> new HashSet<>()).add(image);
-            }
-
-            analysedRowSection.getLocalDictionaryAdds().keySet().forEach(key -> {
             if (dictionary.isRecommendation(key)) {
-                    analysedRowSection.getLocalDictionaryAdds().get(key).forEach(value -> {
+                analysedSection.getLocalDictionaryAdds().get(key).forEach(value -> {
                     if (!dictionary.containsValue(key, value)) {
                         dictionary.getLocalAccessMap().get(key).getLocalEntries().add(value);
                     }
                 });
             } else {
-                    analysedRowSection.getLocalDictionaryAdds().get(key).forEach(value -> {
+                analysedSection.getLocalDictionaryAdds().get(key).forEach(value -> {
 
                     if (dictionary.getLocalAccessMap().get(key) == null) {
                         log.warn("Dictionary {} is null", key);
@@ -153,214 +192,12 @@ public class EntityRedactionService {
                 });
             }
         });
-
-        });
-
-        return documentEntities;
     }
 
 
-    private List<SectionSearchableTextPair> processTablePerRow(Document classifiedDoc, Table table,
+    private Set<Entity> findEntities(SearchableText searchableText, String headline, int sectionNumber,
-                                                               ManualRedactions manualRedactions,
+                                     Dictionary dictionary, boolean local, NerEntities nerEntities,
-                                                               AtomicInteger sectionNumber, Dictionary dictionary,
+                                     List<Integer> cellstarts) {
-                                                               boolean local,
-                                                               Map<Integer, Set<Entity>> hintsPerSectionNumber) {
-
-        List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();
-
-        for (List<Cell> row : table.getRows()) {
-            SearchableText searchableRow = new SearchableText();
-            Map<String, CellValue> tabularData = new HashMap<>();
-            int start = 0;
-            List<Integer> cellStarts = new ArrayList<>();
-            SectionText sectionText = new SectionText();
-            for (Cell cell : row) {
-
-                if (CollectionUtils.isEmpty(cell.getTextBlocks())) {
-                    continue;
-                }
-
-                SectionArea sectionArea = new SectionArea(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
-                        .getWidth(), (float) cell.getHeight(), cell.getTextBlocks()
-                        .get(0)
-                        .getSequences()
-                        .get(0)
-                        .getPage());
-                sectionText.getSectionAreas().add(sectionArea);
-                sectionText.getTextBlocks().addAll(cell.getTextBlocks());
-
-                addSectionToManualRedactions(cell.getTextBlocks(), manualRedactions, table.getHeadline(), sectionNumber.intValue());
-                int cellStart = start;
-
-                if (!cell.isHeaderCell()) {
-                    cell.getHeaderCells().forEach(headerCell -> {
-                        StringBuilder headerBuilder = new StringBuilder();
-                        headerCell.getTextBlocks().forEach(textBlock -> headerBuilder.append(textBlock.getText()));
-                        String headerName = headerBuilder.toString()
-                                .replaceAll("\n", "")
-                                .replaceAll(" ", "")
-                                .replaceAll("-", "");
-                        sectionArea.setHeader(headerName);
-                        tabularData.put(headerName, new CellValue(cell.getTextBlocks(), cellStart));
-                    });
-                }
-
-                for (TextBlock textBlock : cell.getTextBlocks()) {
-                    // TODO avoid cell overlap merging.
-                    searchableRow.addAll(textBlock.getSequences());
-                }
-                cellStarts.add(cellStart);
-                start = start + cell.toString().trim().length() + 1;
-
-            }
-            Set<Entity> rowEntities = findEntities(searchableRow, table.getHeadline(), sectionNumber.intValue(), dictionary, local);
-            surroundingWordsService.addSurroundingText(rowEntities, searchableRow, dictionary, cellStarts);
-
-            sectionSearchableTextPairs.add(new SectionSearchableTextPair(Section.builder()
-                    .isLocal(local)
-                    .dictionaryTypes(dictionary.getTypes())
-                    .entities(hintsPerSectionNumber != null && hintsPerSectionNumber.containsKey(sectionNumber.intValue()) ? Stream
-                            .concat(rowEntities.stream(), hintsPerSectionNumber.get(sectionNumber.intValue()).stream())
-                            .collect(Collectors.toSet()) : rowEntities)
-                    .text(searchableRow.getAsStringWithLinebreaks())
-                    .searchText(searchableRow.toString())
-                    .headline(table.getHeadline())
-                    .sectionNumber(sectionNumber.intValue())
-                    .tabularData(tabularData)
-                    .searchableText(searchableRow)
-                    .dictionary(dictionary)
-                    .build(), searchableRow));
-
-            if (!local) {
-                sectionText.setText(searchableRow.toString());
-                sectionText.setHeadline(table.getHeadline());
-                sectionText.setSectionNumber(sectionNumber.intValue());
-                sectionText.setTable(true);
-                sectionText.setTabularData(tabularData);
-                sectionText.setCellStarts(cellStarts);
-                classifiedDoc.getSectionText().add(sectionText);
-            }
-
-            sectionNumber.incrementAndGet();
-        }
-
-        return sectionSearchableTextPairs;
-    }
-
-
-    private List<SectionSearchableTextPair> processTableAsOneText(Document classifiedDoc, Table table,
-                                                                  ManualRedactions manualRedactions,
-                                                                  AtomicInteger sectionNumber, Dictionary dictionary,
-                                                                  boolean local,
-                                                                  Map<Integer, Set<Entity>> hintsPerSectionNumber) {
-
-        List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();
-        SearchableText entireTableText = new SearchableText();
-        SectionText sectionText = new SectionText();
-        for (List<Cell> row : table.getRows()) {
-            for (Cell cell : row) {
-                if (CollectionUtils.isEmpty(cell.getTextBlocks())) {
-                    continue;
-                }
-
-                if (!local) {
-                    SectionArea sectionArea = new SectionArea(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
-                            .getWidth(), (float) cell.getHeight(), cell.getTextBlocks()
-                            .get(0)
-                            .getSequences()
-                            .get(0)
-                            .getPage());
-                    sectionText.getTextBlocks().addAll(cell.getTextBlocks());
-                    sectionText.getSectionAreas().add(sectionArea);
-                }
-
-                for (TextBlock textBlock : cell.getTextBlocks()) {
-                    entireTableText.addAll(textBlock.getSequences());
-                }
-                addSectionToManualRedactions(cell.getTextBlocks(), manualRedactions, table.getHeadline(), sectionNumber.intValue());
-            }
-        }
-
-        Set<Entity> rowEntities = findEntities(entireTableText, table.getHeadline(), sectionNumber.intValue(), dictionary, local);
-        surroundingWordsService.addSurroundingText(rowEntities, entireTableText, dictionary);
-
-        sectionSearchableTextPairs.add(new SectionSearchableTextPair(Section.builder()
-                .isLocal(local)
-                .dictionaryTypes(dictionary.getTypes())
-                .entities(hintsPerSectionNumber != null && hintsPerSectionNumber.containsKey(sectionNumber.intValue()) ? Stream
-                        .concat(rowEntities.stream(), hintsPerSectionNumber.get(sectionNumber.intValue()).stream())
-                        .collect(Collectors.toSet()) : rowEntities)
-                .text(entireTableText.getAsStringWithLinebreaks())
-                .searchText(entireTableText.toString())
-                .headline(table.getHeadline())
-                .sectionNumber(sectionNumber.intValue())
-                .searchableText(entireTableText)
-                .dictionary(dictionary)
-                .build(), entireTableText));
-
-        if (!local) {
-            sectionText.setText(entireTableText.toString());
-            sectionText.setHeadline(table.getHeadline());
-            sectionText.setSectionNumber(sectionNumber.intValue());
-            sectionText.setTable(true);
-            classifiedDoc.getSectionText().add(sectionText);
-        }
-
-        return sectionSearchableTextPairs;
-    }
-
-
-    private SectionSearchableTextPair processText(Document classifiedDoc, SearchableText searchableText,
-                                                  List<TextBlock> paragraphTextBlocks, String headline,
-                                                  ManualRedactions manualRedactions, AtomicInteger sectionNumber,
-                                                  Dictionary dictionary, boolean local,
-                                                  Map<Integer, Set<Entity>> hintsPerSectionNumber,
-                                                  List<PdfImage> images) {
-
-        if (!local) {
-            SectionText sectionText = new SectionText();
-            for (TextBlock paragraphTextBlock : paragraphTextBlocks) {
-                SectionArea sectionArea = new SectionArea(new Point(paragraphTextBlock.getMinX(), paragraphTextBlock.getMinY()), paragraphTextBlock
-                        .getWidth(), paragraphTextBlock.getHeight(), paragraphTextBlock.getPage());
-                sectionText.getSectionAreas().add(sectionArea);
-            }
-
-            sectionText.setText(searchableText.toString());
-            sectionText.setHeadline(headline);
-            sectionText.setSectionNumber(sectionNumber.intValue());
-            sectionText.setTable(false);
-            sectionText.setImages(images.stream()
-                    .map(image -> convert(image, sectionNumber.intValue(), headline))
-                    .collect(Collectors.toSet()));
-            sectionText.setTextBlocks(paragraphTextBlocks);
-            classifiedDoc.getSectionText().add(sectionText);
-        }
-
-        addSectionToManualRedactions(paragraphTextBlocks, manualRedactions, headline, sectionNumber.intValue());
-        Set<Entity> entities = findEntities(searchableText, headline, sectionNumber.intValue(), dictionary, local);
-        surroundingWordsService.addSurroundingText(entities, searchableText, dictionary);
-
-        return new SectionSearchableTextPair(Section.builder()
-                .isLocal(local)
-                .dictionaryTypes(dictionary.getTypes())
-                .entities(hintsPerSectionNumber != null && hintsPerSectionNumber.containsKey(sectionNumber.intValue()) ? Stream
-                        .concat(entities.stream(), hintsPerSectionNumber.get(sectionNumber.intValue()).stream())
-                        .collect(Collectors.toSet()) : entities)
-                .text(searchableText.getAsStringWithLinebreaks())
-                .searchText(searchableText.toString())
-                .headline(headline)
-                .sectionNumber(sectionNumber.intValue())
-                .searchableText(searchableText)
-                .dictionary(dictionary)
-                .images(images.stream()
-                        .map(image -> convert(image, sectionNumber.intValue(), headline))
-                        .collect(Collectors.toSet()))
-                .build(), searchableText);
-    }
-
-
-    public Set<Entity> findEntities(SearchableText searchableText, String headline, int sectionNumber,
-                                    Dictionary dictionary, boolean local) {
 
         Set<Entity> found = new HashSet<>();
         String searchableString = searchableText.toString();
@@ -371,47 +208,52 @@ public class EntityRedactionService {
         String lowercaseInputString = searchableString.toLowerCase();
         for (DictionaryModel model : dictionary.getDictionaryModels()) {
             if (model.isCaseInsensitive()) {
-                found.addAll(EntitySearchUtils.find(lowercaseInputString, model.getValues(local), model.getType(), headline, sectionNumber, local, model.isDossierDictionary()));
+                EntitySearchUtils.addOrAddEngine(found, EntitySearchUtils.find(lowercaseInputString, model.getValues(local), model
+                        .getType(), headline, sectionNumber, !local, model.isDossierDictionary(), Engine.DICTIONARY));
             } else {
-                found.addAll(EntitySearchUtils.find(searchableString, model.getValues(local), model.getType(), headline, sectionNumber, local, model.isDossierDictionary()));
+                EntitySearchUtils.addOrAddEngine(found, EntitySearchUtils.find(searchableString, model.getValues(local), model
+                        .getType(), headline, sectionNumber, !local, model.isDossierDictionary(), Engine.DICTIONARY));
             }
         }
 
+        if (!local) {
+            Map<String, Set<String>> nerValuesPerType = getNerValues(sectionNumber, nerEntities, cellstarts);
+            nerValuesPerType.entrySet().forEach(entry -> {
+                EntitySearchUtils.addOrAddEngine(found, EntitySearchUtils.find(searchableString, entry.getValue(), entry
+                        .getKey(), headline, sectionNumber, false, false, Engine.NER));
+            });
+        }
+
         return EntitySearchUtils.clearAndFindPositions(found, searchableText, dictionary);
     }
 
 
-    private void addSectionToManualRedactions(List<TextBlock> textBlocks, ManualRedactions manualRedactions,
+    private Map<String, Set<String>> getNerValues(int sectionNumber, NerEntities nerEntities,
-                                              String section, int sectionNumber) {
+                                                  List<Integer> cellstarts) {
 
-        if (manualRedactions == null || manualRedactions.getEntriesToAdd().isEmpty()) {
+        Map<String, Set<String>> nerValuesPerType = new HashMap<>();
-            return;
-        }
 
-        for (TextBlock textBlock : textBlocks) {
+        if (redactionServiceSettings.isEnableEntityRecognition() && nerEntities.getResult()
-            for (ManualRedactionEntry manualRedactionEntry : manualRedactions.getEntriesToAdd()) {
+                .containsKey(sectionNumber)) {
-                for (Rectangle rectangle : manualRedactionEntry.getPositions()) {
+            nerEntities.getResult().get(sectionNumber).forEach(res -> {
-                    if (textBlock.contains(rectangle)) {
+                if (cellstarts == null || cellstarts.isEmpty()) {
-                        manualRedactionEntry.setSection(section);
+                    nerValuesPerType.computeIfAbsent(res.getType(), (a) -> new HashSet<>())
-                        manualRedactionEntry.setSectionNumber(sectionNumber);
+                            .add(new String(Base64.decodeBase64(res.getValue().getBytes())));
+                } else {
+                    boolean intersectsCellStart = false;
+                    for (Integer cellStart : cellstarts) {
+                        if (res.getStartOffset() < cellStart && cellStart < res.getEndOffset()) {
+                            intersectsCellStart = true;
                         }
                     }
+                    if (!intersectsCellStart) {
+                        nerValuesPerType.computeIfAbsent(res.getType(), (a) -> new HashSet<>())
+                                .add(new String(Base64.decodeBase64(res.getValue().getBytes())));
                     }
                 }
+            });
         }
-
+        return nerValuesPerType;
-
-    private Image convert(PdfImage pdfImage, int sectionNumber, String headline) {
-
-        return Image.builder()
-                .type(pdfImage.getImageType().equals(ImageType.OTHER) ? "image" : pdfImage.getImageType()
-                        .name()
-                        .toLowerCase(Locale.ROOT))
-                .position(pdfImage.getPosition())
-                .sectionNumber(sectionNumber)
-                .section(headline)
-                .page(pdfImage.getPage())
-                .build();
     }
 
 }

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

@@ -0,0 +1,51 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import java.util.stream.Collectors;
+
+import org.apache.commons.codec.binary.Base64;
+import org.springframework.stereotype.Service;
+
+import com.iqser.red.service.persistence.service.v1.api.model.FileType;
+import com.iqser.red.service.redaction.v1.server.client.EntityRecognitionClient;
+import com.iqser.red.service.redaction.v1.server.client.model.EntityRecognitionRequest;
+import com.iqser.red.service.redaction.v1.server.client.model.EntityRecognitionSection;
+import com.iqser.red.service.redaction.v1.server.settings.RedactionServiceSettings;
+import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
+
+import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
+@Service
+@RequiredArgsConstructor
+public class NerAnalyserService {
+
+    private final RedactionStorageService redactionStorageService;
+    private final EntityRecognitionClient entityRecognitionClient;
+    private final RedactionServiceSettings redactionServiceSettings;
+
+    public void computeNerEntities(String dossierId, String fileId) {
+
+        if (redactionServiceSettings.isEnableEntityRecognition()) {
+            var text = redactionStorageService.getText(dossierId, fileId);
+
+            long start = System.currentTimeMillis();
+
+            var nerRequest = EntityRecognitionRequest.builder()
+                    .data(text.getSectionTexts()
+                            .stream()
+                            .map(sectionText -> new EntityRecognitionSection(sectionText.getSectionNumber(), new String(Base64
+                                    .encodeBase64(sectionText
+                                    .getText().getBytes()))))
+                            .collect(Collectors.toList()))
+                    .build();
+
+            var nerResponse = entityRecognitionClient.findAuthors(nerRequest);
+
+            log.info("Computing NER entities took: {} ms for dossierId {} and fileId {}", System.currentTimeMillis() - start, dossierId, fileId);
+
+            redactionStorageService.storeObject(dossierId, fileId, FileType.NER_ENTITIES, nerResponse);
+        }
+    }
+
+}

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

@@ -1,296 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.redaction.service;
-
-import com.iqser.red.service.file.management.v1.api.model.FileType;
-import com.iqser.red.service.redaction.v1.model.*;
-import com.iqser.red.service.redaction.v1.server.classification.model.Document;
-import com.iqser.red.service.redaction.v1.server.classification.model.SectionText;
-import com.iqser.red.service.redaction.v1.server.classification.model.Text;
-import com.iqser.red.service.redaction.v1.server.client.LegalBasisClient;
-import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
-import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
-import com.iqser.red.service.redaction.v1.server.redaction.model.*;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.EntitySearchUtils;
-import com.iqser.red.service.redaction.v1.server.segmentation.PdfSegmentationService;
-import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
-
-import lombok.RequiredArgsConstructor;
-import lombok.SneakyThrows;
-import lombok.extern.slf4j.Slf4j;
-
-import org.kie.api.runtime.KieContainer;
-import org.springframework.stereotype.Service;
-import org.springframework.web.bind.annotation.RequestBody;
-
-import java.util.*;
-import java.util.stream.Collectors;
-import java.util.stream.Stream;
-
-@Slf4j
-@Service
-@RequiredArgsConstructor
-public class ReanalyzeService {
-
-    private final DictionaryService dictionaryService;
-    private final DroolsExecutionService droolsExecutionService;
-    private final SurroundingWordsService surroundingWordsService;
-    private final EntityRedactionService entityRedactionService;
-    private final RedactionLogCreatorService redactionLogCreatorService;
-    private final RedactionStorageService redactionStorageService;
-    private final PdfSegmentationService pdfSegmentationService;
-    private final RedactionChangeLogService redactionChangeLogService;
-    private final AnalyzeResponseService analyzeResponseService;
-    private final LegalBasisClient legalBasisClient;
-
-    public AnalyzeResult analyze(AnalyzeRequest analyzeRequest) {
-
-        long startTime = System.currentTimeMillis();
-
-        var pageCount = 0;
-        Document classifiedDoc;
-
-        try {
-            var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(analyzeRequest
-                    .getProjectId(), analyzeRequest.getFileId(), FileType.ORIGIN));
-            classifiedDoc = pdfSegmentationService.parseDocument(storedObjectStream);
-            pageCount = classifiedDoc.getPages().size();
-        } catch (Exception e) {
-            throw new RedactionException(e);
-        }
-        log.info("Document structure analysis successful, starting redaction analysis...");
-
-        entityRedactionService.processDocument(classifiedDoc, analyzeRequest.getRuleSetId(), analyzeRequest.getManualRedactions(), analyzeRequest
-                .getProjectId());
-        redactionLogCreatorService.createRedactionLog(classifiedDoc, pageCount, analyzeRequest.getManualRedactions(), analyzeRequest
-                .getRuleSetId());
-
-        log.info("Redaction analysis successful...");
-
-        var legalBasis = legalBasisClient.getLegalBasisMapping(analyzeRequest.getRuleSetId());
-        var redactionLog = new RedactionLog(classifiedDoc.getRedactionLogEntities(),legalBasis,
-                analyzeRequest.getRuleSetId(),
-                classifiedDoc.getDictionaryVersion().getRulesetVersion(),
-                classifiedDoc.getRulesVersion(),
-                classifiedDoc.getDictionaryVersion().getDossierVersion(),
-                legalBasisClient.getVersion(analyzeRequest.getRuleSetId()));
-
-        log.info("Analyzed with rules {} and dictionary {} for ruleSet: {}", classifiedDoc.getRulesVersion(), classifiedDoc
-                .getDictionaryVersion(), analyzeRequest.getRuleSetId());
-
-        // first create changelog - this only happens when we migrate files analyzed via the old process and we don't want to loose changeLog data
-        var changeLog = redactionChangeLogService.createAndStoreChangeLog(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), redactionLog);
-        // store redactionLog
-        redactionStorageService.storeObject(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), FileType.REDACTION_LOG, redactionLog);
-        redactionStorageService.storeObject(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), FileType.TEXT, new Text(pageCount, classifiedDoc
-                .getSectionText()));
-        redactionStorageService.storeObject(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), FileType.SECTION_GRID, classifiedDoc
-                .getSectionGrid());
-
-        long duration = System.currentTimeMillis() - startTime;
-        return analyzeResponseService.createAnalyzeResponse(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), duration, pageCount, redactionLog, changeLog);
-    }
-
-
-    @SneakyThrows
-    public AnalyzeResult reanalyze(@RequestBody AnalyzeRequest analyzeRequest) {
-
-        long startTime = System.currentTimeMillis();
-
-        var redactionLog = redactionStorageService.getRedactionLog(analyzeRequest.getProjectId(), analyzeRequest.getFileId());
-        var text = redactionStorageService.getText(analyzeRequest.getProjectId(), analyzeRequest.getFileId());
-
-        // not yet ready for reanalysis
-        if (redactionLog == null || text == null || text.getNumberOfPages() == 0) {
-            return analyze(analyzeRequest);
-        }
-
-        DictionaryIncrement dictionaryIncrement = dictionaryService.getDictionaryIncrements(analyzeRequest.getRuleSetId(), new DictionaryVersion(redactionLog
-                .getDictionaryVersion(), redactionLog.getDossierDictionaryVersion()), analyzeRequest.getProjectId());
-
-        Set<String> manualForceAndRemoveIds = getForceAndRemoveIds(analyzeRequest.getManualRedactions());
-        Map<String, List<Comment>> comments = null;
-        Set<ManualRedactionEntry> manualAdds = null;
-
-        if (analyzeRequest.getManualRedactions() != null) {
-            // TODO comments will be removed from redactionLog, so we ignore this first.
-            comments = analyzeRequest.getManualRedactions().getComments();
-            manualAdds = analyzeRequest.getManualRedactions().getEntriesToAdd();
-        }
-
-        Set<Integer> sectionsToReanalyse = new HashSet<>();
-        Map<Integer, Set<Image>> imageEntries = new HashMap<>();
-        for (RedactionLogEntry entry : redactionLog.getRedactionLogEntry()) {
-            if (entry.isManual() || manualForceAndRemoveIds.contains(entry.getId())) {
-                sectionsToReanalyse.add(entry.getSectionNumber());
-            }
-            if (entry.isImage() || entry.getType().equals("image")) {
-                imageEntries.computeIfAbsent(entry.getSectionNumber(), x -> new HashSet<>()).add(convert(entry));
-            }
-        }
-
-        for (SectionText sectionText : text.getSectionTexts()) {
-
-            if (EntitySearchUtils.sectionContainsAny(sectionText.getText(), dictionaryIncrement.getValues())) {
-                sectionsToReanalyse.add(sectionText.getSectionNumber());
-            }
-
-            if (manualAdds != null) {
-                for (SectionArea sectionArea : sectionText.getSectionAreas()) {
-                    for (ManualRedactionEntry manualAdd : manualAdds) {
-                        for (Rectangle manualPosition : manualAdd.getPositions()) {
-                            if (sectionArea.contains(manualPosition)) {
-                                manualAdd.setSection(sectionText.getHeadline());
-                                manualAdd.setSectionNumber(sectionText.getSectionNumber());
-                            }
-                        }
-                    }
-                }
-            }
-        }
-
-        log.info("Should reanalyze {} sections for request: {}", sectionsToReanalyse.size(), analyzeRequest);
-
-        if (sectionsToReanalyse.isEmpty() && (manualAdds == null || manualAdds.isEmpty())) {
-            return finalizeAnalysis(analyzeRequest, startTime, redactionLog, text, dictionaryIncrement);
-        }
-
-        List<SectionText> reanalysisSections = new ArrayList<>();
-
-        for (SectionText sectionText : text.getSectionTexts()) {
-
-            if (sectionsToReanalyse.contains(sectionText.getSectionNumber())) {
-                reanalysisSections.add(sectionText);
-            }
-        }
-
-        //--
-
-        KieContainer kieContainer = droolsExecutionService.updateRules(analyzeRequest.getRuleSetId());
-
-        Dictionary dictionary = dictionaryService.getDeepCopyDictionary(analyzeRequest.getRuleSetId(), analyzeRequest.getProjectId());
-
-        List<SectionSearchableTextPair> sectionSearchableTextPairs = new ArrayList<>();
-        for (SectionText reanalysisSection : reanalysisSections) {
-
-            Set<Entity> entities = entityRedactionService.findEntities(reanalysisSection.getSearchableText(), reanalysisSection
-                    .getHeadline(), reanalysisSection.getSectionNumber(), dictionary, false);
-            if (reanalysisSection.getCellStarts() != null) {
-                surroundingWordsService.addSurroundingText(entities, reanalysisSection.getSearchableText(), dictionary, reanalysisSection
-                        .getCellStarts());
-            } else {
-                surroundingWordsService.addSurroundingText(entities, reanalysisSection.getSearchableText(), dictionary);
-            }
-
-            sectionSearchableTextPairs.add(new SectionSearchableTextPair(Section.builder()
-                    .isLocal(false)
-                    .dictionaryTypes(dictionary.getTypes())
-                    .entities(entities)
-                    .text(reanalysisSection.getSearchableText().getAsStringWithLinebreaks())
-                    .searchText(reanalysisSection.getSearchableText().toString())
-                    .headline(reanalysisSection.getHeadline())
-                    .sectionNumber(reanalysisSection.getSectionNumber())
-                    .tabularData(reanalysisSection.getTabularData())
-                    .searchableText(reanalysisSection.getSearchableText())
-                    .dictionary(dictionary)
-                    .images(reanalysisSection.getImages())
-                    .build(), reanalysisSection.getSearchableText()));
-        }
-
-        Set<Entity> entities = new HashSet<>();
-        Map<Integer, Set<Image>> imagesPerPage = new HashMap<>();
-        sectionSearchableTextPairs.forEach(sectionSearchableTextPair -> {
-            Section analysedRowSection = droolsExecutionService.executeRules(kieContainer, sectionSearchableTextPair.getSection());
-            entities.addAll(analysedRowSection.getEntities());
-            EntitySearchUtils.removeEntitiesContainedInLarger(entities);
-
-            for (Image image : analysedRowSection.getImages()) {
-                imagesPerPage.computeIfAbsent(image.getPage(), (a) -> new HashSet<>()).add(image);
-            }
-
-        });
-
-        Map<Integer, List<Entity>> entitiesPerPage = new HashMap<>();
-        for (Entity entity : entities) {
-            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()) {
-                entitiesPerPage.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(), entity
-                                .getLegalBasis(), entity.isDictionaryEntry(), entity.getTextBefore(), entity.getTextAfter(), entity
-                                .getStart(), entity.getEnd(), entity.isDossierDictionaryEntry()));
-            }
-        }
-
-        List<RedactionLogEntry> newRedactionLogEntries = new ArrayList<>();
-        for (int page = 1; page <= text.getNumberOfPages(); page++) {
-            if (entitiesPerPage.get(page) != null) {
-                newRedactionLogEntries.addAll(redactionLogCreatorService.addEntries(entitiesPerPage, analyzeRequest.getManualRedactions(), page, analyzeRequest
-                        .getRuleSetId()));
-            }
-
-            if (imagesPerPage.get(page) != null) {
-                newRedactionLogEntries.addAll(redactionLogCreatorService.addImageEntries(imagesPerPage, analyzeRequest.getManualRedactions(), page, analyzeRequest
-                        .getRuleSetId()));
-            }
-
-            newRedactionLogEntries.addAll(redactionLogCreatorService.addManualAddEntries(manualAdds, comments, page, analyzeRequest
-                    .getRuleSetId()));
-        }
-
-        redactionLog.getRedactionLogEntry()
-                .removeIf(entry -> sectionsToReanalyse.contains(entry.getSectionNumber()));
-        redactionLog.getRedactionLogEntry().addAll(newRedactionLogEntries);
-        return finalizeAnalysis(analyzeRequest, startTime, redactionLog, text, dictionaryIncrement);
-
-    }
-
-
-    private AnalyzeResult finalizeAnalysis(@RequestBody AnalyzeRequest analyzeRequest, long startTime,
-                                           RedactionLog redactionLog, Text text,
-                                           DictionaryIncrement dictionaryIncrement) {
-
-        redactionLog.setDictionaryVersion(dictionaryIncrement.getDictionaryVersion().getRulesetVersion());
-        redactionLog.setDossierDictionaryVersion(dictionaryIncrement.getDictionaryVersion().getDossierVersion());
-
-        var changeLog = redactionChangeLogService.createAndStoreChangeLog(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), redactionLog);
-        redactionStorageService.storeObject(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), FileType.REDACTION_LOG, redactionLog);
-
-        long duration = System.currentTimeMillis() - startTime;
-
-        return analyzeResponseService.createAnalyzeResponse(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), duration, text
-                .getNumberOfPages(), redactionLog, changeLog);
-    }
-
-
-    private Set<String> getForceAndRemoveIds(ManualRedactions manualRedactions) {
-
-        if (manualRedactions == null) {
-            return new HashSet<>();
-        }
-
-        return Stream.concat(manualRedactions.getIdsToRemove()
-                .stream()
-                .map(IdRemoval::getId), manualRedactions.getForceRedacts().stream().map(ManualForceRedact::getId))
-                .collect(Collectors.toSet());
-    }
-
-
-    public Image convert(RedactionLogEntry entry) {
-
-        Rectangle position = entry.getPositions().get(0);
-
-        return Image.builder()
-                .type(entry.getType())
-                .position(new RedRectangle2D(position.getTopLeft().getX(), position.getTopLeft()
-                        .getY(), position.getWidth(), position.getHeight()))
-                .sectionNumber(entry.getSectionNumber())
-                .section(entry.getSection())
-                .page(position.getPage())
-                .build();
-    }
-
-}

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

@@ -1,19 +1,25 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;
 
-import com.iqser.red.service.file.management.v1.api.model.FileType;
+import java.time.OffsetDateTime;
-import com.iqser.red.service.redaction.v1.model.ChangeType;
+import java.util.ArrayList;
-import com.iqser.red.service.redaction.v1.model.RedactionChangeLog;
+import java.util.HashMap;
-import com.iqser.red.service.redaction.v1.model.RedactionChangeLogEntry;
+import java.util.HashSet;
-import com.iqser.red.service.redaction.v1.model.RedactionLog;
+import java.util.List;
-import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import java.util.Map;
-import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
+import java.util.Set;
-import lombok.RequiredArgsConstructor;
+import java.util.stream.Collectors;
-import lombok.extern.slf4j.Slf4j;
+
 import org.springframework.stereotype.Service;
 
-import java.util.ArrayList;
+import com.iqser.red.service.redaction.v1.model.Change;
-import java.util.List;
+import com.iqser.red.service.redaction.v1.model.ChangeType;
-import java.util.stream.Collectors;
+import com.iqser.red.service.redaction.v1.model.RedactionLog;
+import com.iqser.red.service.redaction.v1.model.RedactionLogChanges;
+import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
+
+import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;
 
 @Slf4j
 @Service
@@ -22,72 +28,80 @@ public class RedactionChangeLogService {
 
     private final RedactionStorageService redactionStorageService;
 
-    public RedactionChangeLog createAndStoreChangeLog(String projectId, String fileId, RedactionLog currentRedactionLog) {
 
-        try {
+    public RedactionLogChanges computeChanges(String dossierId, String fileId, RedactionLog currentRedactionLog) {
-            RedactionLog previousRedactionLog = redactionStorageService.getRedactionLog(projectId, fileId);
-            var changeLog = createChangeLog(currentRedactionLog, previousRedactionLog);
-            redactionStorageService.storeObject(projectId, fileId, FileType.REDACTION_CHANGELOG, changeLog);
-            return changeLog;
-        } catch (Exception e) {
-            log.debug("Previous redaction log not available");
-            return null;
-        }
 
-    }
+        long start = System.currentTimeMillis();
-
-
-    private RedactionChangeLog createChangeLog(RedactionLog currentRedactionLog, RedactionLog previousRedactionLog) {
 
+        RedactionLog previousRedactionLog = redactionStorageService.getRedactionLog(dossierId, fileId);
 
         if (previousRedactionLog == null) {
-            return null;
+            currentRedactionLog.getRedactionLogEntry().forEach(entry -> {
+                entry.getChanges().add(new Change(ChangeType.ADDED, OffsetDateTime.now()));
+            });
+            return new RedactionLogChanges(currentRedactionLog, false);
         }
 
-        List<RedactionLogEntry> added = new ArrayList<>(currentRedactionLog.getRedactionLogEntry());
+        List<RedactionLogEntry> notRemovedPreviousEntries = previousRedactionLog.getRedactionLogEntry()
-        added.removeAll(previousRedactionLog.getRedactionLogEntry());
+                .stream()
+                .filter(entry -> !entry.getChanges()
+                        .get(entry.getChanges().size() - 1)
+                        .getType()
+                        .equals(ChangeType.REMOVED))
+                .collect(Collectors.toList());
 
-        List<RedactionLogEntry> removed = new ArrayList<>(previousRedactionLog.getRedactionLogEntry());
+        Set<RedactionLogEntry> added = new HashSet<>(currentRedactionLog.getRedactionLogEntry());
+        added.removeAll(notRemovedPreviousEntries);
+
+        Set<RedactionLogEntry> removed = new HashSet<>(notRemovedPreviousEntries);
         removed.removeAll(currentRedactionLog.getRedactionLogEntry());
 
-        List<RedactionChangeLogEntry> changeLogEntries = added.stream()
+        Map<String, RedactionLogEntry> addedIds = new HashMap<>();
-                .map(entry -> convert(entry, ChangeType.ADDED))
+        added.forEach(entry -> {
-                .collect(Collectors.toList());
+            addedIds.put(entry.getId(), entry);
-        changeLogEntries.addAll(removed.stream()
+        });
-                .map(entry -> convert(entry, ChangeType.REMOVED))
-                .collect(Collectors.toList()));
 
-        return new RedactionChangeLog(changeLogEntries, currentRedactionLog.getDictionaryVersion(), currentRedactionLog.getRulesVersion(), currentRedactionLog
+        Set<String> removedIds = new HashSet<>();
-                .getRuleSetId());
+        removed.forEach(entry -> {
+            removedIds.add(entry.getId());
+        });
+
+        List<RedactionLogEntry> newRedactionLogEntries = previousRedactionLog.getRedactionLogEntry();
+
+        List<RedactionLogEntry> toRemove = new ArrayList<>();
+        newRedactionLogEntries.forEach(entry -> {
+            if (removedIds.contains(entry.getId()) && addedIds.containsKey(entry.getId())) {
+                List<Change> changes = entry.getChanges();
+                changes.add(new Change(ChangeType.CHANGED, OffsetDateTime.now()));
+                var newEntry = addedIds.get(entry.getId());
+                newEntry.setChanges(changes);
+                addedIds.put(entry.getId(), newEntry);
+                toRemove.add(entry);
+            } else if (removedIds.contains(entry.getId())) {
+                entry.getChanges().add(new Change(ChangeType.REMOVED, OffsetDateTime.now()));
+            } else if (addedIds.containsKey(entry.getId())) {
+                List<Change> changes = entry.getChanges();
+                changes.add(new Change(ChangeType.ADDED, OffsetDateTime.now()));
+                var newEntry = addedIds.get(entry.getId());
+                newEntry.setChanges(changes);
+                addedIds.put(entry.getId(), newEntry);
+                toRemove.add(entry);
             }
+        });
 
+        newRedactionLogEntries.removeAll(toRemove);
 
-    private RedactionChangeLogEntry convert(RedactionLogEntry entry, ChangeType changeType) {
+        addedIds.forEach((k, v) -> {
+            if(v.getChanges().isEmpty()) {
+                v.getChanges().add(new Change(ChangeType.ADDED, OffsetDateTime.now()));
+            }
+            newRedactionLogEntries.add(v);
+        });
 
-        return RedactionChangeLogEntry.builder()
+        currentRedactionLog.setRedactionLogEntry(newRedactionLogEntries);
-                .id(entry.getId())
+
-                .type(entry.getType())
+        log.info("Change computation took: {}", System.currentTimeMillis() - start);
-                .value(entry.getValue())
+        return new RedactionLogChanges(currentRedactionLog, !addedIds.isEmpty() || !removedIds.isEmpty());
-                .reason(entry.getReason())
-                .matchedRule(entry.getMatchedRule())
-                .legalBasis(entry.getLegalBasis())
-                .redacted(entry.isRedacted())
-                .isHint(entry.isHint())
-                .isRecommendation(entry.isRecommendation())
-                .section(entry.getSection())
-                .color(entry.getColor())
-                .positions(entry.getPositions())
-                .sectionNumber(entry.getSectionNumber())
-                .manual(entry.isManual())
-                .status(entry.getStatus())
-                .manualRedactionType(entry.getManualRedactionType())
-                .isDictionaryEntry(entry.isDictionaryEntry())
-                .textBefore(entry.getTextBefore())
-                .textAfter(entry.getTextAfter())
-                .comments(entry.getComments())
-                .changeType(changeType)
-                .isDossierDictionaryEntry(entry.isDossierDictionaryEntry())
-                .build();
     }
 
 }

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

@@ -1,22 +1,5 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;
 
-import com.iqser.red.service.redaction.v1.model.*;
-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.parsing.model.RedTextPosition;
-import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
-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.Image;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.IdBuilder;
-import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
-import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
-import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
-import lombok.RequiredArgsConstructor;
-import org.apache.commons.collections4.CollectionUtils;
-import org.springframework.stereotype.Service;
-
 import java.util.ArrayList;
 import java.util.HashSet;
 import java.util.List;
@@ -24,6 +7,22 @@ import java.util.Map;
 import java.util.Set;
 import java.util.stream.Collectors;
 
+import org.apache.commons.collections4.CollectionUtils;
+import org.springframework.stereotype.Service;
+
+import com.iqser.red.service.redaction.v1.model.Point;
+import com.iqser.red.service.redaction.v1.model.Rectangle;
+import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import com.iqser.red.service.redaction.v1.server.parsing.model.RedTextPosition;
+import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
+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.Image;
+import com.iqser.red.service.redaction.v1.server.redaction.model.PageEntities;
+import com.iqser.red.service.redaction.v1.server.redaction.utils.IdBuilder;
+
+import lombok.RequiredArgsConstructor;
+
 @Service
 @RequiredArgsConstructor
 public class RedactionLogCreatorService {
@@ -31,35 +30,27 @@ public class RedactionLogCreatorService {
     private final DictionaryService dictionaryService;
 
 
-    public void createRedactionLog(Document classifiedDoc, int numberOfPages, ManualRedactions manualRedactions,
+    public List<RedactionLogEntry> createRedactionLog(PageEntities pageEntities, int numberOfPages,
-                                   String ruleSetId) {
+                                                      String dossierTemplateId) {
 
-        Set<Integer> manualRedactionPages = getManualRedactionPages(manualRedactions);
+        List<RedactionLogEntry> entries = new ArrayList<>();
 
         for (int page = 1; page <= numberOfPages; page++) {
-
+            if (pageEntities.getEntitiesPerPage().get(page) != null) {
-            addSectionGrid(classifiedDoc, page);
+                entries.addAll(addEntries(pageEntities.getEntitiesPerPage(), page, dossierTemplateId));
-
-            if (classifiedDoc.getEntities().get(page) != null) {
-                classifiedDoc.getRedactionLogEntities()
-                        .addAll(addEntries(classifiedDoc.getEntities(), manualRedactions, page, ruleSetId));
             }
 
-            if (manualRedactionPages.contains(page)) {
+            if (pageEntities.getImagesPerPage().get(page) != null) {
-                classifiedDoc.getRedactionLogEntities()
+                entries.addAll(addImageEntries(pageEntities.getImagesPerPage(), page, dossierTemplateId));
-                        .addAll(addManualAddEntries(manualRedactions.getEntriesToAdd(), manualRedactions.getComments(), page, ruleSetId));
+            }
         }
 
-            if (classifiedDoc.getImages().get(page) != null && !classifiedDoc.getImages().get(page).isEmpty()) {
+        return entries;
-                classifiedDoc.getRedactionLogEntities()
-                        .addAll(addImageEntries(classifiedDoc.getImages(), manualRedactions, page, ruleSetId));
-            }
-        }
     }
 
 
-    public List<RedactionLogEntry> addImageEntries(Map<Integer, Set<Image>> images, ManualRedactions manualRedactions,
+    public List<RedactionLogEntry> addImageEntries(Map<Integer, Set<Image>> images, int pageNumber,
-                                                   int pageNumber, String ruleSetId) {
+                                                   String dossierTemplateId) {
 
         List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
 
@@ -69,14 +60,14 @@ public class RedactionLogCreatorService {
 
             RedactionLogEntry redactionLogEntry = RedactionLogEntry.builder()
                     .id(id)
-                    .color(getColorForImage(image, ruleSetId, false))
+                    .color(getColor(image.getType(), dossierTemplateId, image.isRedaction()))
                     .isImage(true)
                     .type(image.getType())
                     .redacted(image.isRedaction())
                     .reason(image.getRedactionReason())
                     .legalBasis(image.getLegalBasis())
                     .matchedRule(image.getMatchedRule())
-                    .isHint(dictionaryService.isHint(image.getType(), ruleSetId))
+                    .isHint(dictionaryService.isHint(image.getType(), dossierTemplateId))
                     .manual(false)
                     .isDictionaryEntry(false)
                     .isRecommendation(false)
@@ -85,62 +76,9 @@ public class RedactionLogCreatorService {
                             .getWidth(), (float) image.getPosition().getHeight(), pageNumber)))
                     .sectionNumber(image.getSectionNumber())
                     .section(image.getSection())
+                    .imageHasTransparency(image.isHasTransparency())
                     .build();
 
-            if (manualRedactions != null && !manualRedactions.getIdsToRemove().isEmpty()) {
-                for (IdRemoval manualRemoval : manualRedactions.getIdsToRemove()) {
-                    if (manualRemoval.getId().equals(id)) {
-                        String manualOverrideReason = null;
-                        if (manualRemoval.getStatus().equals(Status.APPROVED)) {
-                            image.setRedaction(false);
-                            redactionLogEntry.setRedacted(false);
-                            redactionLogEntry.setStatus(Status.APPROVED);
-                            manualOverrideReason = image.getRedactionReason() + ", removed by manual override";
-                            redactionLogEntry.setColor(getColorForImage(image, ruleSetId, false));
-                        } else if (manualRemoval.getStatus().equals(Status.REQUESTED)) {
-                            manualOverrideReason = image.getRedactionReason() + ", requested to remove";
-                            redactionLogEntry.setStatus(Status.REQUESTED);
-                            redactionLogEntry.setColor(getColorForImage(image, ruleSetId, true));
-                        } else {
-                            redactionLogEntry.setStatus(Status.DECLINED);
-                        }
-
-                        image.setRedactionReason(manualOverrideReason != null ? manualOverrideReason : image.getRedactionReason());
-                        redactionLogEntry.setReason(manualOverrideReason);
-                        redactionLogEntry.setManual(true);
-                        redactionLogEntry.setManualRedactionType(ManualRedactionType.REMOVE);
-                    }
-                }
-            }
-
-            if (manualRedactions != null && !manualRedactions.getForceRedacts().isEmpty()) {
-                for (ManualForceRedact manualForceRedact : manualRedactions.getForceRedacts()) {
-                    if (manualForceRedact.getId().equals(id)) {
-                        String manualOverrideReason = null;
-                        if (manualForceRedact.getStatus().equals(Status.APPROVED)) {
-                            image.setRedaction(true);
-                            redactionLogEntry.setRedacted(true);
-                            redactionLogEntry.setStatus(Status.APPROVED);
-                            redactionLogEntry.setColor(getColorForImage(image, ruleSetId, false));
-                            manualOverrideReason = image.getRedactionReason() + ", forced by manual override";
-                            redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
-                        } else if (manualForceRedact.getStatus().equals(Status.REQUESTED)) {
-                            manualOverrideReason = image.getRedactionReason() + ", requested to force redact";
-                            redactionLogEntry.setStatus(Status.REQUESTED);
-                            redactionLogEntry.setColor(getColorForImage(image, ruleSetId, true));
-                            redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
-                        } else {
-                            redactionLogEntry.setStatus(Status.DECLINED);
-                        }
-
-                        image.setRedactionReason(manualOverrideReason != null ? manualOverrideReason : image.getRedactionReason());
-                        redactionLogEntry.setReason(manualOverrideReason);
-                        redactionLogEntry.setManual(true);
-                        redactionLogEntry.setManualRedactionType(ManualRedactionType.FORCE_REDACT);
-                    }
-                }
-            }
-
             redactionLogEntities.add(redactionLogEntry);
         }
 
@@ -148,25 +86,7 @@ public class RedactionLogCreatorService {
     }
 
 
-    private Set<Integer> getManualRedactionPages(ManualRedactions manualRedactions) {
+    public List<RedactionLogEntry> addEntries(Map<Integer, List<Entity>> entities, int page, String dossierTemplateId) {
-
-        Set<Integer> manualRedactionPages = new HashSet<>();
-
-        if (manualRedactions == null) {
-            return manualRedactionPages;
-        }
-
-        manualRedactions.getEntriesToAdd().forEach(entry -> {
-            entry.getPositions().forEach(pos -> {
-                manualRedactionPages.add(pos.getPage());
-            });
-        });
-        return manualRedactionPages;
-    }
-
-
-    public List<RedactionLogEntry> addEntries(Map<Integer, List<Entity>> entities, ManualRedactions manualRedactions,
-                                              int page, String ruleSetId) {
 
         List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
 
@@ -176,11 +96,9 @@ public class RedactionLogCreatorService {
         entityLoop:
         for (Entity entity : entities.get(page)) {
 
-            List<Comment> comments = null;
-
             for (EntityPositionSequence entityPositionSequence : entity.getPositionSequences()) {
 
-                RedactionLogEntry redactionLogEntry = createRedactionLogEntry(entity, ruleSetId);
+                RedactionLogEntry redactionLogEntry = createRedactionLogEntry(entity, dossierTemplateId);
                 if (processedIds.contains(entityPositionSequence.getId())) {
 
                     // TODO refactor this outer loop jump as soon as we have the time.
@@ -189,60 +107,7 @@ public class RedactionLogCreatorService {
                     processedIds.add(entityPositionSequence.getId());
                 }
 
-                if (manualRedactions != null && !manualRedactions.getIdsToRemove().isEmpty()) {
+                redactionLogEntry.setId(entityPositionSequence.getId());
-                    for (IdRemoval manualRemoval : manualRedactions.getIdsToRemove()) {
-                        if (manualRemoval.getId().equals(entityPositionSequence.getId())) {
-                            comments = manualRedactions.getComments().get(manualRemoval.getId());
-                            String manualOverrideReason = null;
-                            if (manualRemoval.getStatus().equals(Status.APPROVED)) {
-                                entity.setRedaction(false);
-                                redactionLogEntry.setRedacted(false);
-                                redactionLogEntry.setStatus(Status.APPROVED);
-                                manualOverrideReason = entity.getRedactionReason() + ", removed by manual override";
-                                redactionLogEntry.setColor(getColor(entity, ruleSetId, false));
-                            } else if (manualRemoval.getStatus().equals(Status.REQUESTED)) {
-                                manualOverrideReason = entity.getRedactionReason() + ", requested to remove";
-                                redactionLogEntry.setStatus(Status.REQUESTED);
-                                redactionLogEntry.setColor(getColor(entity, ruleSetId, true));
-                            } else {
-                                redactionLogEntry.setStatus(Status.DECLINED);
-                            }
-
-                            entity.setRedactionReason(manualOverrideReason != null ? manualOverrideReason : entity.getRedactionReason());
-                            redactionLogEntry.setReason(manualOverrideReason);
-                            redactionLogEntry.setManual(true);
-                            redactionLogEntry.setManualRedactionType(ManualRedactionType.REMOVE);
-                        }
-                    }
-                }
-
-                if (manualRedactions != null && !manualRedactions.getForceRedacts().isEmpty()) {
-                    for (ManualForceRedact manualForceRedact : manualRedactions.getForceRedacts()) {
-                        if (manualForceRedact.getId().equals(entityPositionSequence.getId())) {
-                            String manualOverrideReason = null;
-                            if (manualForceRedact.getStatus().equals(Status.APPROVED)) {
-                                entity.setRedaction(true);
-                                redactionLogEntry.setRedacted(true);
-                                redactionLogEntry.setStatus(Status.APPROVED);
-                                redactionLogEntry.setColor(getColor(entity, ruleSetId, false));
-                                manualOverrideReason = entity.getRedactionReason() + ", forced by manual override";
-                                redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
-                            } else if (manualForceRedact.getStatus().equals(Status.REQUESTED)) {
-                                manualOverrideReason = entity.getRedactionReason() + ", requested to force redact";
-                                redactionLogEntry.setStatus(Status.REQUESTED);
-                                redactionLogEntry.setColor(getColor(entity, ruleSetId, true));
-                                redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
-                            } else {
-                                redactionLogEntry.setStatus(Status.DECLINED);
-                            }
-
-                            entity.setRedactionReason(manualOverrideReason != null ? manualOverrideReason : entity.getRedactionReason());
-                            redactionLogEntry.setReason(manualOverrideReason);
-                            redactionLogEntry.setManual(true);
-                            redactionLogEntry.setManualRedactionType(ManualRedactionType.FORCE_REDACT);
-                        }
-                    }
-                }
 
                 if (CollectionUtils.isNotEmpty(entityPositionSequence.getSequences())) {
                     List<Rectangle> rectanglesPerLine = getRectanglesPerLine(entityPositionSequence.getSequences()
@@ -250,17 +115,10 @@ public class RedactionLogCreatorService {
                             .flatMap(seq -> seq.getTextPositions().stream())
                             .collect(Collectors.toList()), page);
 
-                    if (manualRedactions != null) {
-                        comments = manualRedactions.getComments().get(entityPositionSequence.getId());
-                    }
-
-                    redactionLogEntry.setComments(comments);
                     redactionLogEntry.getPositions().addAll(rectanglesPerLine);
 
                 }
 
-                redactionLogEntry.setId(entityPositionSequence.getId());
-
                 // FIXME ids should never be null. Figure out why this happens.
                 if (redactionLogEntry.getId() != null) {
                     redactionLogEntities.add(redactionLogEntry);
@@ -276,20 +134,22 @@ public class RedactionLogCreatorService {
 
         List<Rectangle> rectangles = new ArrayList<>();
         if (textPositions.size() == 1) {
-            rectangles.add( TextPositionSequence.fromData(textPositions, page).getRectangle());
+            rectangles.add(TextPositionSequence.fromData(textPositions, page).getRectangle());
         } else {
             float y = textPositions.get(0).getYDirAdj();
             int startIndex = 0;
             for (int i = 1; i < textPositions.size(); i++) {
                 float yDirAdj = textPositions.get(i).getYDirAdj();
                 if (yDirAdj != y) {
-                    rectangles.add( TextPositionSequence.fromData(textPositions.subList(startIndex, i), page).getRectangle());
+                    rectangles.add(TextPositionSequence.fromData(textPositions.subList(startIndex, i), page)
+                            .getRectangle());
                     y = yDirAdj;
                     startIndex = i;
                 }
             }
             if (startIndex != textPositions.size()) {
-                rectangles.add( TextPositionSequence.fromData(textPositions.subList(startIndex, textPositions.size()), page).getRectangle());
+                rectangles.add(TextPositionSequence.fromData(textPositions.subList(startIndex, textPositions.size()), page)
+                        .getRectangle());
             }
         }
 
@@ -297,80 +157,20 @@ public class RedactionLogCreatorService {
     }
 
 
-    public List<RedactionLogEntry> addManualAddEntries(Set<ManualRedactionEntry> manualAdds,
+    private RedactionLogEntry createRedactionLogEntry(Entity entity, String dossierTemplateId) {
-                                                       Map<String, List<Comment>> comments, int page,
-                                                       String ruleSetId) {
 
-        List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
+        Set<String> referenceIds = new HashSet<>();
-
+        entity.getReferences().forEach(ref -> ref.getPositionSequences().forEach(pos -> referenceIds.add(pos.getId())));
-        if (manualAdds == null) {
-            return redactionLogEntities;
-        }
-
-        for (ManualRedactionEntry manualRedactionEntry : manualAdds) {
-
-            String id = manualRedactionEntry.getId();
-
-            RedactionLogEntry redactionLogEntry = createRedactionLogEntry(manualRedactionEntry, id, ruleSetId);
-
-            List<Rectangle> rectanglesOnPage = new ArrayList<>();
-            for (Rectangle rectangle : manualRedactionEntry.getPositions()) {
-                if (page == rectangle.getPage()) {
-                    rectanglesOnPage.add(rectangle);
-                    redactionLogEntry.getPositions().add(rectangle);
-                }
-            }
-
-            redactionLogEntry.setComments(comments.get(id));
-            if (!rectanglesOnPage.isEmpty() && !approvedAndShouldBeInDictionary(manualRedactionEntry)) {
-                redactionLogEntities.add(redactionLogEntry);
-            }
-        }
-
-        return redactionLogEntities;
-    }
-
-
-    private boolean approvedAndShouldBeInDictionary(ManualRedactionEntry manualRedactionEntry) {
-
-        return manualRedactionEntry.getStatus().equals(Status.APPROVED) && manualRedactionEntry.isAddToDictionary();
-    }
-
-
-    private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry, String id,
-                                                      String ruleSetId) {
 
         return RedactionLogEntry.builder()
-                .id(id)
+                .color(getColor(entity.getType(), dossierTemplateId, entity.isRedaction()))
-                .color(getColorForManualAdd(manualRedactionEntry.getType(), ruleSetId, manualRedactionEntry.getStatus()))
-                .reason(manualRedactionEntry.getReason())
-                .legalBasis(manualRedactionEntry.getLegalBasis())
-                .value(manualRedactionEntry.getValue())
-                .type(manualRedactionEntry.getType())
-                .redacted(true)
-                .isHint(false)
-                .section(manualRedactionEntry.getSection())
-                .sectionNumber(manualRedactionEntry.getSectionNumber())
-                .manual(true)
-                .status(manualRedactionEntry.getStatus())
-                .manualRedactionType(ManualRedactionType.ADD)
-                .isDictionaryEntry(false)
-                .isDossierDictionaryEntry(manualRedactionEntry.isAddToDossierDictionary())
-                .build();
-    }
-
-
-    private RedactionLogEntry createRedactionLogEntry(Entity entity, String ruleSetId) {
-
-        return RedactionLogEntry.builder()
-                .color(getColor(entity, ruleSetId, false))
                 .reason(entity.getRedactionReason())
                 .legalBasis(entity.getLegalBasis())
                 .value(entity.getWord())
                 .type(entity.getType())
                 .redacted(entity.isRedaction())
-                .isHint(isHint(entity, ruleSetId))
+                .isHint(isHint(entity.getType(), dossierTemplateId))
-                .isRecommendation(isRecommendation(entity, ruleSetId))
+                .isRecommendation(isRecommendation(entity.getType(), dossierTemplateId))
                 .section(entity.getHeadline())
                 .sectionNumber(entity.getSectionNumber())
                 .matchedRule(entity.getMatchedRule())
@@ -380,104 +180,30 @@ public class RedactionLogCreatorService {
                 .startOffset(entity.getStart())
                 .endOffset(entity.getEnd())
                 .isDossierDictionaryEntry(entity.isDossierDictionaryEntry())
+                .engines(entity.getEngines())
+                .reference(referenceIds)
                 .build();
     }
 
 
-    private float[] getColor(Entity entity, String ruleSetId, boolean requestedToRemove) {
+    private float[] getColor(String type, String dossierTemplateId, boolean isRedaction) {
 
-        if (requestedToRemove) {
+        if (!isRedaction && !isHint(type, dossierTemplateId)) {
-            return dictionaryService.getRequestRemoveColor(ruleSetId);
+            return dictionaryService.getNotRedactedColor(dossierTemplateId);
         }
-        if (!entity.isRedaction() && !isHint(entity, ruleSetId)) {
+        return dictionaryService.getColor(type, dossierTemplateId);
-            return dictionaryService.getNotRedactedColor(ruleSetId);
-        }
-        return dictionaryService.getColor(entity.getType(), ruleSetId);
     }
 
 
-    private float[] getColorForManualAdd(String type, String ruleSetId, Status status) {
+    private boolean isHint(String type, String dossierTemplateId) {
 
-        if (status.equals(Status.REQUESTED)) {
+        return dictionaryService.isHint(type, dossierTemplateId);
-            return dictionaryService.getRequestAddColor(ruleSetId);
-        } else if (status.equals(Status.DECLINED)) {
-            return dictionaryService.getNotRedactedColor(ruleSetId);
-        }
-        return getColor(type, ruleSetId);
     }
 
 
-    private float[] getColor(String type, String ruleSetId) {
+    private boolean isRecommendation(String type, String dossierTemplateId) {
 
-        return dictionaryService.getColor(type, ruleSetId);
+        return dictionaryService.isRecommendation(type, dossierTemplateId);
-    }
-
-
-    private float[] getColorForImage(Image image, String ruleSetId, boolean requestedToRemove) {
-
-        if (requestedToRemove) {
-            return dictionaryService.getRequestRemoveColor(ruleSetId);
-        }
-        if (!image.isRedaction() && !dictionaryService.isHint(image.getType(), ruleSetId)) {
-            return dictionaryService.getNotRedactedColor(ruleSetId);
-        }
-        return dictionaryService.getColor(image.getType(), ruleSetId);
-    }
-
-
-    private boolean isHint(Entity entity, String ruleSetId) {
-
-        return dictionaryService.isHint(entity.getType(), ruleSetId);
-    }
-
-
-    private boolean isRecommendation(Entity entity, String ruleSetId) {
-
-        return dictionaryService.isRecommendation(entity.getType(), ruleSetId);
-    }
-
-
-    private void addSectionGrid(Document classifiedDoc, int page) {
-
-        for (Paragraph paragraph : classifiedDoc.getParagraphs()) {
-
-            for (int i = 0; i <= paragraph.getPageBlocks().size() - 1; i++) {
-
-                AbstractTextContainer textBlock = paragraph.getPageBlocks().get(i);
-
-                if (textBlock.getPage() != page) {
-                    continue;
-                }
-
-                if (textBlock instanceof TextBlock) {
-
-                    classifiedDoc.getSectionGrid()
-                            .getRectanglesPerPage()
-                            .computeIfAbsent(page, (x) -> new ArrayList<>())
-                            .add(new SectionRectangle(new Point(textBlock.getMinX(), textBlock.getMinY()), textBlock.getWidth(), textBlock
-                                    .getHeight(), i + 1, paragraph.getPageBlocks().size()));
-
-                } else if (textBlock instanceof Table) {
-
-                    List<CellRectangle> cellRectangles = new ArrayList<>();
-                    for (List<Cell> row : ((Table) textBlock).getRows()) {
-                        for (Cell cell : row) {
-                            if (cell != null) {
-                                cellRectangles.add(new CellRectangle(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
-                                        .getWidth(), (float) cell.getHeight()));
-                            }
-                        }
-                    }
-
-                    classifiedDoc.getSectionGrid()
-                            .getRectanglesPerPage()
-                            .computeIfAbsent(page, (x) -> new ArrayList<>())
-                            .add(new SectionRectangle(new Point(textBlock.getMinX(), textBlock.getMinY()), textBlock.getWidth(), textBlock
-                                    .getHeight(), i + 1, paragraph.getPageBlocks().size(), cellRectangles));
-
-                }
-            }
-        }
     }
 
 }

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

@@ -0,0 +1,328 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import java.time.OffsetDateTime;
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+import java.util.stream.Collectors;
+
+import org.springframework.stereotype.Service;
+
+import com.iqser.red.service.redaction.v1.model.Comment;
+import com.iqser.red.service.redaction.v1.model.IdRemoval;
+import com.iqser.red.service.redaction.v1.model.ManualForceRedact;
+import com.iqser.red.service.redaction.v1.model.ManualImageRecategorization;
+import com.iqser.red.service.redaction.v1.model.ManualLegalBasisChange;
+import com.iqser.red.service.redaction.v1.model.ManualRedactionEntry;
+import com.iqser.red.service.redaction.v1.model.ManualRedactionType;
+import com.iqser.red.service.redaction.v1.model.ManualRedactions;
+import com.iqser.red.service.redaction.v1.model.RedactionLog;
+import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
+import com.iqser.red.service.redaction.v1.model.Status;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
+@Service
+@RequiredArgsConstructor
+public class RedactionLogMergeService {
+
+    private final DictionaryService dictionaryService;
+
+
+    public RedactionLog mergeRedactionLogData(RedactionLog redactionLog, String dossierTemplateId,
+                                              ManualRedactions manualRedactions, Set<Integer> excludedPages) {
+
+        log.info("Merging Redaction log with manual redactions {}", manualRedactions);
+        if (manualRedactions != null) {
+
+            var manualRedactionLogEntries = addManualAddEntries(manualRedactions.getEntriesToAdd(), manualRedactions.getComments(), dossierTemplateId);
+
+            redactionLog.getRedactionLogEntry().addAll(manualRedactionLogEntries);
+
+            var manualRedactionWrappers = createManualRedactionWrappers(manualRedactions);
+
+            for (RedactionLogEntry entry : redactionLog.getRedactionLogEntry()) {
+
+                processRedactionLogEntry(manualRedactionWrappers.stream()
+                        .filter(mr -> entry.getId().equals(mr.getId()))
+                        .collect(Collectors.toList()), dossierTemplateId, entry);
+
+                entry.setComments(manualRedactions.getComments().get(entry.getId()));
+
+                if (excludedPages != null && !excludedPages.isEmpty()) {
+                    entry.getPositions().forEach(pos -> {
+                        if (excludedPages.contains(pos.getPage())) {
+                            entry.setExcluded(true);
+                        }
+                    });
+                }
+            }
+
+        }
+
+        return redactionLog;
+    }
+
+
+    private List<ManualRedactionWrapper> createManualRedactionWrappers(ManualRedactions manualRedactions) {
+
+        List<ManualRedactionWrapper> manualRedactionWrappers = new ArrayList<>();
+
+        manualRedactions.getImageRecategorizations().forEach(item -> {
+            if (item.getSoftDeletedTime() == null) {
+                manualRedactionWrappers.add(new ManualRedactionWrapper(item.getId(), item.getRequestDate(), item));
+            }
+        });
+
+        manualRedactions.getIdsToRemove().forEach(item -> {
+            if (item.getSoftDeletedTime() == null) {
+                manualRedactionWrappers.add(new ManualRedactionWrapper(item.getId(), item.getRequestDate(), item));
+            }
+        });
+
+        manualRedactions.getForceRedacts().forEach(item -> {
+            if (item.getSoftDeletedTime() == null) {
+                manualRedactionWrappers.add(new ManualRedactionWrapper(item.getId(), item.getRequestDate(), item));
+            }
+        });
+
+        manualRedactions.getManualLegalBasisChanges().forEach(item -> {
+            if (item.getSoftDeletedTime() == null) {
+                manualRedactionWrappers.add(new ManualRedactionWrapper(item.getId(), item.getRequestDate(), item));
+            }
+        });
+
+        Collections.sort(manualRedactionWrappers);
+
+        return manualRedactionWrappers;
+    }
+
+
+    private void processRedactionLogEntry(List<ManualRedactionWrapper> manualRedactionWrappers,
+                                          String dossierTemplateId, RedactionLogEntry redactionLogEntry) {
+
+        manualRedactionWrappers.forEach(mrw -> {
+
+            if (mrw.getItem() instanceof ManualImageRecategorization) {
+                var imageRecategorization = (ManualImageRecategorization) mrw.getItem();
+                String manualOverrideReason = null;
+                if (imageRecategorization.getStatus().equals(Status.APPROVED)) {
+                    redactionLogEntry.setStatus(Status.APPROVED);
+                    redactionLogEntry.setType(imageRecategorization.getType());
+                    manualOverrideReason = mergeReasonIfNecessary(redactionLogEntry.getReason(), ", recategorized by manual override");
+                } else if (imageRecategorization.getStatus().equals(Status.REQUESTED)) {
+                    manualOverrideReason = mergeReasonIfNecessary(redactionLogEntry.getReason(), ", requested to recategorize");
+                    redactionLogEntry.setStatus(Status.REQUESTED);
+                    redactionLogEntry.setColor(getColor(redactionLogEntry.getType(), dossierTemplateId, false, redactionLogEntry
+                            .isRedacted(), false));
+                    redactionLogEntry.setRecategorizationType(imageRecategorization.getType());
+                } else {
+                    redactionLogEntry.setStatus(Status.DECLINED);
+                }
+
+                redactionLogEntry.setManualRedactionUserId(imageRecategorization.getUser());
+                redactionLogEntry.setReason(manualOverrideReason);
+                redactionLogEntry.setManual(true);
+                redactionLogEntry.setManualRedactionType(ManualRedactionType.RECATEGORIZE);
+            }
+
+            if (mrw.getItem() instanceof IdRemoval) {
+                var manualRemoval = (IdRemoval) mrw.getItem();
+                String manualOverrideReason = null;
+                if (manualRemoval.getStatus().equals(Status.APPROVED)) {
+                    redactionLogEntry.setRedacted(false);
+                    redactionLogEntry.setStatus(Status.APPROVED);
+                    manualOverrideReason = mergeReasonIfNecessary(redactionLogEntry.getReason(), ", removed by manual override");
+                    redactionLogEntry.setColor(getColor(redactionLogEntry.getType(), dossierTemplateId, false, redactionLogEntry
+                            .isRedacted(), true));
+                } else if (manualRemoval.getStatus().equals(Status.REQUESTED)) {
+                    manualOverrideReason = mergeReasonIfNecessary(redactionLogEntry.getReason(), ", requested to remove");
+                    redactionLogEntry.setStatus(Status.REQUESTED);
+                    redactionLogEntry.setColor(getColor(redactionLogEntry.getType(), dossierTemplateId, true, redactionLogEntry
+                            .isRedacted(), false));
+                } else {
+                    redactionLogEntry.setStatus(Status.DECLINED);
+                }
+
+                redactionLogEntry.setReason(manualOverrideReason);
+                redactionLogEntry.setManual(true);
+                redactionLogEntry.setManualRedactionUserId(manualRemoval.getUser());
+                redactionLogEntry.setManualRedactionType(ManualRedactionType.REMOVE);
+                redactionLogEntry.setDictionaryEntry(manualRemoval.isRemoveFromDictionary());
+                redactionLogEntry.setDossierDictionaryEntry(manualRemoval.isRemoveFromDictionary());
+            }
+
+            if (mrw.getItem() instanceof ManualForceRedact) {
+                var manualForceRedact = (ManualForceRedact) mrw.getItem();
+                String manualOverrideReason = null;
+                if (manualForceRedact.getStatus().equals(Status.APPROVED)) {
+                    redactionLogEntry.setRedacted(true);
+                    redactionLogEntry.setStatus(Status.APPROVED);
+                    redactionLogEntry.setColor(getColor(redactionLogEntry.getType(), dossierTemplateId, false, redactionLogEntry
+                            .isRedacted(), false));
+                    manualOverrideReason = mergeReasonIfNecessary(redactionLogEntry.getReason(), ", forced by manual override");
+                    redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
+                } else if (manualForceRedact.getStatus().equals(Status.REQUESTED)) {
+                    manualOverrideReason = mergeReasonIfNecessary(redactionLogEntry.getReason(), ", requested to force redact");
+                    redactionLogEntry.setStatus(Status.REQUESTED);
+                    redactionLogEntry.setColor(getColor(redactionLogEntry.getType(), dossierTemplateId, true, redactionLogEntry
+                            .isRedacted(), false));
+                    redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
+                } else {
+                    redactionLogEntry.setStatus(Status.DECLINED);
+                }
+
+                redactionLogEntry.setManualRedactionUserId(manualForceRedact.getUser());
+                redactionLogEntry.setReason(manualOverrideReason);
+                redactionLogEntry.setManual(true);
+                redactionLogEntry.setManualRedactionType(ManualRedactionType.FORCE_REDACT);
+
+            }
+
+            if (mrw.getItem() instanceof ManualLegalBasisChange) {
+                var manualLegalBasisChange = (ManualLegalBasisChange) mrw.getItem();
+                String manualOverrideReason = null;
+                if (manualLegalBasisChange.getStatus().equals(Status.APPROVED)) {
+                    redactionLogEntry.setStatus(Status.APPROVED);
+                    manualOverrideReason = mergeReasonIfNecessary(redactionLogEntry.getReason(), ", legal basis was manually changed");
+                    redactionLogEntry.setLegalBasis(manualLegalBasisChange.getLegalBasis());
+                } else if (manualLegalBasisChange.getStatus().equals(Status.REQUESTED)) {
+                    manualOverrideReason = mergeReasonIfNecessary(redactionLogEntry.getReason(), ", legal basis change requested");
+                    redactionLogEntry.setStatus(Status.REQUESTED);
+                    redactionLogEntry.setColor(getColor(redactionLogEntry.getType(), dossierTemplateId, true, redactionLogEntry
+                            .isRedacted(), false));
+                    redactionLogEntry.setLegalBasisChangeValue(manualLegalBasisChange.getLegalBasis());
+                } else {
+                    redactionLogEntry.setStatus(Status.DECLINED);
+                }
+
+                redactionLogEntry.setManualRedactionUserId(manualLegalBasisChange.getUser());
+                redactionLogEntry.setReason(manualOverrideReason);
+                redactionLogEntry.setManual(true);
+                redactionLogEntry.setManualRedactionType(ManualRedactionType.LEGAL_BASIS_CHANGE);
+            }
+
+        });
+
+    }
+
+
+    private String mergeReasonIfNecessary(String currentReason, String addition) {
+
+        if (currentReason != null) {
+            if (!currentReason.contains(addition)) {
+                return currentReason + addition;
+            }
+            return currentReason;
+        } else {
+            return "";
+        }
+    }
+
+
+    public List<RedactionLogEntry> addManualAddEntries(Set<ManualRedactionEntry> manualAdds,
+                                                       Map<String, List<Comment>> comments, String dossierTemplateId) {
+
+        List<RedactionLogEntry> redactionLogEntries = new ArrayList<>();
+
+        for (ManualRedactionEntry manualRedactionEntry : manualAdds) {
+
+            if (!approvedAndShouldBeInDictionary(manualRedactionEntry)) {
+                RedactionLogEntry redactionLogEntry = createRedactionLogEntry(manualRedactionEntry, manualRedactionEntry
+                        .getId(), dossierTemplateId);
+                redactionLogEntry.setPositions(manualRedactionEntry.getPositions());
+                redactionLogEntry.setComments(comments.get(manualRedactionEntry.getId()));
+                redactionLogEntries.add(redactionLogEntry);
+            }
+        }
+
+        return redactionLogEntries;
+    }
+
+
+    private boolean approvedAndShouldBeInDictionary(ManualRedactionEntry manualRedactionEntry) {
+
+        return manualRedactionEntry.getStatus()
+                .equals(Status.APPROVED) && (manualRedactionEntry.isAddToDictionary() || manualRedactionEntry.isAddToDossierDictionary());
+    }
+
+
+    private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry, String id,
+                                                      String dossierTemplateId) {
+
+        return RedactionLogEntry.builder()
+                .id(id)
+                .color(getColorForManualAdd(manualRedactionEntry.getType(), dossierTemplateId, manualRedactionEntry.getStatus()))
+                .reason(manualRedactionEntry.getReason())
+                .isDictionaryEntry(manualRedactionEntry.isAddToDictionary())
+                .isDossierDictionaryEntry(manualRedactionEntry.isAddToDossierDictionary())
+                .legalBasis(manualRedactionEntry.getLegalBasis())
+                .value(manualRedactionEntry.getValue())
+                .type(manualRedactionEntry.getType())
+                .redacted(true)
+                .isHint(false)
+                .section(null)
+                .sectionNumber(-1)
+                .manual(true)
+                .status(manualRedactionEntry.getStatus())
+                .manualRedactionType(ManualRedactionType.ADD)
+                .manualRedactionUserId(manualRedactionEntry.getUser())
+                .build();
+    }
+
+
+    private float[] getColor(String type, String dossierTemplateId, boolean requested, boolean isRedaction,
+                             boolean skipped) {
+
+        if (requested) {
+            return dictionaryService.getRequestRemoveColor(dossierTemplateId);
+        }
+        if (skipped || !isRedaction && !dictionaryService.isHint(type, dossierTemplateId)) {
+            return dictionaryService.getNotRedactedColor(dossierTemplateId);
+        }
+        return dictionaryService.getColor(type, dossierTemplateId);
+    }
+
+
+    private float[] getColorForManualAdd(String type, String dossierTemplateId, Status status) {
+
+        if (status.equals(Status.REQUESTED)) {
+            return dictionaryService.getRequestAddColor(dossierTemplateId);
+        } else if (status.equals(Status.DECLINED)) {
+            return dictionaryService.getNotRedactedColor(dossierTemplateId);
+        }
+        return getColor(type, dossierTemplateId);
+    }
+
+
+    private float[] getColor(String type, String dossierTemplateId) {
+
+        return dictionaryService.getColor(type, dossierTemplateId);
+    }
+
+
+    @Data
+    @AllArgsConstructor
+    private static class ManualRedactionWrapper implements Comparable<ManualRedactionWrapper> {
+
+        private String id;
+        private OffsetDateTime date;
+        private Object item;
+
+
+        @Override
+        public int compareTo(ManualRedactionWrapper o) {
+
+            return this.date.compareTo(o.date);
+        }
+
+    }
+
+}
+

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

@@ -0,0 +1,76 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import org.springframework.stereotype.Service;
+
+import com.iqser.red.service.redaction.v1.model.CellRectangle;
+import com.iqser.red.service.redaction.v1.model.Point;
+import com.iqser.red.service.redaction.v1.model.SectionRectangle;
+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.tableextraction.model.AbstractTextContainer;
+import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
+import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
+
+import lombok.RequiredArgsConstructor;
+
+@Service
+@RequiredArgsConstructor
+public class SectionGridCreatorService {
+
+
+    public void createSectionGrid(Document classifiedDoc, int numberOfPages) {
+
+        for (int page = 1; page <= numberOfPages; page++) {
+            addSectionGrid(classifiedDoc, page);
+        }
+    }
+
+
+    private void addSectionGrid(Document classifiedDoc, int page) {
+
+        for (Paragraph paragraph : classifiedDoc.getParagraphs()) {
+
+            for (int i = 0; i <= paragraph.getPageBlocks().size() - 1; i++) {
+
+                AbstractTextContainer textBlock = paragraph.getPageBlocks().get(i);
+
+                if (textBlock.getPage() != page) {
+                    continue;
+                }
+
+                if (textBlock instanceof TextBlock) {
+
+                    classifiedDoc.getSectionGrid()
+                            .getRectanglesPerPage()
+                            .computeIfAbsent(page, (x) -> new ArrayList<>())
+                            .add(new SectionRectangle(new Point(textBlock.getMinX(), textBlock.getMinY()), textBlock.getWidth(), textBlock
+                                    .getHeight(), i + 1, paragraph.getPageBlocks().size()));
+
+                } else if (textBlock instanceof Table) {
+
+                    List<CellRectangle> cellRectangles = new ArrayList<>();
+                    for (List<Cell> row : ((Table) textBlock).getRows()) {
+                        for (Cell cell : row) {
+                            if (cell != null) {
+                                cellRectangles.add(new CellRectangle(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
+                                        .getWidth(), (float) cell.getHeight()));
+                            }
+                        }
+                    }
+
+                    classifiedDoc.getSectionGrid()
+                            .getRectanglesPerPage()
+                            .computeIfAbsent(page, (x) -> new ArrayList<>())
+                            .add(new SectionRectangle(new Point(textBlock.getMinX(), textBlock.getMinY()), textBlock.getWidth(), textBlock
+                                    .getHeight(), i + 1, paragraph.getPageBlocks().size(), cellRectangles));
+
+                }
+            }
+        }
+    }
+
+}

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

@@ -0,0 +1,220 @@
+package com.iqser.red.service.redaction.v1.server.redaction.service;
+
+import java.util.ArrayList;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Locale;
+import java.util.Map;
+import java.util.concurrent.atomic.AtomicInteger;
+import java.util.stream.Collectors;
+
+import org.apache.commons.collections4.CollectionUtils;
+import org.springframework.stereotype.Service;
+
+import com.iqser.red.service.redaction.v1.model.Point;
+import com.iqser.red.service.redaction.v1.model.SectionArea;
+import com.iqser.red.service.redaction.v1.server.classification.model.Document;
+import com.iqser.red.service.redaction.v1.server.classification.model.Footer;
+import com.iqser.red.service.redaction.v1.server.classification.model.Header;
+import com.iqser.red.service.redaction.v1.server.classification.model.Paragraph;
+import com.iqser.red.service.redaction.v1.server.classification.model.SectionText;
+import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
+import com.iqser.red.service.redaction.v1.server.classification.model.UnclassifiedText;
+import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Image;
+import com.iqser.red.service.redaction.v1.server.redaction.model.ImageType;
+import com.iqser.red.service.redaction.v1.server.redaction.model.PdfImage;
+import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
+import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
+import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
+
+import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
+@Service
+@RequiredArgsConstructor
+public class SectionTextBuilderService {
+
+    public List<SectionText> buildSectionText(Document classifiedDoc) {
+
+        List<SectionText> sectionTexts = new ArrayList<>();
+        AtomicInteger sectionNumber = new AtomicInteger(1);
+        for (Paragraph paragraph : classifiedDoc.getParagraphs()) {
+
+            List<Table> tables = paragraph.getTables();
+            for (Table table : tables) {
+                if (table.getColCount() == 2) {
+                    sectionTexts.add(processTableAsOneText(table, sectionNumber));
+                } else {
+                    sectionTexts.addAll(processTablePerRow(table, sectionNumber));
+                }
+                sectionNumber.incrementAndGet();
+            }
+            sectionTexts.add(processText(paragraph.getSearchableText(), paragraph.getTextBlocks(), paragraph.getHeadline(), sectionNumber, paragraph
+                    .getImages()));
+            sectionNumber.incrementAndGet();
+        }
+
+        for (Header header : classifiedDoc.getHeaders()) {
+            sectionTexts.add(processText(header.getSearchableText(), header.getTextBlocks(), "Header", sectionNumber, new ArrayList<>()));
+            sectionNumber.incrementAndGet();
+        }
+
+        for (Footer footer : classifiedDoc.getFooters()) {
+            sectionTexts.add(processText(footer.getSearchableText(), footer.getTextBlocks(), "Footer", sectionNumber, new ArrayList<>()));
+            sectionNumber.incrementAndGet();
+        }
+
+        for (UnclassifiedText unclassifiedText : classifiedDoc.getUnclassifiedTexts()) {
+            sectionTexts.add(processText(unclassifiedText.getSearchableText(), unclassifiedText.getTextBlocks(), "", sectionNumber, new ArrayList<>()));
+            sectionNumber.incrementAndGet();
+        }
+
+        return sectionTexts;
+    }
+
+
+    private List<SectionText> processTablePerRow(Table table, AtomicInteger sectionNumber) {
+
+        List<SectionText> sectionTexts = new ArrayList<>();
+        for (List<Cell> row : table.getRows()) {
+            SearchableText searchableRow = new SearchableText();
+            Map<String, CellValue> tabularData = new HashMap<>();
+            int start = 0;
+            List<Integer> cellStarts = new ArrayList<>();
+            SectionText sectionText = new SectionText();
+            for (Cell cell : row) {
+
+                if (CollectionUtils.isEmpty(cell.getTextBlocks())) {
+                    continue;
+                }
+
+                SectionArea sectionArea = new SectionArea(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
+                        .getWidth(), (float) cell.getHeight(), cell.getTextBlocks()
+                        .get(0)
+                        .getSequences()
+                        .get(0)
+                        .getPage());
+                sectionText.getSectionAreas().add(sectionArea);
+                sectionText.getTextBlocks().addAll(cell.getTextBlocks());
+
+                int cellStart = start;
+
+                if (!cell.isHeaderCell()) {
+                    cell.getHeaderCells().forEach(headerCell -> {
+                        StringBuilder headerBuilder = new StringBuilder();
+                        headerCell.getTextBlocks().forEach(textBlock -> headerBuilder.append(textBlock.getText()));
+                        String headerName = headerBuilder.toString()
+                                .replaceAll("\n", "")
+                                .replaceAll(" ", "")
+                                .replaceAll("-", "");
+                        sectionArea.setHeader(headerName);
+                        tabularData.put(headerName, new CellValue(cell.getTextBlocks(), cellStart));
+                    });
+                }
+
+                for (TextBlock textBlock : cell.getTextBlocks()) {
+                    // TODO avoid cell overlap merging.
+                    searchableRow.addAll(textBlock.getSequences());
+                }
+                cellStarts.add(cellStart);
+                start = start + cell.toString().trim().length() + 1;
+
+            }
+
+            sectionText.setText(searchableRow.toString());
+            sectionText.setHeadline(table.getHeadline());
+            sectionText.setSectionNumber(sectionNumber.intValue());
+            sectionText.setTable(true);
+            sectionText.setTabularData(tabularData);
+            sectionText.setCellStarts(cellStarts);
+            sectionTexts.add(sectionText);
+
+            sectionNumber.incrementAndGet();
+        }
+
+        return sectionTexts;
+    }
+
+
+    private SectionText processTableAsOneText(Table table, AtomicInteger sectionNumber) {
+
+        SearchableText entireTableText = new SearchableText();
+        SectionText sectionText = new SectionText();
+        int start = 0;
+        List<Integer> cellStarts = new ArrayList<>();
+        for (List<Cell> row : table.getRows()) {
+
+
+            for (Cell cell : row) {
+                if (CollectionUtils.isEmpty(cell.getTextBlocks())) {
+                    continue;
+                }
+
+                SectionArea sectionArea = new SectionArea(new Point((float) cell.getX(), (float) cell.getY()), (float) cell
+                        .getWidth(), (float) cell.getHeight(), cell.getTextBlocks()
+                        .get(0)
+                        .getSequences()
+                        .get(0)
+                        .getPage());
+                sectionText.getTextBlocks().addAll(cell.getTextBlocks());
+                sectionText.getSectionAreas().add(sectionArea);
+
+                for (TextBlock textBlock : cell.getTextBlocks()) {
+                    entireTableText.addAll(textBlock.getSequences());
+                }
+
+
+                cellStarts.add(start);
+                start = start + cell.toString().trim().length() + 1;
+            }
+
+        }
+        sectionText.setCellStarts(cellStarts);
+
+        sectionText.setText(entireTableText.toString());
+        sectionText.setHeadline(table.getHeadline());
+        sectionText.setSectionNumber(sectionNumber.intValue());
+        sectionText.setTable(true);
+        return sectionText;
+    }
+
+
+    private SectionText processText(SearchableText searchableText, List<TextBlock> paragraphTextBlocks, String headline,
+                                    AtomicInteger sectionNumber, List<PdfImage> images) {
+
+        SectionText sectionText = new SectionText();
+        for (TextBlock paragraphTextBlock : paragraphTextBlocks) {
+            SectionArea sectionArea = new SectionArea(new Point(paragraphTextBlock.getMinX(), paragraphTextBlock.getMinY()), paragraphTextBlock
+                    .getWidth(), paragraphTextBlock.getHeight(), paragraphTextBlock.getPage());
+            sectionText.getSectionAreas().add(sectionArea);
+        }
+
+        sectionText.setText(searchableText.toString());
+        sectionText.setHeadline(headline);
+        sectionText.setSectionNumber(sectionNumber.intValue());
+        sectionText.setTable(false);
+        sectionText.setImages(images.stream()
+                .map(image -> convertImage(image, sectionNumber.intValue(), headline))
+                .collect(Collectors.toSet()));
+        sectionText.setTextBlocks(paragraphTextBlocks);
+        return sectionText;
+    }
+
+
+    private Image convertImage(PdfImage pdfImage, int sectionNumber, String headline) {
+
+        return Image.builder()
+                .type(pdfImage.getImageType().equals(ImageType.OTHER) ? "image" : pdfImage.getImageType()
+                        .name()
+                        .toLowerCase(Locale.ROOT))
+                .position(pdfImage.getPosition())
+                .sectionNumber(sectionNumber)
+                .section(headline)
+                .page(pdfImage.getPage())
+                .hasTransparency(pdfImage.isHasTransparency())
+                .build();
+    }
+
+}

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

@@ -1,10 +1,12 @@
 package com.iqser.red.service.redaction.v1.server.redaction.utils;
 
+import com.iqser.red.service.redaction.v1.model.Engine;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Dictionary;
 import com.iqser.red.service.redaction.v1.server.redaction.model.DictionaryIncrementValue;
 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 lombok.experimental.UtilityClass;
 import lombok.extern.slf4j.Slf4j;
 
@@ -17,7 +19,6 @@ import java.util.stream.Collectors;
 @SuppressWarnings("PMD")
 public class EntitySearchUtils {
 
-
     public boolean sectionContainsAny(String sectionText, Set<DictionaryIncrementValue> values) {
 
         String inputString = sectionText.toLowerCase(Locale.ROOT);
@@ -38,10 +39,8 @@ public class EntitySearchUtils {
 
                 if (startIndex > -1 && (startIndex == 0 || Character.isWhitespace(inputString.charAt(startIndex - 1)) || isSeparator(inputString
                         .charAt(startIndex - 1))) && (stopIndex == inputString.length() || isSeparator(inputString.charAt(stopIndex)))) {
-                    if (value.isCaseinsensitive() || !value.isCaseinsensitive() && sectionText.substring(startIndex, stopIndex).equals(value.getValue())) {
                     return true;
                 }
-                }
             } while (startIndex > -1);
         }
         return false;
@@ -49,7 +48,7 @@ public class EntitySearchUtils {
 
 
     public Set<Entity> find(String inputString, Set<String> values, String type, String headline, int sectionNumber,
-                            boolean local, boolean isDossierDictionary) {
+                            boolean isDictionaryEntry, boolean isDossierDictionary, Engine engine) {
 
         Set<Entity> found = new HashSet<>();
 
@@ -69,7 +68,7 @@ public class EntitySearchUtils {
 
                 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, !local, isDossierDictionary));
+                    found.add(new Entity(inputString.substring(startIndex, stopIndex), type, startIndex, stopIndex, headline, sectionNumber, isDictionaryEntry, isDossierDictionary, engine));
                 }
             } while (startIndex > -1);
         }
@@ -98,8 +97,8 @@ public class EntitySearchUtils {
                     .sorted(Comparator.comparing(Entity::getStart))
                     .collect(Collectors.toList());
             Entity firstEntity = orderedEntities.get(0);
-            List<EntityPositionSequence> positionSequences = text.getSequences(firstEntity.getWord().trim(), dictionary.isCaseInsensitiveDictionary(firstEntity
+            List<EntityPositionSequence> positionSequences = text.getSequences(firstEntity.getWord()
-                    .getType()), firstEntity.getTargetSequences());
+                    .trim(), dictionary.isCaseInsensitiveDictionary(firstEntity.getType()), firstEntity.getTargetSequences());
 
             for (int i = 0; i <= orderedEntities.size() - 1; i++) {
                 try {
@@ -133,6 +132,7 @@ public class EntitySearchUtils {
 
 
     public void addEntitiesWithHigherRank(Set<Entity> entities, Set<Entity> found, Dictionary dictionary) {
+
         found.forEach(f -> addEntitiesWithHigherRank(entities, f, dictionary));
     }
 
@@ -143,14 +143,33 @@ public class EntitySearchUtils {
             Entity existing = entities.stream().filter(entity -> entity.equals(found)).findFirst().get();
             if (dictionary.getDictionaryRank(existing.getType()) <= dictionary.getDictionaryRank(found.getType())) {
                 entities.remove(found);
+                entities.add(found);
+            } else {
+                existing.getEngines().addAll(found.getEngines());
             }
-        }
+        } else {
             entities.add(found);
         }
+    }
+
 
     public void addEntitiesIgnoreRank(Set<Entity> entities, Set<Entity> found) {
         // HashSet keeps old value but we want the new.
         entities.removeAll(found);
         entities.addAll(found);
     }
+
+
+    public void addOrAddEngine(Set<Entity> existing, Set<Entity> toBeAdded){
+
+        for(Entity toAdd: toBeAdded){
+            if (existing.contains(toAdd)) {
+                Entity existingEntity = existing.stream().filter(entity -> entity.equals(toAdd)).findFirst().get();
+                existingEntity.getEngines().addAll(toAdd.getEngines());
+            } else {
+                existing.add(toAdd);
+            }
+        }
+    }
+
 }

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

@@ -12,7 +12,7 @@ public class TextNormalizationUtilities {
      * @return Text without line-break hyphenation.
      */
     public static String removeHyphenLineBreaks(String text) {
-        return text.replaceAll("([^\\s\\d\\-]{2,})[\\-\\u00AD]\\R|\n\r(.+ )", "$1$2");
+        return text.replaceAll("([^\\s\\d\\-]{2,})[\\-\\u00AD]\\R", "$1");
     }
 
 }

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

@@ -0,0 +1,165 @@
+package com.iqser.red.service.redaction.v1.server.segmentation;
+
+import java.awt.Graphics;
+import java.awt.geom.Rectangle2D;
+import java.awt.image.BufferedImage;
+import java.util.ArrayList;
+import java.util.List;
+
+import org.springframework.stereotype.Service;
+
+import com.iqser.red.service.redaction.v1.server.redaction.model.PdfImage;
+
+import lombok.RequiredArgsConstructor;
+import lombok.extern.slf4j.Slf4j;
+
+@Slf4j
+@Service
+@RequiredArgsConstructor
+public class ImageMergeService {
+
+
+    public List<PdfImage> mergeImages(List<PdfImage> images, int rotation){
+
+        List<PdfImage> mergedList = processImages(images, rotation);
+
+        List<PdfImage> imagesInImage = new ArrayList<>();
+        for(PdfImage image: mergedList){
+            for (PdfImage inner: mergedList){
+                if(image != inner && image.getPosition().contains(inner.getPosition().getX(), inner.getPosition().getY(), inner.getPosition().getWidth(), inner.getPosition().getHeight())){
+                    imagesInImage.add(inner);
+                }
+            }
+        }
+        mergedList.removeAll(imagesInImage);
+
+        return mergedList;
+    }
+
+
+    //merge images, if they are separated during pdf import, return new list of Pdfimages
+    private List<PdfImage> processImages(List<PdfImage> imageList, int rotation) {
+        if (imageList.size() > 1) {
+            List<PdfImage> mergedList = new ArrayList<>();
+            int countElementsInList = 0;
+            boolean beginImage = true;
+
+            // a List of Boolean, true = candidate for merging, false = no merging
+            List<Boolean> candidatesList = getCandidatesList(imageList, rotation);
+
+            // loop through list, if there are candidates for merging (true), merge images and add it to mergedList
+            for (int i = 0; i < candidatesList.size(); i++) {
+                if (candidatesList.get(i)) {
+                    if (beginImage) {
+                        //begin of image, merge two parts of imageList
+                        PdfImage mergedImage = mergeTwoImages(imageList.get(i), imageList.get(i + 1), rotation);
+                        // image merge successful
+                        if (mergedImage != null) {
+                            mergedList.add(mergedImage);
+                            countElementsInList++;
+                        }
+                    } else {
+                        //middle of an image, merge current piece auf mergedList with image of imageList
+                        PdfImage mergedImage = mergeTwoImages(mergedList.get(countElementsInList - 1), imageList.get(i + 1), rotation);
+                        // image merge successful
+                        if (mergedImage != null) {
+                            mergedList.set(countElementsInList - 1, mergedImage);
+                        }
+                    }
+                    beginImage = false;
+                } else {
+                    // if the last candidate is false, then both images i and i+1 must be added
+                    if (i == candidatesList.size() - 1) {
+                        if (countElementsInList > 0 && mergedList.get(countElementsInList - 1) == imageList.get(i)) {
+                            mergedList.add(imageList.get(i + 1));
+                        } else {
+                            mergedList.add(imageList.get(i));
+                            mergedList.add(imageList.get(i + 1));
+                        }
+                    } else {
+                        //first image is not splitted, add i to resultlist
+                        if (beginImage) {
+                            mergedList.add(imageList.get(i));
+                            countElementsInList++;
+                        } else {
+                            // i is the end of an image, add begin of new image
+                            mergedList.add(imageList.get(i + 1));
+                            countElementsInList++;
+                            beginImage = false;
+                        }
+                    }
+                }
+            }
+            return mergedList;
+        } else {
+            return imageList;
+        }
+    }
+
+    private PdfImage mergeTwoImages(PdfImage image1, PdfImage image2, int rotation) {
+
+        // diese Angaben von getPosition scheinen nicht richtig zu sein, damit werden teile des Bildes abgeschnitten
+        double width = image1.getPosition().getWidth();
+        double width2 = image2.getPosition().getWidth();
+        double height1 = image1.getPosition().getHeight();
+        double height2 = image2.getPosition().getHeight();
+        // mit  den Werten, die unter Image gespeichert sind, funktioniert es
+        double img1height = image1.getImage().getHeight();
+        double img1width = image1.getImage().getWidth();
+        double img2height = image2.getImage().getHeight();
+
+        BufferedImage mergedImage = new BufferedImage((int) img1width, (int) (img1height + img2height), BufferedImage.TYPE_INT_RGB);
+        Graphics mergedImageGraphics = mergedImage.getGraphics();
+        try {
+            mergedImageGraphics.drawImage(image1.getImage(), 0, 0, null);
+            mergedImageGraphics.drawImage(image2.getImage(), 0, (int) (img1height), null);
+
+            // set Image, Position and type for merged Image
+            //set position for merged image with values of image1 and the height of both
+            Rectangle2D pos = new Rectangle2D.Float();
+            pos.setRect(image1.getPosition().getX(), image2.getPosition().getY(), rotation == 90 ? width + width2: width, rotation == 90 ? height1 : height1 + height2);
+            PdfImage newPdfImage = new PdfImage(mergedImage, pos, image1.getPage(), image1.isHasTransparency() || image2.isHasTransparency());
+            // Graphics need to be disposed
+
+            image1.getImage().flush();
+            image2.getImage().flush();
+
+            mergedImage.flush();
+            mergedImageGraphics.dispose();
+
+            return newPdfImage;
+        } catch (Exception e) {
+            // failed to merge image
+            log.error("Failed to merge image", e);
+            return null;
+        }
+
+
+    }
+
+    //make a list of true and false, if the image is a candidate for merging
+    private List<Boolean> getCandidatesList(List<PdfImage> imageList, int rotation) {
+        List<Boolean> candidatesList = new ArrayList<>();
+        for (int i = 0; i < imageList.size(); i++) {
+            if (i >= 1) {
+                candidatesList.add(isCandidateForMerging(imageList.get(i - 1), imageList.get(i), rotation));
+            }
+        }
+        return candidatesList;
+    }
+
+    // evaluate if two images are candidates for merging, depending on their coordinates, width and height
+    private boolean isCandidateForMerging(PdfImage image1, PdfImage image2, int rotation) {
+        double x1 = rotation == 90 ? image1.getPosition().getY() : image1.getPosition().getX();
+        double y1 = rotation == 90 ? image1.getPosition().getX() : image1.getPosition().getY();
+        double width1 = rotation == 90 ? image1.getPosition().getHeight() : image1.getPosition().getWidth();
+        double x2 = rotation == 90 ? image2.getPosition().getY() : image2.getPosition().getX();
+        double y2 = rotation == 90 ? image2.getPosition().getX() : image2.getPosition().getY();
+        double width2 = rotation == 90 ? image2.getPosition().getHeight() : image2.getPosition().getWidth();
+        double height2 = rotation == 90 ? image2.getPosition().getWidth() : image2.getPosition().getHeight();
+        //if the x-coordinates and widths of images are equal and the height is equal to difference between y-coordinates,
+        // then it is the same picture and has to be merged -> return true
+        return x1 == x2 && width1 == width2 && Math.ceil(height2) == Math.ceil(rotation == 90 ? y2 - y1 : y1 - y2) && width2 > (height2 / 6);
+    }
+
+}

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

@@ -1,6 +1,19 @@
 package com.iqser.red.service.redaction.v1.server.segmentation;
 
-import com.iqser.red.service.redaction.v1.model.Rectangle;
+import java.io.File;
+import java.io.FileOutputStream;
+import java.io.IOException;
+import java.io.InputStream;
+import java.util.ArrayList;
+import java.util.List;
+
+import org.apache.commons.io.IOUtils;
+import org.apache.pdfbox.io.MemoryUsageSetting;
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.pdmodel.common.PDRectangle;
+import org.springframework.stereotype.Service;
+
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
 import com.iqser.red.service.redaction.v1.server.classification.model.Page;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
@@ -15,24 +28,9 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractT
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.CleanRulings;
 import com.iqser.red.service.redaction.v1.server.tableextraction.service.RulingCleaningService;
 import com.iqser.red.service.redaction.v1.server.tableextraction.service.TableExtractionService;
+
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
-import org.apache.commons.io.IOUtils;
-import org.apache.pdfbox.io.MemoryUsageSetting;
-import org.apache.pdfbox.pdmodel.PDDocument;
-import org.apache.pdfbox.pdmodel.PDPage;
-import org.apache.pdfbox.pdmodel.common.PDRectangle;
-import org.springframework.stereotype.Service;
-
-import java.awt.Graphics;
-import java.awt.geom.Rectangle2D;
-import java.awt.image.BufferedImage;
-import java.io.File;
-import java.io.FileOutputStream;
-import java.io.IOException;
-import java.io.InputStream;
-import java.util.ArrayList;
-import java.util.List;
 
 @Slf4j
 @Service
@@ -47,13 +45,17 @@ public class PdfSegmentationService {
     private final ClassificationService classificationService;
     private final SectionsBuilderService sectionsBuilderService;
     private final ImageClassificationService imageClassificationService;
+    private final ImageMergeService imageMergeService;
 
 
     public Document parseDocument(InputStream documentInputStream) throws IOException {
+
         return parseDocument(documentInputStream, false);
     }
 
+
     public Document parseDocument(InputStream documentInputStream, boolean ignoreImages) throws IOException {
+
         PDDocument pdDocument = null;
         try {
             //create tempFile
@@ -64,7 +66,6 @@ public class PdfSegmentationService {
             Document document = new Document();
             List<Page> pages = new ArrayList<>();
 
-
             pdDocument = reinitializePDDocument(tempFile, null);
             long pageCount = pdDocument.getNumberOfPages();
 
@@ -101,32 +102,19 @@ public class PdfSegmentationService {
                 page.setRotation(rotation);
                 page.setLandscape(isLandscape || isRotated);
                 page.setPageNumber(pageNumber);
-                List<PdfImage> mergedList = processImages(stripper.getImages());
-
-                List<PdfImage> imagesInImage = new ArrayList<>();
-                for(PdfImage image: mergedList){
-                    for (PdfImage inner: mergedList){
-                        if(image != inner && image.getPosition().contains(inner.getPosition().getX(), inner.getPosition().getY(), inner.getPosition().getWidth(), inner.getPosition().getHeight())){
-                            imagesInImage.add(inner);
-                        }
-                    }
-                }
-                mergedList.removeAll(imagesInImage);
 
+                List<PdfImage> mergedList = imageMergeService.mergeImages(stripper.getImages(), rotation);
                 page.setImages(mergedList);
 
                 tableExtractionService.extractTables(cleanRulings, page);
                 buildPageStatistics(page);
                 increaseDocumentStatistics(page, document);
 
-
                 if (!ignoreImages) {
                     imageClassificationService.classifyImages(page);
                 }
 
                 pages.add(page);
-
-
             }
 
             document.setPages(pages);
@@ -149,7 +137,9 @@ public class PdfSegmentationService {
         }
     }
 
+
     private PDDocument reinitializePDDocument(File tempFile, PDDocument pdDocument) throws IOException {
+
         if (pdDocument != null) {
             pdDocument.close();
         }
@@ -164,130 +154,6 @@ public class PdfSegmentationService {
         return newPDDocument;
     }
 
-    //merge images, if they are separated during pdf import, return new list of Pdfimages
-    private List<PdfImage> processImages(List<PdfImage> imageList) {
-        if (imageList.size() > 1) {
-            List<PdfImage> mergedList = new ArrayList<>();
-            int countElementsInList = 0;
-            boolean beginImage = true;
-
-            // a List of Boolean, true = candidate for merging, false = no merging
-            List<Boolean> candidatesList = getCandidatesList(imageList);
-
-            // loop through list, if there are candidates for merging (true), merge images and add it to mergedList
-            for (int i = 0; i < candidatesList.size(); i++) {
-                if (candidatesList.get(i)) {
-                    if (beginImage) {
-                        //begin of image, merge two parts of imageList
-                        PdfImage mergedImage = mergeTwoImages(imageList.get(i), imageList.get(i + 1));
-                        // image merge successful
-                        if (mergedImage != null) {
-                            mergedList.add(mergedImage);
-                            countElementsInList++;
-                        }
-                    } else {
-                        //middle of an image, merge current piece auf mergedList with image of imageList
-                        PdfImage mergedImage = mergeTwoImages(mergedList.get(countElementsInList - 1), imageList.get(i + 1));
-                        // image merge successful
-                        if (mergedImage != null) {
-                            mergedList.set(countElementsInList - 1, mergedImage);
-                        }
-                    }
-                    beginImage = false;
-                } else {
-                    // if the last candidate is false, then both images i and i+1 must be added
-                    if (i == candidatesList.size() - 1) {
-                        if (countElementsInList > 0 && mergedList.get(countElementsInList - 1) == imageList.get(i)) {
-                            mergedList.add(imageList.get(i + 1));
-                        } else {
-                            mergedList.add(imageList.get(i));
-                            mergedList.add(imageList.get(i + 1));
-                        }
-                    } else {
-                        //first image is not splitted, add i to resultlist
-                        if (beginImage) {
-                            mergedList.add(imageList.get(i));
-                            countElementsInList++;
-                        } else {
-                            // i is the end of an image, add begin of new image
-                            mergedList.add(imageList.get(i + 1));
-                            countElementsInList++;
-                            beginImage = false;
-                        }
-                    }
-                }
-            }
-            return mergedList;
-        } else {
-            return imageList;
-        }
-    }
-
-    private PdfImage mergeTwoImages(PdfImage image1, PdfImage image2) {
-
-        // diese Angaben von getPosition scheinen nicht richtig zu sein, damit werden teile des Bildes abgeschnitten
-        double width = image1.getPosition().getWidth();
-        double height1 = image1.getPosition().getHeight();
-        double height2 = image2.getPosition().getHeight();
-        // mit  den Werten, die unter Image gespeichert sind, funktioniert es
-        double img1height = image1.getImage().getHeight();
-        double img1width = image1.getImage().getWidth();
-        double img2height = image2.getImage().getHeight();
-
-        BufferedImage mergedImage = new BufferedImage((int) img1width, (int) (img1height + img2height), BufferedImage.TYPE_INT_RGB);
-        Graphics mergedImageGraphics = mergedImage.getGraphics();
-        try {
-            mergedImageGraphics.drawImage(image1.getImage(), 0, 0, null);
-            mergedImageGraphics.drawImage(image2.getImage(), 0, (int) (img1height), null);
-
-            // set Image, Position and type for merged Image
-            //set position for merged image with values of image1 and the height of both
-            Rectangle2D pos = new Rectangle2D.Float();
-            pos.setRect(image1.getPosition().getX(), image2.getPosition().getY(), width, height1 + height2);
-            PdfImage newPdfImage = new PdfImage(mergedImage, pos, image1.getPage());
-            // Graphics need to be disposed
-
-            image1.getImage().flush();
-            image2.getImage().flush();
-
-            mergedImage.flush();
-            mergedImageGraphics.dispose();
-
-            return newPdfImage;
-        } catch (Exception e) {
-            // failed to merge image
-            log.error("Failed to merge image", e);
-            return null;
-        }
-
-
-    }
-
-    //make a list of true and false, if the image is a candidate for merging
-    private List<Boolean> getCandidatesList(List<PdfImage> imageList) {
-        List<Boolean> candidatesList = new ArrayList<>();
-        for (int i = 0; i < imageList.size(); i++) {
-            if (i >= 1) {
-                candidatesList.add(isCandidateForMerging(imageList.get(i - 1), imageList.get(i)));
-            }
-        }
-        return candidatesList;
-    }
-
-    // evaluate if two images are candidates for merging, depending on their coordinates, width and height
-    private boolean isCandidateForMerging(PdfImage image1, PdfImage image2) {
-        double x1 = image1.getPosition().getX();
-        double y1 = image1.getPosition().getY();
-        double width1 = image1.getPosition().getWidth();
-        double x2 = image2.getPosition().getX();
-        double y2 = image2.getPosition().getY();
-        double width2 = image2.getPosition().getWidth();
-        double height2 = image2.getPosition().getHeight();
-        //if the x-coordinates and widths of images are equal and the height is equal to difference between y-coordinates,
-        // then it is the same picture and has to be merged -> return true
-        return x1 == x2 && width1 == width2 && Math.ceil(height2) == Math.ceil(y1 - y2) && width2 > (height2 / 6);
-    }
-
 
     private void increaseDocumentStatistics(Page page, Document document) {
 
@@ -319,5 +185,4 @@ public class PdfSegmentationService {
 
     }
 
-
 }

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

@@ -15,4 +15,8 @@ public class RedactionServiceSettings {
 
     private float maxImageCropboxRatio = 0.9f;
 
+    private int analysisVersion = 1;
+
+    private boolean enableEntityRecognition = true;
+
 }

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

@@ -1,10 +1,11 @@
 package com.iqser.red.service.redaction.v1.server.storage;
 
 import com.fasterxml.jackson.databind.ObjectMapper;
-import com.iqser.red.service.file.management.v1.api.model.FileType;
+import com.iqser.red.service.persistence.service.v1.api.model.FileType;
 import com.iqser.red.service.redaction.v1.model.RedactionLog;
 import com.iqser.red.service.redaction.v1.model.SectionGrid;
 import com.iqser.red.service.redaction.v1.server.classification.model.Text;
+import com.iqser.red.service.redaction.v1.server.client.model.NerEntities;
 import com.iqser.red.storage.commons.exception.StorageObjectDoesNotExist;
 import com.iqser.red.storage.commons.service.StorageService;
 import lombok.Getter;
@@ -32,16 +33,16 @@ public class RedactionStorageService {
 
 
     @SneakyThrows
-    public void storeObject(String projectId, String fileId, FileType fileType, Object any) {
+    public void storeObject(String dossierId, String fileId, FileType fileType, Object any) {
-        storageService.storeObject(StorageIdUtils.getStorageId(projectId, fileId, fileType), objectMapper.writeValueAsBytes(any));
+        storageService.storeObject(StorageIdUtils.getStorageId(dossierId, fileId, fileType), objectMapper.writeValueAsBytes(any));
     }
 
 
-    public RedactionLog getRedactionLog(String projectId, String fileId) {
+    public RedactionLog getRedactionLog(String dossierId, String fileId) {
 
         InputStreamResource inputStreamResource;
         try {
-            inputStreamResource = storageService.getObject(StorageIdUtils.getStorageId(projectId, fileId, FileType.REDACTION_LOG));
+            inputStreamResource = storageService.getObject(StorageIdUtils.getStorageId(dossierId, fileId, FileType.REDACTION_LOG));
         } catch (StorageObjectDoesNotExist e) {
             log.debug("Text not available.");
             return null;
@@ -55,11 +56,11 @@ public class RedactionStorageService {
     }
 
 
-    public Text getText(String projectId, String fileId) {
+    public Text getText(String dossierId, String fileId) {
 
         InputStreamResource inputStreamResource;
         try {
-            inputStreamResource = storageService.getObject(StorageIdUtils.getStorageId(projectId, fileId, FileType.TEXT));
+            inputStreamResource = storageService.getObject(StorageIdUtils.getStorageId(dossierId, fileId, FileType.TEXT));
         } catch (StorageObjectDoesNotExist e) {
             log.debug("Text not available.");
             return null;
@@ -73,9 +74,28 @@ public class RedactionStorageService {
     }
 
 
-    public SectionGrid getSectionGrid(String projectId, String fileId) {
+    public NerEntities getNerEntities(String dossierId, String fileId) {
 
-        var sectionGrid = storageService.getObject(StorageIdUtils.getStorageId(projectId, fileId, FileType.SECTION_GRID));
+        InputStreamResource inputStreamResource;
+        try {
+            inputStreamResource = storageService.getObject(StorageIdUtils.getStorageId(dossierId, fileId, FileType.NER_ENTITIES));
+        } catch (StorageObjectDoesNotExist e) {
+            log.debug("NER Entities not available.");
+            return null;
+        }
+
+        try {
+            return objectMapper.readValue(inputStreamResource.getInputStream(), NerEntities.class);
+        } catch (IOException e) {
+            throw new RuntimeException("Could not convert NerEntities", e);
+        }
+    }
+
+
+
+    public SectionGrid getSectionGrid(String dossierId, String fileId) {
+
+        var sectionGrid = storageService.getObject(StorageIdUtils.getStorageId(dossierId, fileId, FileType.SECTION_GRID));
         try {
             return objectMapper.readValue(sectionGrid.getInputStream(), SectionGrid.class);
         } catch (IOException e) {
@@ -95,8 +115,8 @@ public class RedactionStorageService {
 
     public static class StorageIdUtils {
 
-        public static String getStorageId(String projectId, String fileId, FileType fileType) {
+        public static String getStorageId(String dossierId, String fileId, FileType fileType) {
-            return projectId + "/" + fileId + "." + fileType.name() + fileType.getExtension();
+            return dossierId + "/" + fileId + "." + fileType.name() + fileType.getExtension();
         }
 
     }

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

@@ -2,6 +2,8 @@ package com.iqser.red.service.redaction.v1.server.tableextraction.model;
 
 import com.fasterxml.jackson.annotation.JsonIgnore;
 import com.iqser.red.service.redaction.v1.model.Rectangle;
+import com.iqser.red.service.redaction.v1.server.classification.model.Orientation;
+
 import lombok.AllArgsConstructor;
 import lombok.Data;
 import lombok.NoArgsConstructor;
@@ -18,6 +20,8 @@ public abstract class AbstractTextContainer {
     protected String classification;
     protected int page;
 
+    private Orientation orientation = Orientation.NONE;
+
     public abstract String getText();
 
     public boolean contains(AbstractTextContainer other) {

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

@@ -246,8 +246,12 @@ public class Ruling extends Line2D.Float {
 
     public Ruling expand(float amount) {
         Ruling r = (Ruling) this.clone();
+        try {
             r.setStart(this.getStart() - amount);
             r.setEnd(this.getEnd() + amount);
+        } catch (UnsupportedOperationException e){
+            log.warn("Could not expand ruling!");
+        }
         return r;
     }
 

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

@@ -102,7 +102,7 @@ public class PdfVisualisationService {
 
             contentStream.newLineAtOffset(textBlock.getMinX(), textBlock.getMaxY());
 
-            contentStream.showText(textBlock.getClassification());
+            contentStream.showText(textBlock.getClassification() + textBlock.getOrientation());
 
             contentStream.endText();
         }

redaction-service-v1/redaction-service-server-v1/src/main/resources/application.yml

@@ -1,9 +1,9 @@
 info:
   description: Redaction Service Server V1
 
-configuration-service.url: "http://configuration-service-v1:8080"
+persistence-service.url: "http://persistence-service-v1:8080"
-file-management-service.url: "http://file-management-service-v1:8080"
 image-service.url: "http://image-service-v1:8080"
+entity-recognition-service.url: "http://entity-recognition-service-v1:8080"
 
 server:
   port: 8080

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

@@ -2,9 +2,11 @@ package com.iqser.red.service.redaction.v1.server;
 
 import com.amazonaws.services.s3.AmazonS3;
 import com.fasterxml.jackson.databind.ObjectMapper;
-import com.iqser.red.service.configuration.v1.api.model.*;
+import com.iqser.red.service.persistence.service.v1.api.model.FileType;
-import com.iqser.red.service.configuration.v1.api.resource.DictionaryResource;
+import com.iqser.red.service.persistence.service.v1.api.model.JSONPrimitive;
-import com.iqser.red.service.file.management.v1.api.model.FileType;
+import com.iqser.red.service.persistence.service.v1.api.model.data.configuration.Colors;
+import com.iqser.red.service.persistence.service.v1.api.model.data.configuration.DictionaryEntry;
+import com.iqser.red.service.persistence.service.v1.api.model.data.configuration.Type;
 import com.iqser.red.service.redaction.v1.model.*;
 import com.iqser.red.service.redaction.v1.server.classification.model.SectionText;
 import com.iqser.red.service.redaction.v1.server.client.DictionaryClient;
@@ -13,12 +15,14 @@ import com.iqser.red.service.redaction.v1.server.client.LegalBasisClient;
 import com.iqser.red.service.redaction.v1.server.client.RulesClient;
 import com.iqser.red.service.redaction.v1.server.controller.RedactionController;
 import com.iqser.red.service.redaction.v1.server.memory.MemoryStats;
-import com.iqser.red.service.redaction.v1.server.redaction.service.ReanalyzeService;
+import com.iqser.red.service.redaction.v1.server.redaction.service.AnalyzeService;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
 import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
 import com.iqser.red.storage.commons.service.StorageService;
+
 import lombok.SneakyThrows;
+
 import org.apache.commons.io.IOUtils;
 import org.junit.After;
 import org.junit.Before;
@@ -84,12 +88,11 @@ public class RedactionIntegrationTest {
 
     private static final String PII = "PII";
 
-
     @Autowired
     private RedactionController redactionController;
 
     @Autowired
-    private ReanalyzeService reanalyzeService;
+    private AnalyzeService analyzeService;
 
     @Autowired
     private ObjectMapper objectMapper;
@@ -127,9 +130,10 @@ public class RedactionIntegrationTest {
     private final Map<String, Integer> rankTypeMap = new HashMap<>();
     private final Colors colors = new Colors();
     private final Map<String, Long> reanlysisVersions = new HashMap<>();
+    private final Set<String> deleted = new HashSet<>();
 
-    private final static String TEST_RULESET_ID = "123";
+    private final static String TEST_DOSSIER_TEMPLATE_ID = "123";
-    private final static String TEST_PROJECT_ID = "123";
+    private final static String TEST_DOSSIER_ID = "123";
     private final static String TEST_FILE_ID = "123";
 
     @Configuration
@@ -152,18 +156,20 @@ public class RedactionIntegrationTest {
             return kieServices.newKieContainer(kieModule.getReleaseId());
         }
 
+
         @Bean
         @Primary
         public StorageService inmemoryStorage() {
+
             return new FileSystemBackedStorageService();
         }
 
-
     }
 
 
     @After
     public void cleanupStorage() {
+
         if (this.storageService instanceof FileSystemBackedStorageService) {
             ((FileSystemBackedStorageService) this.storageService).clearStorage();
         }
@@ -173,51 +179,64 @@ public class RedactionIntegrationTest {
     @Before
     public void stubClients() {
 
-        when(rulesClient.getVersion(TEST_RULESET_ID)).thenReturn(0L);
+        when(rulesClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(0L);
-        when(rulesClient.getRules(TEST_RULESET_ID)).thenReturn(new RulesResponse(RULES));
+        when(rulesClient.getRules(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(JSONPrimitive.of(RULES));
 
         loadDictionaryForTest();
         loadTypeForTest();
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(0L);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(0L);
-        when(dictionaryClient.getAllTypes(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(TypeResponse.builder()
+        when(dictionaryClient.getAllTypesForDossierTemplate(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(getTypeResponse());
-                .types(getTypeResponse())
-                .build());
 
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, TEST_PROJECT_ID)).thenReturn(0L);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(0L);
-        when(dictionaryClient.getAllTypes(TEST_RULESET_ID, TEST_PROJECT_ID)).thenReturn(TypeResponse.builder()
+        when(dictionaryClient.getAllTypesForDossier(TEST_DOSSIER_ID)).thenReturn(List.of(Type.builder()
-                .types(List.of(TypeResult.builder()
+                .id(DOSSIER_REDACTIONS +":"+ TEST_DOSSIER_TEMPLATE_ID)
                 .type(DOSSIER_REDACTIONS)
-                        .ruleSetId(TEST_RULESET_ID)
+                        .dossierTemplateId(TEST_DOSSIER_ID)
-                        .hexColor( "#ffe187")
+                        .hexColor("#ffe187")
                         .isHint(hintTypeMap.get(DOSSIER_REDACTIONS))
                         .isCaseInsensitive(caseInSensitiveMap.get(DOSSIER_REDACTIONS))
                         .isRecommendation(recommendationTypeMap.get(DOSSIER_REDACTIONS))
                         .rank(rankTypeMap.get(DOSSIER_REDACTIONS))
-                        .build()))
+                        .build()));
-                .build());
 
-        when(dictionaryClient.getDictionaryForType(VERTEBRATE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(VERTEBRATE, false));
+        when(dictionaryClient.getDictionaryForType(VERTEBRATE + ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(ADDRESS, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(ADDRESS, false));
+                .thenReturn(getDictionaryResponse(VERTEBRATE, false));
-        when(dictionaryClient.getDictionaryForType(AUTHOR, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(AUTHOR, false));
+        when(dictionaryClient.getDictionaryForType(ADDRESS+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(SPONSOR, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(SPONSOR, false));
+                .thenReturn(getDictionaryResponse(ADDRESS, false));
-        when(dictionaryClient.getDictionaryForType(NO_REDACTION_INDICATOR, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(NO_REDACTION_INDICATOR, false));
+        when(dictionaryClient.getDictionaryForType(AUTHOR+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(REDACTION_INDICATOR, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(REDACTION_INDICATOR, false));
+                .thenReturn(getDictionaryResponse(AUTHOR, false));
-        when(dictionaryClient.getDictionaryForType(HINT_ONLY, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(HINT_ONLY, false));
+        when(dictionaryClient.getDictionaryForType(SPONSOR+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(MUST_REDACT, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(MUST_REDACT, false));
+                .thenReturn(getDictionaryResponse(SPONSOR, false));
-        when(dictionaryClient.getDictionaryForType(PUBLISHED_INFORMATION, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(PUBLISHED_INFORMATION, false));
+        when(dictionaryClient.getDictionaryForType(NO_REDACTION_INDICATOR+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(TEST_METHOD, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(TEST_METHOD, false));
+                .thenReturn(getDictionaryResponse(NO_REDACTION_INDICATOR, false));
-        when(dictionaryClient.getDictionaryForType(PII, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(PII, false));
+        when(dictionaryClient.getDictionaryForType(REDACTION_INDICATOR+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_AUTHOR, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(RECOMMENDATION_AUTHOR, false));
+                .thenReturn(getDictionaryResponse(REDACTION_INDICATOR, false));
-        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_ADDRESS, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(RECOMMENDATION_ADDRESS, false));
+        when(dictionaryClient.getDictionaryForType(HINT_ONLY+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(FALSE_POSITIVE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(FALSE_POSITIVE, false));
+                .thenReturn(getDictionaryResponse(HINT_ONLY, false));
-        when(dictionaryClient.getDictionaryForType(PURITY, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(PURITY, false));
+        when(dictionaryClient.getDictionaryForType(MUST_REDACT+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(IMAGE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(IMAGE, false));
+                .thenReturn(getDictionaryResponse(MUST_REDACT, false));
-        when(dictionaryClient.getDictionaryForType(OCR, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(OCR, false));
+        when(dictionaryClient.getDictionaryForType(PUBLISHED_INFORMATION+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(LOGO, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(LOGO, false));
+                .thenReturn(getDictionaryResponse(PUBLISHED_INFORMATION, false));
-        when(dictionaryClient.getDictionaryForType(SIGNATURE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(SIGNATURE, false));
+        when(dictionaryClient.getDictionaryForType(TEST_METHOD+ ":" + TEST_DOSSIER_TEMPLATE_ID))
-        when(dictionaryClient.getDictionaryForType(FORMULA, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(FORMULA, false));
+                .thenReturn(getDictionaryResponse(TEST_METHOD, false));
-        when(dictionaryClient.getDictionaryForType(DOSSIER_REDACTIONS, TEST_RULESET_ID, TEST_PROJECT_ID)).thenReturn(getDictionaryResponse(DOSSIER_REDACTIONS, true));
+        when(dictionaryClient.getDictionaryForType(PII+ ":" + TEST_DOSSIER_TEMPLATE_ID)).thenReturn(getDictionaryResponse(PII, false));
-        when(dictionaryClient.getColors(TEST_RULESET_ID)).thenReturn(colors);
+        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_AUTHOR+ ":" + TEST_DOSSIER_TEMPLATE_ID))
+                .thenReturn(getDictionaryResponse(RECOMMENDATION_AUTHOR, false));
+        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_ADDRESS+ ":" + TEST_DOSSIER_TEMPLATE_ID))
+                .thenReturn(getDictionaryResponse(RECOMMENDATION_ADDRESS, false));
+        when(dictionaryClient.getDictionaryForType(FALSE_POSITIVE+ ":" + TEST_DOSSIER_TEMPLATE_ID))
+                .thenReturn(getDictionaryResponse(FALSE_POSITIVE, false));
+        when(dictionaryClient.getDictionaryForType(PURITY+ ":" + TEST_DOSSIER_TEMPLATE_ID))
+                .thenReturn(getDictionaryResponse(PURITY, false));
+        when(dictionaryClient.getDictionaryForType(IMAGE+ ":" + TEST_DOSSIER_TEMPLATE_ID)).thenReturn(getDictionaryResponse(IMAGE, false));
+        when(dictionaryClient.getDictionaryForType(OCR+ ":" + TEST_DOSSIER_TEMPLATE_ID)).thenReturn(getDictionaryResponse(OCR, false));
+        when(dictionaryClient.getDictionaryForType(LOGO+ ":" + TEST_DOSSIER_TEMPLATE_ID)).thenReturn(getDictionaryResponse(LOGO, false));
+        when(dictionaryClient.getDictionaryForType(SIGNATURE+ ":" + TEST_DOSSIER_TEMPLATE_ID))
+                .thenReturn(getDictionaryResponse(SIGNATURE, false));
+        when(dictionaryClient.getDictionaryForType(FORMULA+ ":" + TEST_DOSSIER_TEMPLATE_ID))
+                .thenReturn(getDictionaryResponse(FORMULA, false));
+        when(dictionaryClient.getDictionaryForType(DOSSIER_REDACTIONS+ ":" + TEST_DOSSIER_TEMPLATE_ID)).thenReturn(getDictionaryResponse(DOSSIER_REDACTIONS, true));
+        when(dictionaryClient.getColors(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(colors);
     }
 
 
@@ -455,13 +474,14 @@ public class RedactionIntegrationTest {
     }
 
 
-    private List<TypeResult> getTypeResponse() {
+    private List<Type> getTypeResponse() {
 
         return typeColorMap.entrySet()
                 .stream()
-                .map(typeColor -> TypeResult.builder()
+                .map(typeColor -> Type.builder()
+                        .id(typeColor.getKey() + ":" + TEST_DOSSIER_TEMPLATE_ID)
                         .type(typeColor.getKey())
-                        .ruleSetId(TEST_RULESET_ID)
+                        .dossierTemplateId(TEST_DOSSIER_TEMPLATE_ID)
                         .hexColor(typeColor.getValue())
                         .isHint(hintTypeMap.get(typeColor.getKey()))
                         .isCaseInsensitive(caseInSensitiveMap.get(typeColor.getKey()))
@@ -473,11 +493,13 @@ public class RedactionIntegrationTest {
     }
 
 
-    private DictionaryResponse getDictionaryResponse(String type, boolean isDossierDictionary) {
+    private Type getDictionaryResponse(String type, boolean isDossierDictionary) {
 
-        return DictionaryResponse.builder()
+        return Type.builder()
+                .id(type + ":" +TEST_DOSSIER_TEMPLATE_ID)
                 .hexColor(typeColorMap.get(type))
-                .entries(isDossierDictionary ? toDictionaryEntry(dossierDictionary.get(type)) : toDictionaryEntry(dictionary.get(type)))
+                .entries(isDossierDictionary ? toDictionaryEntry(dossierDictionary.get(type)) : toDictionaryEntry(dictionary
+                        .get(type)))
                 .isHint(hintTypeMap.get(type))
                 .isCaseInsensitive(caseInSensitiveMap.get(type))
                 .isRecommendation(recommendationTypeMap.get(type))
@@ -490,7 +512,11 @@ public class RedactionIntegrationTest {
 
         List<DictionaryEntry> dictionaryEntries = new ArrayList<>();
         entries.forEach(entry -> {
-            dictionaryEntries.add(new DictionaryEntry(entry, reanlysisVersions.containsKey(entry) ? reanlysisVersions.get(entry) : 0L, false));
+            dictionaryEntries.add(DictionaryEntry.builder()
+                    .value(entry)
+                    .version(reanlysisVersions.containsKey(entry) ? reanlysisVersions.get(entry) : 0L)
+                    .deleted(deleted
+                            .contains(entry) ? true : false).build());
         });
         return dictionaryEntries;
     }
@@ -498,9 +524,11 @@ public class RedactionIntegrationTest {
 
     @Test
     public void test270Rotated() {
+
         AnalyzeRequest request = prepareStorage("files/Minimal Examples/270Rotated.pdf");
         MemoryStats.printMemoryStats();
-        AnalyzeResult result = reanalyzeService.analyze(request);
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        AnalyzeResult result = analyzeService.analyze(request);
         assertThat(result).isNotNull();
     }
 
@@ -508,12 +536,15 @@ public class RedactionIntegrationTest {
     @Test
     @Ignore
     public void testLargeScannedFileOOM() {
+
         AnalyzeRequest request = prepareStorage("scanned/VV-377031.pdf");
         MemoryStats.printMemoryStats();
-        AnalyzeResult result = reanalyzeService.analyze(request);
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        AnalyzeResult result = analyzeService.analyze(request);
         assertThat(result).isNotNull();
     }
 
+
     @Test
     public void testMergedImages() throws IOException {
 
@@ -521,11 +552,12 @@ public class RedactionIntegrationTest {
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/merge_images.pdf");
 
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
-        AnalyzeResult result = reanalyzeService.analyze(request);
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        AnalyzeResult result = analyzeService.analyze(request);
 
         Map<String, List<RedactionLogEntry>> duplicates = new HashMap<>();
 
-        var redactionLog = redactionStorageService.getRedactionLog(TEST_PROJECT_ID, TEST_FILE_ID);
+        var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
 
         redactionLog.getRedactionLogEntry().forEach(entry -> {
             duplicates.computeIfAbsent(entry.getId(), v -> new ArrayList<>()).add(entry);
@@ -536,10 +568,10 @@ public class RedactionIntegrationTest {
         });
 
         dictionary.get(AUTHOR).add("Drinking water");
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(1L);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(1L);
 
         AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
-                .projectId(TEST_PROJECT_ID)
+                .dossierId(TEST_DOSSIER_ID)
                 .fileId(TEST_FILE_ID)
                 .build());
 
@@ -547,19 +579,18 @@ public class RedactionIntegrationTest {
             fileOutputStream.write(annotateResponse.getDocument());
         }
         long rstart = System.currentTimeMillis();
-        reanalyzeService.reanalyze(request);
+        analyzeService.reanalyze(request);
 
         long rend = System.currentTimeMillis();
         System.out.println("reanalysis analysis duration: " + (rend - rstart));
 
-
         long end = System.currentTimeMillis();
 
         System.out.println("duration: " + (end - start));
 
-
     }
 
+
     @Test
     @Ignore
     public void noExceptionShouldBeThrownForAnyFiles() throws IOException {
@@ -577,11 +608,15 @@ public class RedactionIntegrationTest {
 
             AnalyzeRequest request = prepareStorage(new FileInputStream((path)));
             System.out.println("Redacting file : " + path.getName());
-            AnalyzeResult result = reanalyzeService.analyze(request);
+            analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+
+            long fstart = System.currentTimeMillis();
+            AnalyzeResult result = analyzeService.analyze(request);
+            System.out.println("analysis analysis duration: " + (System.currentTimeMillis() - fstart));
 
             Map<String, List<RedactionLogEntry>> duplicates = new HashMap<>();
 
-            var redactionLog = redactionStorageService.getRedactionLog(TEST_PROJECT_ID, TEST_FILE_ID);
+            var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
 
             redactionLog.getRedactionLogEntry().forEach(entry -> {
                 duplicates.computeIfAbsent(entry.getId(), v -> new ArrayList<>()).add(entry);
@@ -592,10 +627,10 @@ public class RedactionIntegrationTest {
             });
 
             dictionary.get(AUTHOR).add("Drinking water");
-            when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(1L);
+            when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(1L);
 
             long rstart = System.currentTimeMillis();
-            reanalyzeService.reanalyze(request);
+            analyzeService.reanalyze(request);
 
             long rend = System.currentTimeMillis();
             System.out.println("reanalysis analysis duration: " + (rend - rstart));
@@ -631,13 +666,22 @@ public class RedactionIntegrationTest {
     public void redactionTest() throws IOException {
 
         long start = System.currentTimeMillis();
-        ClassPathResource pdfFileResource = new ClassPathResource("files/new/Single Study - Oral (Gavage) Mouse.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/S11.pdf");
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
+        request.setExcludedPages(Set.of(1));
 
-        AnalyzeResult result = reanalyzeService.analyze(request);
+        request.setFileAttributes(List.of(FileAttribute.builder()
+                .id("fileAttributeId")
+                .label("Vertebrate Study")
+                .placeholder("{fileattributes.vertebrateStudy}")
+                .value("true")
+                .build()));
 
-        var redactionLog = redactionStorageService.getRedactionLog(TEST_PROJECT_ID, TEST_FILE_ID);
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
-        var text = redactionStorageService.getText(TEST_PROJECT_ID, TEST_FILE_ID);
+        AnalyzeResult result = analyzeService.analyze(request);
+
+        var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
+        var text = redactionStorageService.getText(TEST_DOSSIER_ID, TEST_FILE_ID);
 
         redactionLog.getRedactionLogEntry().forEach(entry -> {
             if (entry.isImage()) {
@@ -650,7 +694,7 @@ public class RedactionIntegrationTest {
         System.out.println("first analysis duration: " + (end - start));
 
         try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Test.json")) {
-            fileOutputStream.write(objectMapper.writeValueAsBytes(redactionStorageService.getText(TEST_PROJECT_ID, TEST_FILE_ID)));
+            fileOutputStream.write(objectMapper.writeValueAsBytes(redactionStorageService.getText(TEST_DOSSIER_ID, TEST_FILE_ID)));
         }
 
         int correctFound = 0;
@@ -680,21 +724,43 @@ public class RedactionIntegrationTest {
         dictionary.get(AUTHOR).add("physical");
         reanlysisVersions.put("physical", 2L);
 
-        dictionary.get(VERTEBRATE).add("s-metolachlor");
+        deleted.add("David Chubb");
-        reanlysisVersions.put("s-metolachlor", 3L);
+        deleted.add("mouse");
 
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(3L);
+        dictionary.get(FALSE_POSITIVE).add("David Chubb");
+        reanlysisVersions.put("David Chubb", 3L);
 
-        when(dictionaryClient.getDictionaryForType(VERTEBRATE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(VERTEBRATE, false));
+        reanlysisVersions.put("mouse", 3L);
+
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(3L);
+
+        when(dictionaryClient.getDictionaryForType(VERTEBRATE))
+                .thenReturn(getDictionaryResponse(VERTEBRATE, false));
+
+        when(dictionaryClient.getDictionaryForType(FALSE_POSITIVE))
+                .thenReturn(getDictionaryResponse(FALSE_POSITIVE, false));
 
         start = System.currentTimeMillis();
-        AnalyzeResult reanalyzeResult = reanalyzeService.reanalyze(request);
+
+        ManualRedactions manualRedactions = new ManualRedactions();
+
+        manualRedactions.setImageRecategorizations(Set.of(ManualImageRecategorization.builder()
+                .id("37eee3e9d589a5cc529bfec38c3ba479")
+                .status(Status.APPROVED)
+                .type("signature")
+                .build()));
+
+        request.setManualRedactions(manualRedactions);
+
+        AnalyzeResult reanalyzeResult = analyzeService.reanalyze(request);
+
+        redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
 
         end = System.currentTimeMillis();
         System.out.println("reanalysis analysis duration: " + (end - start));
 
         AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
-                .projectId(TEST_PROJECT_ID)
+                .dossierId(TEST_DOSSIER_ID)
                 .fileId(TEST_FILE_ID)
                 .build());
 
@@ -702,6 +768,20 @@ public class RedactionIntegrationTest {
             fileOutputStream.write(annotateResponse.getDocument());
         }
 
+
+        deleted.remove("mouse");
+        reanlysisVersions.put("mouse", 4L);
+
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(4L);
+
+        when(dictionaryClient.getDictionaryForType(VERTEBRATE))
+                .thenReturn(getDictionaryResponse(VERTEBRATE, false));
+
+        analyzeService.reanalyze(request);
+
+        redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
+
+        System.out.println("hi");
     }
 
 
@@ -712,10 +792,11 @@ public class RedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = prepareStorage("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");
-        AnalyzeResult result = reanalyzeService.analyze(request);
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
-                .projectId(TEST_PROJECT_ID)
+                .dossierId(TEST_DOSSIER_ID)
                 .fileId(TEST_FILE_ID)
                 .build());
 
@@ -771,26 +852,31 @@ public class RedactionIntegrationTest {
 
 //        manualRedactions.getEntriesToAdd().add(manualRedactionEntry);
 
-
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
         request.setManualRedactions(manualRedactions);
-        AnalyzeResult result = reanalyzeService.analyze(request);
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        AnalyzeResult result = analyzeService.analyze(request);
 
         manualRedactions.getEntriesToAdd().add(manualRedactionEntry);
         manualRedactions.setIdsToRemove(Set.of(IdRemoval.builder()
                 .id("5b940b2cb401ed9f5be6fc24f6e77bcf")
                 .status(Status.APPROVED)
                 .build()));
+        manualRedactions.setManualLegalBasisChanges(Set.of(ManualLegalBasisChange.builder()
+                .id("675eba69b0c2917de55462c817adaa05")
+                .legalBasis("Manual Legal Basis Change")
+                .status(Status.APPROVED)
+                .build()));
 
-        reanalyzeService.reanalyze(request);
+        analyzeService.reanalyze(request);
 
+        var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
 
         AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
-                .projectId(TEST_PROJECT_ID)
+                .dossierId(TEST_DOSSIER_ID)
                 .fileId(TEST_FILE_ID)
                 .build());
 
-
         try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Annotated.pdf")) {
             fileOutputStream.write(annotateResponse.getDocument());
         }
@@ -805,15 +891,14 @@ public class RedactionIntegrationTest {
     public void classificationTest() throws IOException {
 
         System.out.println("classificationTest");
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Trinexapac/93 Trinexapac-ethyl_RAR_03_Volume_3CA_B-1_2017-03-31.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/S11.pdf");
-
 
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
         RedactionRequest redactionRequest = RedactionRequest.builder()
-                .projectId(request.getProjectId())
+                .dossierId(request.getDossierId())
                 .fileId(request.getFileId())
-                .ruleSetId(request.getRuleSetId())
+                .dossierTemplateId(request.getDossierTemplateId())
                 .build();
 
         RedactionResult result = redactionController.classify(redactionRequest);
@@ -828,14 +913,14 @@ public class RedactionIntegrationTest {
     public void sectionsTest() throws IOException {
 
         System.out.println("sectionsTest");
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Fludioxonil/51 " + "Fludioxonil_RAR_02_Volume_2_2018-02-21.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/S11.pdf");
 
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
         RedactionRequest redactionRequest = RedactionRequest.builder()
-                .projectId(request.getProjectId())
+                .dossierId(request.getDossierId())
                 .fileId(request.getFileId())
-                .ruleSetId(request.getRuleSetId())
+                .dossierTemplateId(request.getDossierTemplateId())
                 .build();
 
         RedactionResult result = redactionController.sections(redactionRequest);
@@ -850,14 +935,14 @@ public class RedactionIntegrationTest {
     public void htmlTablesTest() throws IOException {
 
         System.out.println("htmlTablesTest");
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");
+        ClassPathResource pdfFileResource = new ClassPathResource("files/S11.pdf");
 
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
         RedactionRequest redactionRequest = RedactionRequest.builder()
-                .projectId(request.getProjectId())
+                .dossierId(request.getDossierId())
                 .fileId(request.getFileId())
-                .ruleSetId(request.getRuleSetId())
+                .dossierTemplateId(request.getDossierTemplateId())
                 .build();
 
         RedactionResult result = redactionController.htmlTables(redactionRequest);
@@ -877,9 +962,9 @@ public class RedactionIntegrationTest {
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
         RedactionRequest redactionRequest = RedactionRequest.builder()
-                .projectId(request.getProjectId())
+                .dossierId(request.getDossierId())
                 .fileId(request.getFileId())
-                .ruleSetId(request.getRuleSetId())
+                .dossierTemplateId(request.getDossierTemplateId())
                 .build();
 
         RedactionResult result = redactionController.htmlTables(redactionRequest);
@@ -897,9 +982,10 @@ public class RedactionIntegrationTest {
 
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
-        AnalyzeResult result = reanalyzeService.analyze(request);
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        AnalyzeResult result = analyzeService.analyze(request);
 
-        var redactionLog = redactionStorageService.getRedactionLog(TEST_PROJECT_ID, TEST_FILE_ID);
+        var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
 
         redactionLog.getRedactionLogEntry().forEach(entry -> {
             if (!entry.isHint()) {
@@ -908,8 +994,10 @@ public class RedactionIntegrationTest {
         });
     }
 
+
     @SneakyThrows
     private AnalyzeRequest prepareStorage(String file) {
+
         ClassPathResource pdfFileResource = new ClassPathResource(file);
 
         return prepareStorage(pdfFileResource.getInputStream());
@@ -920,15 +1008,15 @@ public class RedactionIntegrationTest {
     private AnalyzeRequest prepareStorage(InputStream stream) {
 
         AnalyzeRequest request = AnalyzeRequest.builder()
-                .ruleSetId(TEST_RULESET_ID)
+                .dossierTemplateId(TEST_DOSSIER_TEMPLATE_ID)
-                .projectId(TEST_PROJECT_ID)
+                .dossierId(TEST_DOSSIER_ID)
                 .fileId(TEST_FILE_ID)
                 .lastProcessed(OffsetDateTime.now())
                 .build();
 
         var bytes = IOUtils.toByteArray(stream);
 
-        storageService.storeObject(RedactionStorageService.StorageIdUtils.getStorageId(TEST_PROJECT_ID, TEST_FILE_ID, FileType.ORIGIN), bytes);
+        storageService.storeObject(RedactionStorageService.StorageIdUtils.getStorageId(TEST_DOSSIER_ID, TEST_FILE_ID, FileType.ORIGIN), bytes);
 
         return request;
 
@@ -941,13 +1029,13 @@ public class RedactionIntegrationTest {
         long start = System.currentTimeMillis();
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/sponsor_companies.pdf");
 
-
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
-        AnalyzeResult result = reanalyzeService.analyze(request);
+        analyzeService.analyzeDocumentStructure(new StructureAnalyzeRequest(request.getDossierId(), request.getFileId()));
+        AnalyzeResult result = analyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
-                .projectId(TEST_PROJECT_ID)
+                .dossierId(TEST_DOSSIER_ID)
                 .fileId(TEST_FILE_ID)
                 .build());
 

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

@@ -0,0 +1,18 @@
+package com.iqser.red.service.redaction.v1.server.redaction.rulebuilder;
+
+import com.iqser.red.service.redaction.v1.model.RuleBuilderModel;
+import org.junit.Test;
+
+import static org.assertj.core.api.AssertionsForClassTypes.assertThat;
+
+public class RuleBuilderModelServiceTest {
+
+    @Test
+    public void testRuleBuilderModelProvider() {
+
+        RuleBuilderModel model = new RuleBuilderModelService().getRuleBuilderModel();
+
+        assertThat(model.getWhenClauses().size()).isGreaterThan(1);
+        assertThat(model.getThenConditions().size()).isGreaterThan(1);
+    }
+}

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

@@ -1,511 +0,0 @@
-package com.iqser.red.service.redaction.v1.server.redaction.service;
-
-import com.amazonaws.services.s3.AmazonS3;
-import com.iqser.red.service.configuration.v1.api.model.*;
-import com.iqser.red.service.configuration.v1.api.resource.DictionaryResource;
-import com.iqser.red.service.redaction.v1.server.Application;
-import com.iqser.red.service.redaction.v1.server.FileSystemBackedStorageService;
-import com.iqser.red.service.redaction.v1.server.classification.model.Document;
-import com.iqser.red.service.redaction.v1.server.client.DictionaryClient;
-import com.iqser.red.service.redaction.v1.server.client.LegalBasisClient;
-import com.iqser.red.service.redaction.v1.server.client.RulesClient;
-import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.EntitySearchUtils;
-import com.iqser.red.service.redaction.v1.server.redaction.utils.ResourceLoader;
-import com.iqser.red.service.redaction.v1.server.segmentation.PdfSegmentationService;
-import com.iqser.red.storage.commons.service.StorageService;
-import org.junit.Before;
-import org.junit.Ignore;
-import org.junit.Test;
-import org.junit.runner.RunWith;
-import org.kie.api.KieServices;
-import org.kie.api.builder.KieBuilder;
-import org.kie.api.builder.KieFileSystem;
-import org.kie.api.builder.KieModule;
-import org.kie.api.runtime.KieContainer;
-import org.springframework.beans.factory.annotation.Autowired;
-import org.springframework.boot.autoconfigure.EnableAutoConfiguration;
-import org.springframework.boot.autoconfigure.amqp.RabbitAutoConfiguration;
-import org.springframework.boot.test.context.SpringBootTest;
-import org.springframework.boot.test.mock.mockito.MockBean;
-import org.springframework.context.annotation.Bean;
-import org.springframework.context.annotation.Configuration;
-import org.springframework.context.annotation.Import;
-import org.springframework.context.annotation.Primary;
-import org.springframework.core.io.ClassPathResource;
-import org.springframework.test.context.junit4.SpringRunner;
-
-import java.io.BufferedReader;
-import java.io.ByteArrayInputStream;
-import java.io.IOException;
-import java.io.InputStream;
-import java.io.InputStreamReader;
-import java.net.URL;
-import java.nio.charset.StandardCharsets;
-import java.util.*;
-import java.util.concurrent.atomic.AtomicLong;
-
-import static org.assertj.core.api.Assertions.assertThat;
-import static org.mockito.Mockito.when;
-
-@RunWith(SpringRunner.class)
-@SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT)
-@Import(EntityRedactionServiceTest.RedactionIntegrationTestConfiguration.class)
-public class EntityRedactionServiceTest {
-
-    private static final String DEFAULT_RULES = loadFromClassPath("drools/rules.drl");
-    private static final String AUTHOR_CODE = "author";
-    private static final String ADDRESS_CODE = "address";
-    private static final String SPONSOR_CODE = "sponsor";
-
-    private static final AtomicLong DICTIONARY_VERSION = new AtomicLong();
-    private static final AtomicLong RULES_VERSION = new AtomicLong();
-
-    @MockBean
-    private DictionaryClient dictionaryClient;
-
-    @MockBean
-    private RulesClient rulesClient;
-
-    @Autowired
-    private EntityRedactionService entityRedactionService;
-
-    @Autowired
-    private PdfSegmentationService pdfSegmentationService;
-
-    @Autowired
-    private DroolsExecutionService droolsExecutionService;
-
-    @MockBean
-    private AmazonS3 amazonS3;
-
-    @MockBean
-    private LegalBasisClient legalBasisClient;
-
-    private final static String TEST_RULESET_ID = "123";
-
-    @Configuration
-    @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class})
-    public static class RedactionIntegrationTestConfiguration {
-
-        @Bean
-        public KieContainer kieContainer() {
-
-            KieServices kieServices = KieServices.Factory.get();
-
-            KieFileSystem kieFileSystem = kieServices.newKieFileSystem();
-            InputStream input = new ByteArrayInputStream(DEFAULT_RULES.getBytes(StandardCharsets.UTF_8));
-            kieFileSystem.write("src/test/resources/drools/rules.drl", kieServices.getResources()
-                    .newInputStreamResource(input));
-            KieBuilder kieBuilder = kieServices.newKieBuilder(kieFileSystem);
-            kieBuilder.buildAll();
-            KieModule kieModule = kieBuilder.getKieModule();
-
-            return kieServices.newKieContainer(kieModule.getReleaseId());
-        }
-
-
-        @Bean
-        @Primary
-        public StorageService inmemoryStorage() {
-            return new FileSystemBackedStorageService();
-        }
-
-    }
-
-
-    @Test
-    public void testNestedEntitiesRemoval() {
-
-        Set<Entity> entities = new HashSet<>();
-        Entity nested = new Entity("nested", "fake type", 10, 16, "fake headline", 0, false, false);
-        Entity nesting = new Entity("nesting nested", "fake type", 2, 16, "fake headline", 0, false, false);
-        entities.add(nested);
-        entities.add(nesting);
-        EntitySearchUtils.removeEntitiesContainedInLarger(entities);
-
-        assertThat(entities.size()).isEqualTo(1);
-        assertThat(entities).contains(nesting);
-
-    }
-
-
-    @Test
-    public void testTableRedaction() throws IOException {
-
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Single Table.pdf");
-
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Arrays.asList("Casey, H.W.", "O’Loughlin,  C.K.", "Salamon, C.M.", "Smith, S.H.")))
-                .build();
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Collections.singletonList("Toxigenics, Inc., Decatur, IL 62526, USA")))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-
-        DictionaryResponse sponsorResponse = DictionaryResponse.builder()
-                .entries(Collections.emptyList())
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
-
-        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-        assertThat(classifiedDoc.getEntities().get(1)).hasSize(7);// 3 author cells, 1 address, 1 Y and 2 N entities
-    }
-
-
-    @Test
-    public void testNestedRedaction() throws IOException {
-
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/nested_redaction.pdf");
-
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Arrays.asList("Casey, H.W.", "O’Loughlin,  C.K.", "Salamon, C.M.", "Smith, S.H.")))
-                .build();
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Collections.singletonList("Toxigenics, Inc., Decatur, IL 62526, USA")))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-        DictionaryResponse sponsorResponse = DictionaryResponse.builder()
-                .entries(Collections.emptyList())
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
-
-        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-        assertThat(classifiedDoc.getEntities().get(1)).hasSize(7);// 3 author cells, 1 address, 1 Y and 2 N entities
-    }
-
-
-    @Test
-    public void testTrueNegativesInTable() throws IOException {
-
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Cyprodinil/40 Cyprodinil - EU AIR3 - LCA Section 1" +
-                " Supplement - Identity of the active substance - Reference list.pdf");
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(new ArrayList<>(ResourceLoader.load("dictionaries/CBI_author.txt"))))
-                .build();
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(new ArrayList<>(ResourceLoader.load("dictionaries/CBI_address.txt"))))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-        DictionaryResponse sponsorResponse = DictionaryResponse.builder()
-                .entries(Collections.emptyList())
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
-        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()
-                .entrySet()
-                .stream()
-                .noneMatch(entry -> entry.getValue().stream().anyMatch(e -> e.getMatchedRule() == 9))).isTrue();
-        pdfFileResource = new ClassPathResource("files/Compounds/27 A8637C - EU AIR3 - MCP Section 1 - Identity of " +
-                "the plant protection product.pdf");
-        classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()
-                .entrySet()
-                .stream()
-                .noneMatch(entry -> entry.getValue().stream().anyMatch(e -> e.getMatchedRule() == 9))).isTrue();
-    }
-
-    @Test
-    public void testFalsePositiveInWrongCell() throws IOException {
-
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Row With Ambiguous Redaction.pdf");
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(new ArrayList<>(ResourceLoader.load("dictionaries/CBI_author.txt"))))
-                .build();
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(new ArrayList<>(ResourceLoader.load("dictionaries/CBI_address.txt"))))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-        DictionaryResponse sponsorResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(new ArrayList<>(ResourceLoader.load("dictionaries/CBI_sponsor.txt"))))
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
-        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-        assertThat(classifiedDoc.getEntities().get(1).stream()
-                .filter(entity -> entity.getMatchedRule() == 9)
-                .count()).isEqualTo(10);
-
-    }
-
-
-    @Test
-    public void testApplicantInTableRedaction() throws IOException {
-
-        String tableRules = "package drools\n" +
-                "\n" +
-                "import com.iqser.red.service.redaction.v1.server.redaction.model.Section\n" +
-                "\n" +
-                "global Section section\n" +
-                "rule \"6: Redact contact information if applicant is found\"\n" +
-                "   when\n" +
-                "       eval(section.headlineContainsWord(\"applicant\") || section.getText().contains(\"Applicant\"));\n" +
-                "   then\n" +
-                "       section.redactLineAfter(\"Name:\", \"address\", 6,true, \"Applicant information was found\", \"Reg" +
-                " (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactBetween(\"Address:\", \"Contact\", \"address\", 6,true, \"Applicant information was found\", \"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Contact point:\",  \"address\", 6,true, \"Applicant information was found\", \"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Phone:\", \"address\", 6,true, \"Applicant information was found\", " +
-                "\"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Fax:\", \"address\", 6,true, \"Applicant information was found\", \"Reg " +
-                "(EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Tel.:\", \"address\", 6,true, \"Applicant information was found\", \"Reg" +
-                " (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Tel:\", \"address\", 6,true, \"Applicant information was found\", \"Reg " +
-                "(EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"E-mail:\", \"address\", 6,true, \"Applicant information was found\", " +
-                "\"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Email:\", \"address\", 6,true, \"Applicant information was found\", " +
-                "\"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Contact:\", \"address\", 6,true, \"Applicant information was found\", " +
-                "\"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Telephone number:\", \"address\", 6,true, \"Applicant information was found\", \"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Fax number:\", \"address\", 6,true, \"Applicant information was found\"," +
-                " \"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactLineAfter(\"Telephone:\", \"address\", 6,true, \"Applicant information was found\", " +
-                "\"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactBetween(\"No:\", \"Fax\", \"address\", 6,true, \"Applicant information was found\", " +
-                "\"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "       section.redactBetween(\"Contact:\", \"Tel.:\", \"address\", 6,true, \"Applicant information was found\", \"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "   end";
-        when(rulesClient.getVersion(TEST_RULESET_ID)).thenReturn(RULES_VERSION.incrementAndGet());
-        when(rulesClient.getRules(TEST_RULESET_ID)).thenReturn(new RulesResponse(tableRules));
-        droolsExecutionService.updateRules(TEST_RULESET_ID);
-
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Applicant Producer Table.pdf");
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(new ArrayList<>(ResourceLoader.load("dictionaries/CBI_author.txt"))))
-                .build();
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(new ArrayList<>(ResourceLoader.load("dictionaries/CBI_address.txt"))))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-        DictionaryResponse sponsorResponse = DictionaryResponse.builder()
-                .entries(Collections.emptyList())
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
-        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-        assertThat(classifiedDoc.getEntities().get(1).stream()
-                .filter(entity -> entity.getMatchedRule() == 6)
-                .count()).isEqualTo(13);
-
-    }
-
-
-    @Test
-    public void testSponsorInCell() throws IOException {
-
-        String tableRules = "package drools\n" +
-                "\n" +
-                "import com.iqser.red.service.redaction.v1.server.redaction.model.Section\n" +
-                "\n" +
-                "global Section section\n" + "rule \"11: Redact sponsor company\"\n" + "    when\n" + "        " +
-                "Section(searchText.toLowerCase().contains(\"batches produced at\"))\n" + "    then\n" + "        section" +
-                ".redactIfPrecededBy(\"batches produced at\", \"sponsor\", 11, \"Redacted because it represents a " +
-                "sponsor company\", \"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" + "    end";
-        when(rulesClient.getVersion(TEST_RULESET_ID)).thenReturn(RULES_VERSION.incrementAndGet());
-        when(rulesClient.getRules(TEST_RULESET_ID)).thenReturn(new RulesResponse(tableRules));
-        droolsExecutionService.updateRules(TEST_RULESET_ID);
-
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/batches_new_line.pdf");
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(Collections.emptyList())
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-        DictionaryResponse authorResponse = DictionaryResponse.builder()
-                .entries(Collections.emptyList())
-                .build();
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(authorResponse);
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(new ArrayList<>(ResourceLoader.load("dictionaries/CBI_sponsor.txt"))))
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-        assertThat(classifiedDoc.getEntities().get(1).stream()
-                .filter(entity -> entity.getMatchedRule() == 11)
-                .count()).isEqualTo(1);
-
-    }
-
-
-    @Test
-    public void headerPropagation() throws IOException {
-
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Header Propagation.pdf");
-
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Arrays.asList("Bissig R.", "Thanei P.")))
-                .build();
-
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Collections.singletonList("Novartis Crop Protection AG, Basel, Switzerland")))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-
-        DictionaryResponse sponsorResponse = DictionaryResponse.builder()
-                .entries(Collections.emptyList())
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
-        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()).hasSize(2); // two pages
-        assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(8);
-        assertThat(classifiedDoc.getEntities().get(2).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(5); // 2 names, 1 address, 2 Y
-
-        pdfFileResource = new ClassPathResource("files/Minimal Examples/Header Propagation2.pdf");
-
-        dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Arrays.asList("Tribolet, R.", "Muir, G.", "Kühne-Thu, H.", "Close, C.")))
-                .build();
-
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        addressResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Collections.singletonList("Novartis Crop Protection AG, Basel, Switzerland")))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-
-        classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-        assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(3);
-        assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(9);
-    }
-
-
-    @Test
-    @Ignore
-    public void testNGuideline() throws IOException {
-
-        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Empty Tabular Data.pdf");
-
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Collections.singletonList("Aldershof S.")))
-                .build();
-
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .entries(toDictionaryEntry(Collections.singletonList("Novartis Crop Protection AG, Basel, Switzerland")))
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-
-        DictionaryResponse sponsorResponse = DictionaryResponse.builder()
-                .entries(Collections.emptyList())
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
-        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
-        entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null, "dossierId");
-        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-        assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(6);
-    }
-
-
-    @Before
-    public void stubRedaction() {
-        String tableRules = "package drools\n" +
-                "\n" +
-                "import com.iqser.red.service.redaction.v1.server.redaction.model.Section\n" +
-                "\n" +
-                "global Section section\n" +
-                "rule \"8: Not redacted because Vertebrate Study = N\"\n" +
-                "    when\n" +
-                "        Section(rowEquals(\"Vertebrate study Y/N\", \"N\") || rowEquals(\"Vertebrate study Y/N\", \"No\"))\n" +
-                "    then\n" +
-                "        section.redactNotCell(\"Author(s)\", 8,  \"name\", false, \"Not redacted because row is not a vertebrate study\");\n" +
-                "        section.redactNot(\"address\", 8, \"Not redacted because row is not a vertebrate study\");\n" +
-                "        section.highlightCell(\"Vertebrate study Y/N\", 8, \"hint_only\");\n" +
-                "    end\n" +
-                "rule \"9: Redact Authors and Addresses in Reference Table, if it is a Vertebrate study\"\n" +
-                "    when\n" +
-                "        Section(rowEquals(\"Vertebrate study Y/N\", \"Y\") || rowEquals(\"Vertebrate study Y/N\", " +
-                "\"Yes\"))\n" +
-                "    then\n" +
-                "        section.redactCell(\"Author(s)\", 9,  \"name\", false, \"Redacted because row is a vertebrate study\", \"Reg (EC) No 1107/2009 Art. 63 (2g)\");\n" +
-                "        section.redact(\"address\", 9, \"Redacted because row is a vertebrate study\", \"Reg (EC) No" +
-                " 1107/2009 Art. 63 (2g)\");\n" +
-                "        section.highlightCell(\"Vertebrate study Y/N\", 9, \"must_redact\");\n" +
-                "    end";
-        when(rulesClient.getVersion(TEST_RULESET_ID)).thenReturn(RULES_VERSION.incrementAndGet());
-        when(rulesClient.getRules(TEST_RULESET_ID)).thenReturn(new RulesResponse(tableRules));
-        TypeResponse typeResponse = TypeResponse.builder()
-                .types(Arrays.asList(
-                        TypeResult.builder().ruleSetId(TEST_RULESET_ID).type(AUTHOR_CODE).hexColor("#ffff00").build(),
-                        TypeResult.builder().ruleSetId(TEST_RULESET_ID).type(ADDRESS_CODE).hexColor("#ff00ff").build(),
-                        TypeResult.builder().ruleSetId(TEST_RULESET_ID).type(SPONSOR_CODE).hexColor("#00ffff").build()))
-                .build();
-        when(dictionaryClient.getVersion(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getAllTypes(TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(typeResponse);
-
-        // Default empty return to prevent NPEs
-        DictionaryResponse dictionaryResponse = DictionaryResponse.builder()
-                .build();
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
-        DictionaryResponse addressResponse = DictionaryResponse.builder()
-                .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
-        DictionaryResponse sponsorResponse = DictionaryResponse.builder()
-                .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
-
-        Colors colors = new Colors();
-        colors.setDefaultColor("#acfc00");
-        colors.setNotRedacted("#cccccc");
-        colors.setRequestAdd("#04b093");
-        colors.setRequestRemove("#04b093");
-
-        when(dictionaryClient.getColors(TEST_RULESET_ID)).thenReturn(colors);
-    }
-
-
-    private static String loadFromClassPath(String path) {
-
-        URL resource = ResourceLoader.class.getClassLoader().getResource(path);
-        if (resource == null) {
-            throw new IllegalArgumentException("could not load classpath resource: drools/rules.drl");
-        }
-        try (BufferedReader br = new BufferedReader(new InputStreamReader(resource.openStream(), StandardCharsets.UTF_8))) {
-            StringBuilder sb = new StringBuilder();
-            String str;
-            while ((str = br.readLine()) != null) {
-                sb.append(str).append("\n");
-            }
-            return sb.toString();
-        } catch (IOException e) {
-            throw new IllegalArgumentException("could not load classpath resource: " + path, e);
-        }
-    }
-
-    private List<DictionaryEntry> toDictionaryEntry(List<String> entries) {
-        List<DictionaryEntry> dictionaryEntries = new ArrayList<>();
-        entries.forEach(entry -> {
-            dictionaryEntries.add(new DictionaryEntry(entry, 1L, false));
-        });
-        return dictionaryEntries;
-    }
-
-}

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

@@ -0,0 +1,30 @@
+package com.iqser.red.service.redaction.v1.server.redaction.utils;
+
+import static org.assertj.core.api.Assertions.assertThat;
+
+import java.util.HashSet;
+import java.util.Set;
+
+import org.junit.Test;
+
+import com.iqser.red.service.redaction.v1.model.Engine;
+import com.iqser.red.service.redaction.v1.server.redaction.model.Entity;
+
+public class EntitySearchUtilsTest {
+
+    @Test
+    public void testNestedEntitiesRemoval() {
+
+        Set<Entity> entities = new HashSet<>();
+        Entity nested = new Entity("nested", "fake type", 10, 16, "fake headline", 0, false, false, Engine.RULE);
+        Entity nesting = new Entity("nesting nested", "fake type", 2, 16, "fake headline", 0, false, false, Engine.RULE);
+        entities.add(nested);
+        entities.add(nesting);
+        EntitySearchUtils.removeEntitiesContainedInLarger(entities);
+
+        assertThat(entities.size()).isEqualTo(1);
+        assertThat(entities).contains(nesting);
+
+    }
+
+}

redaction-service-v1/redaction-service-server-v1/src/test/resources/application.yml

@@ -1,6 +1,6 @@
-configuration-service.url: "http://configuration-service-v1:8080"
 image-service.url: "http://image-service-v1:8080"
-file-management-service.url: "http://file-management-service-v1:8080"
+persistence-service.url: "http://persistence-service-v1:8080"
+entity-recognition-service.url: "localhost:8080"
 
 ribbon:
   ConnectTimeout: 600000
@@ -17,3 +17,4 @@ platform.multi-tenancy:
 
 redaction-service:
   enable-image-classification: false
+  enable-entity-recognition: false

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

@@ -1652,3 +1652,5 @@ Zoecon Corp.
 Zoecon Corp., Palo Alto, USA
 Zyma SA
 Zyma SA, Nyon, Switzerland
+Mambo-Tox Ltd. Biomedical Sciences Building Bassett Crescent East Southampton SO16 7PX UK
+Syngenta Environmental Sciences Jealott’s Hill International Research Centre Bracknell, Berkshire RG42 6EY UK

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

@@ -1676,7 +1676,6 @@ da Silva Rejane
 Das R
 Das, R.
 Daughtry, CST
-David Chubb
 David Chubb|Lorraine Britton
 David Clarke
 Davies

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

@@ -236,3 +236,4 @@ No details reported
 Not available
 Test facility
 TBD
+David Chubb

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

@@ -56,8 +56,8 @@ rule "5: Do not redact Names and Addresses if no redaction Indicator is containe
     when
         Section(matchesType("vertebrate"), matchesType("published_information"))
     then
-        section.redactNot("CBI_author", 5, "Vertebrate and Published Information found");
+        section.redactNotAndReference("CBI_author","published_information", 5, "Vertebrate and Published Information found");
-        section.redactNot("CBI_address", 5, "Vertebrate and Published Information found");
+        section.redactNotAndReference("CBI_address","published_information", 5, "Vertebrate and Published Information found");
     end
 
 
@@ -268,7 +268,7 @@ rule "18: Redact contact information if Producer is found"
 
 rule "19: Redact AUTHOR(S)"
     when
-        Section(searchText.contains("AUTHOR(S):"))
+        Section(searchText.contains("AUTHOR(S):") && fileAttributeByPlaceholderEquals("{fileattributes.vertebrateStudy}", "true"))
     then
         section.redactLinesBetween("AUTHOR(S):", "COMPLETION DATE:", "PII", 19, true, "AUTHOR(S) was found", "Reg (EC) No 1107/2009 Art. 63 (2e)");
     end