Fixed endless processing

Merge in RED/redaction-service from ParsingOrientationFix to master

* commit 'ab8a710318802d4a836c006638dddb987f5d29c2':
  Fixed text parsing orientation problem

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>
 

bamboo-specs/src/main/java/buildjob/PlanSpec.java

@@ -1,7 +1,5 @@
 package buildjob;
 
-import static com.atlassian.bamboo.specs.builders.task.TestParserTask.createJUnitParserTask;
-
 import com.atlassian.bamboo.specs.api.BambooSpec;
 import com.atlassian.bamboo.specs.api.builders.BambooKey;
 import com.atlassian.bamboo.specs.api.builders.docker.DockerConfiguration;
@@ -23,6 +21,10 @@ import com.atlassian.bamboo.specs.builders.task.VcsTagTask;
 import com.atlassian.bamboo.specs.builders.trigger.BitbucketServerTrigger;
 import com.atlassian.bamboo.specs.model.task.InjectVariablesScope;
 import com.atlassian.bamboo.specs.util.BambooServer;
+import com.atlassian.bamboo.specs.builders.task.ScriptTask;
+import com.atlassian.bamboo.specs.model.task.ScriptTaskProperties.Location;
+
+import static com.atlassian.bamboo.specs.builders.task.TestParserTask.createJUnitParserTask;
 
 /**
  * Plan configuration for Bamboo.
@@ -33,6 +35,8 @@ public class PlanSpec {
 
     private static final String SERVICE_NAME = "redaction-service";
 
+    private static final String JVM_ARGS =" -Xmx4g -XX:+ExitOnOutOfMemoryError -XX:SurvivorRatio=2 -XX:NewRatio=1 -XX:InitialTenuringThreshold=16 -XX:MaxTenuringThreshold=16 -XX:InitiatingHeapOccupancyPercent=35 ";
+
     private static final String SERVICE_KEY = SERVICE_NAME.toUpperCase().replaceAll("-", "");
 
     /**
@@ -82,20 +86,9 @@ public class PlanSpec {
                                                 .checkoutItems(new CheckoutItem().defaultRepository()),
                                         new ScriptTask()
                                                 .description("Build")
-                                                .inlineBody("#!/bin/bash\n" +
-                                                        "set -e\n" +
-
-                                                        "if [[ \"${bamboo.version_tag}\" != \"dev\" ]]; then ${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn --no-transfer-progress -f ${bamboo_build_working_directory}/" + SERVICE_NAME + "-v1/pom.xml versions:set  -DnewVersion=${bamboo.version_tag}; fi\n" +
-                                                        "if [[ \"${bamboo.version_tag}\" != \"dev\" ]]; then ${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn --no-transfer-progress -f ${bamboo_build_working_directory}/" + SERVICE_NAME + "-image-v1/pom.xml versions:set  -DnewVersion=${bamboo.version_tag}; fi\n" +
-
-                                                        "if [[ \"${bamboo.version_tag}\" = \"dev\" ]]; then ${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn -f ${bamboo_build_working_directory}/" + SERVICE_NAME + "-v1/pom.xml --no-transfer-progress clean install -Djava.security.egd=file:/dev/./urandom; fi\n" +
-                                                        "if [[ \"${bamboo.version_tag}\" != \"dev\" ]]; then ${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn -f ${bamboo_build_working_directory}/" + SERVICE_NAME + "-v1/pom.xml --no-transfer-progress clean deploy -e -DdeployAtEnd=true -Dmaven.wagon.http.ssl.insecure=true -Dmaven.wagon.http.ssl.allowall=true -Dmaven.wagon.http.ssl.ignore.validity.dates=true -DaltDeploymentRepository=iqser_release::default::https://nexus.iqser.com/repository/red-platform-releases; fi\n" +
-
-                                                        "${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn --no-transfer-progress -f ${bamboo_build_working_directory}/" + SERVICE_NAME + "-image-v1/pom.xml package\n" +
-                                                        "${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn --no-transfer-progress -f ${bamboo_build_working_directory}/" + SERVICE_NAME + "-image-v1/pom.xml docker:push\n" +
-
-                                                        "if [[ \"${bamboo.version_tag}\" = \"dev\" ]]; then echo \"gitTag=${bamboo.planRepository.1.branch}_${bamboo.buildNumber}\" > git.tag; fi\n" +
-                                                        "if [[ \"${bamboo.version_tag}\" != \"dev\" ]]; then echo \"gitTag=${bamboo.version_tag}\" > git.tag; fi\n"),
+                                                .location(Location.FILE)
+                                                .fileFromPath("bamboo-specs/src/main/resources/scripts/build-java.sh")
+                                                .argument(SERVICE_NAME),
                                         createJUnitParserTask()
                                                 .description("Resultparser")
                                                 .resultDirectories("**/test-reports/*.xml, **/target/surefire-reports/*.xml, **/target/failsafe-reports/*.xml")

bamboo-specs/src/main/resources/scripts/build-java.sh

@@ -0,0 +1,51 @@
+#!/bin/bash
+set -e
+
+SERVICE_NAME=$1
+
+if [[ "${bamboo_version_tag}" = "dev" ]]
+then 
+    ${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn \
+        -f ${bamboo_build_working_directory}/$SERVICE_NAME-v1/pom.xml \
+        --no-transfer-progress \
+        clean install \
+        -Djava.security.egd=file:/dev/./urandomelse
+else        
+    ${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn \
+        --no-transfer-progress \
+        -f ${bamboo_build_working_directory}/$SERVICE_NAME-v1/pom.xml \
+        versions:set \
+        -DnewVersion=${bamboo_version_tag}
+    ${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn \
+        --no-transfer-progress \
+        -f ${bamboo_build_working_directory}/$SERVICE_NAME-image-v1/pom.xml \
+        versions:set  \
+        -DnewVersion=${bamboo_version_tag}
+    ${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn \
+        -f ${bamboo_build_working_directory}/$SERVICE_NAME-v1/pom.xml \
+        --no-transfer-progress \
+        clean deploy \
+        -e \
+        -DdeployAtEnd=true \
+        -Dmaven.wagon.http.ssl.insecure=true \
+        -Dmaven.wagon.http.ssl.allowall=true \
+        -Dmaven.wagon.http.ssl.ignore.validity.dates=true \
+        -DaltDeploymentRepository=iqser_release::default::https://nexus.iqser.com/repository/red-platform-releases
+fi
+
+${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn \
+    --no-transfer-progress \
+    -f ${bamboo_build_working_directory}/$SERVICE_NAME-image-v1/pom.xml \
+    package
+
+${bamboo_capability_system_builder_mvn3_Maven_3}/bin/mvn \
+    --no-transfer-progress \
+    -f ${bamboo_build_working_directory}/$SERVICE_NAME-image-v1/pom.xml \
+    docker:push
+
+if [[ "${bamboo_version_tag}" = "dev" ]]
+then 
+    echo "gitTag=${bamboo_planRepository_1_branch}_${bamboo_buildNumber}" > git.tag
+else
+    echo "gitTag=${bamboo_version_tag}" > git.tag
+fi

redaction-service-v1/pom.xml

@@ -32,7 +32,7 @@
             <dependency>
                 <groupId>com.iqser.red</groupId>
                 <artifactId>platform-commons-dependency</artifactId>
-                <version>1.2.9</version>
+                <version>1.3.1</version>
                 <scope>import</scope>
                 <type>pom</type>
             </dependency>

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

@@ -17,5 +17,16 @@
             <artifactId>spring-web</artifactId>
             <optional>true</optional>
         </dependency>
+        <dependency>
+            <groupId>com.iqser.red.service</groupId>
+            <artifactId>configuration-service-api-v1</artifactId>
+            <version>2.11.0</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

@@ -13,9 +13,10 @@ 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;
 

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

@@ -11,6 +11,9 @@ import lombok.NoArgsConstructor;
 @AllArgsConstructor
 public class AnalyzeResult {
 
+    private String dossierId;
+    private String fileId;
+    private long duration;
     private int numberOfPages;
     private boolean hasHints;
     private boolean hasRequests;
@@ -18,7 +21,10 @@ public class AnalyzeResult {
     private boolean hasImages;
     private boolean hasUpdates;
     private long dictionaryVersion;
+    private long dossierDictionaryVersion;
     private long rulesVersion;
+    private long legalBasisVersion;
+
 
 }
 

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/ChangeType.java

@@ -0,0 +1,5 @@
+package com.iqser.red.service.redaction.v1.model;
+
+public enum ChangeType {
+    ADDED, REMOVED
+}

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

@@ -27,4 +27,6 @@ public class ManualRedactionEntry {
     private String section;
     private int sectionNumber;
 
+    private boolean addToDossierDictionary;
+
 }

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

@@ -0,0 +1,22 @@
+package com.iqser.red.service.redaction.v1.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+import java.util.ArrayList;
+import java.util.List;
+
+@Data
+@AllArgsConstructor
+@NoArgsConstructor
+public class RedactionChangeLog {
+
+    private List<RedactionChangeLogEntry> redactionLogEntry = new ArrayList<>();
+
+    private long dictionaryVersion = -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/RedactionChangeLogEntry.java

@@ -0,0 +1,47 @@
+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

@@ -1,32 +1,21 @@
 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 lombok.NoArgsConstructor;
 
 import java.util.List;
 
 @Data
 @AllArgsConstructor
-@NoArgsConstructor
 public class RedactionLog {
 
     private List<RedactionLogEntry> redactionLogEntry;
+    private List<LegalBasisMapping> legalBasis;
 
     private long dictionaryVersion = -1;
+    private long dossierDictionaryVersion = -1;
     private long rulesVersion = -1;
-
-    private String ruleSetId;
-    private String filename;
-
-
-    public RedactionLog(List<RedactionLogEntry> redactionLogEntry, long dictionaryVersion, long rulesVersion, String ruleSetId) {
-
-        this.redactionLogEntry = redactionLogEntry;
-        this.dictionaryVersion = dictionaryVersion;
-        this.rulesVersion = rulesVersion;
-        this.ruleSetId = ruleSetId;
-    }
-
+    private long legalBasisVersion = -1;
 
 }

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

@@ -3,6 +3,7 @@ package com.iqser.red.service.redaction.v1.model;
 import lombok.AllArgsConstructor;
 import lombok.Builder;
 import lombok.Data;
+import lombok.EqualsAndHashCode;
 import lombok.NoArgsConstructor;
 
 import java.util.ArrayList;
@@ -12,6 +13,7 @@ import java.util.List;
 @Builder
 @NoArgsConstructor
 @AllArgsConstructor
+@EqualsAndHashCode
 public class RedactionLogEntry {
 
     private String id;
@@ -45,4 +47,6 @@ public class RedactionLogEntry {
 
     private boolean isImage;
 
+    private boolean isDossierDictionaryEntry;
+
 }

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

@@ -11,8 +11,8 @@ 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;
 }

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

@@ -10,14 +10,9 @@ public interface RedactionResource {
 
     String SERVICE_NAME = "redaction-service-v1";
 
-    String RULE_SET_PARAMETER_NAME = "ruleSetId";
+    String RULE_SET_PARAMETER_NAME = "dossierTemplateId";
     String RULE_SET_PATH_VARIABLE = "/{" + RULE_SET_PARAMETER_NAME + "}";
 
-    @PostMapping(value = "/analyze", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
-    AnalyzeResult analyze(@RequestBody AnalyzeRequest analyzeRequest);
-
-    @PostMapping(value = "/reanalyze", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
-    AnalyzeResult reanalyze(@RequestBody AnalyzeRequest renalyzeRequest);
 
     @PostMapping(value = "/annotate", produces = MediaType.APPLICATION_JSON_VALUE, consumes = MediaType.APPLICATION_JSON_VALUE)
     AnnotateResponse annotate(@RequestBody AnnotateRequest annotateRequest);
@@ -32,7 +27,7 @@ public interface RedactionResource {
     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);
+    void updateRules(@PathVariable(RULE_SET_PARAMETER_NAME) String dossierTemplateId);
 
     @PostMapping(value = "/rules/test", consumes = MediaType.APPLICATION_JSON_VALUE)
     void testRules(@RequestBody String rules);

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

@@ -21,15 +21,20 @@
             <artifactId>redaction-service-api-v1</artifactId>
             <version>${project.version}</version>
         </dependency>
-        <dependency>
-            <groupId>com.iqser.red.service</groupId>
-            <artifactId>configuration-service-api-v1</artifactId>
-            <version>2.5.0</version>
-        </dependency>
         <dependency>
             <groupId>com.iqser.red.service</groupId>
             <artifactId>file-management-service-api-v1</artifactId>
-            <version>2.6.7</version>
+            <version>2.25.0</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>
@@ -51,6 +56,23 @@
             <artifactId>guava</artifactId>
             <version>29.0-jre</version>
         </dependency>
+
+        <dependency>
+            <groupId>org.apache.pdfbox</groupId>
+            <artifactId>jbig2-imageio</artifactId>
+            <version>3.0.3</version>
+        </dependency>
+        <dependency>
+            <groupId>com.github.jai-imageio</groupId>
+            <artifactId>jai-imageio-core</artifactId>
+            <version>1.4.0</version>
+        </dependency>
+        <dependency>
+            <groupId>com.github.jai-imageio</groupId>
+            <artifactId>jai-imageio-jpeg2000</artifactId>
+            <version>1.4.0</version>
+        </dependency>
+
         <!-- commons -->
         <dependency>
             <groupId>com.iqser.red.commons</groupId>
@@ -83,6 +105,12 @@
             <artifactId>spring-cloud-starter-openfeign</artifactId>
         </dependency>
 
+        <dependency>
+            <groupId>org.springframework.boot</groupId>
+            <artifactId>spring-boot-starter-amqp</artifactId>
+            <version>2.3.1.RELEASE</version>
+        </dependency>
+
         <!-- test dependencies -->
         <dependency>
             <groupId>org.springframework.boot</groupId>
@@ -94,6 +122,12 @@
             <artifactId>test-commons</artifactId>
             <scope>test</scope>
         </dependency>
+        <dependency>
+            <groupId>org.springframework.amqp</groupId>
+            <artifactId>spring-rabbit-test</artifactId>
+            <version>2.3.1</version>
+            <scope>test</scope>
+        </dependency>
     </dependencies>
 
     <build>

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

@@ -18,6 +18,7 @@ import org.springframework.context.annotation.Import;
 public class Application {
 
     public static void main(String[] args) {
+        System.setProperty("org.apache.pdfbox.rendering.UsePureJavaCMYKConversion", "true");
         SpringApplication.run(Application.class, args);
     }
 

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

@@ -2,6 +2,7 @@ package com.iqser.red.service.redaction.v1.server.classification.model;
 
 import com.iqser.red.service.redaction.v1.model.RedactionLogEntry;
 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;
@@ -31,7 +32,7 @@ public class Document {
 
     private List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
     private SectionGrid sectionGrid = new SectionGrid();
-    private long dictionaryVersion;
+    private DictionaryVersion dictionaryVersion;
     private long rulesVersion;
 
     private List<SectionText> sectionText = new ArrayList<>();

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

@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.classification.model;
 
+import com.fasterxml.jackson.annotation.JsonIgnore;
 import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
 import lombok.AllArgsConstructor;
 import lombok.Data;
@@ -12,7 +13,7 @@ public class Footer {
 
     private List<TextBlock> textBlocks;
 
-
+    @JsonIgnore
     public SearchableText getSearchableText() {
 
         SearchableText searchableText = new SearchableText();

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

@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.classification.model;
 
+import com.fasterxml.jackson.annotation.JsonIgnore;
 import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
 import lombok.AllArgsConstructor;
 import lombok.Data;
@@ -12,7 +13,7 @@ public class Header {
 
     private List<TextBlock> textBlocks;
 
-
+    @JsonIgnore
     public SearchableText getSearchableText() {
 
         SearchableText searchableText = new SearchableText();

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/Page.java

@@ -30,6 +30,8 @@ public class Page {
     private StringFrequencyCounter fontCounter = new StringFrequencyCounter();
     private StringFrequencyCounter fontStyleCounter = new StringFrequencyCounter();
 
+    private double cropBoxArea;
+
 
     public boolean isRotated() {
 

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

@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.classification.model;
 
+import com.fasterxml.jackson.annotation.JsonIgnore;
 import com.iqser.red.service.redaction.v1.model.SectionArea;
 import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
 import com.iqser.red.service.redaction.v1.server.redaction.model.Image;
@@ -31,6 +32,12 @@ public class SectionText {
     private List<Integer> cellStarts = new ArrayList<>();
 
 
+    public void setTabularData(Map<String, CellValue> tabularData) {
+        tabularData.remove(null);
+        this.tabularData = tabularData;
+    }
+
+    @JsonIgnore
     public SearchableText getSearchableText() {
 
         SearchableText searchableText = new SearchableText();

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

@@ -1,11 +1,13 @@
 package com.iqser.red.service.redaction.v1.server.classification.model;
 
+import com.fasterxml.jackson.annotation.JsonIgnore;
 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.AbstractTextContainer;
 import lombok.AllArgsConstructor;
 import lombok.Builder;
 import lombok.Data;
+import lombok.NoArgsConstructor;
 
 import java.util.ArrayList;
 import java.util.List;
@@ -13,6 +15,7 @@ import java.util.List;
 @AllArgsConstructor
 @Builder
 @Data
+@NoArgsConstructor
 public class TextBlock extends AbstractTextContainer {
 
     @Builder.Default
@@ -29,6 +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;
@@ -116,6 +120,7 @@ public class TextBlock extends AbstractTextContainer {
     }
 
     @Override
+    @JsonIgnore
     public String getText() {
 
         StringBuilder sb = new StringBuilder();

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

@@ -1,5 +1,6 @@
 package com.iqser.red.service.redaction.v1.server.classification.model;
 
+import com.fasterxml.jackson.annotation.JsonIgnore;
 import com.iqser.red.service.redaction.v1.server.redaction.model.SearchableText;
 import lombok.AllArgsConstructor;
 import lombok.Data;
@@ -12,7 +13,7 @@ public class UnclassifiedText {
 
     private List<TextBlock> textBlocks;
 
-
+    @JsonIgnore
     public SearchableText getSearchableText() {
 
         SearchableText searchableText = new SearchableText();

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
-                    .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 || splitByX || newLineAfterSplit || splittedByRuling)) {
+
+                Orientation prevOrientation = null;
+                if(!chunkBlockList1.isEmpty()) {
+                    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/FileStatusProcessingUpdateClient.java

@@ -0,0 +1,9 @@
+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}")
+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

@@ -0,0 +1,8 @@
+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}")
+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

@@ -3,6 +3,6 @@ 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.SERVICE_NAME, url = "${configuration-service.url}")
+@FeignClient(name = "RulesResource", url = "${configuration-service.url}")
 public interface RulesClient extends RulesResource {
-}
+}

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

@@ -1,7 +1,10 @@
 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.redaction.v1.model.*;
+import com.iqser.red.service.redaction.v1.model.AnnotateRequest;
+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;
@@ -9,7 +12,6 @@ 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.ReanalyzeService;
 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;
@@ -17,6 +19,7 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
 import com.iqser.red.service.redaction.v1.server.visualization.service.PdfVisualisationService;
 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;
@@ -34,32 +37,20 @@ public class RedactionController implements RedactionResource {
     private final DroolsExecutionService droolsExecutionService;
     private final DictionaryService dictionaryService;
     private final AnnotationService annotationService;
-    private final ReanalyzeService reanalyzeService;
     private final PdfSegmentationService pdfSegmentationService;
     private final RedactionStorageService redactionStorageService;
 
-    @Override
-    public AnalyzeResult analyze(@RequestBody AnalyzeRequest analyzeRequest) {
-        return reanalyzeService.analyze(analyzeRequest);
-
-    }
-
-    @Override
-    public AnalyzeResult reanalyze(@RequestBody AnalyzeRequest analyzeRequest) {
-        return reanalyzeService.reanalyze(analyzeRequest);
-    }
-
 
     public AnnotateResponse annotate(@RequestBody AnnotateRequest annotateRequest) {
 
-        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(annotateRequest.getProjectId(), annotateRequest.getFileId(), FileType.ORIGIN));
-        var redactionLog = redactionStorageService.getRedactionLog(annotateRequest.getProjectId(), annotateRequest.getFileId());
-        var sectionsGrid = redactionStorageService.getSectionGrid(annotateRequest.getProjectId(), annotateRequest.getFileId());
-
-        try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
+        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(annotateRequest.getDossierId(), annotateRequest.getFileId(), FileType.ORIGIN));
+        var redactionLog = redactionStorageService.getRedactionLog(annotateRequest.getDossierId(), 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());
+
+            dictionaryService.updateDictionary(annotateRequest.getDossierTemplateId(), annotateRequest.getDossierId());
             annotationService.annotate(pdDocument, redactionLog, sectionsGrid);
 
             try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream()) {
@@ -75,51 +66,62 @@ 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);
 
-        try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
-            pdDocument.setAllSecurityToBeRemoved(true);
+            storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getDossierId(), redactionRequest.getFileId(), FileType.ORIGIN));
+            try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
+                pdDocument.setAllSecurityToBeRemoved(true);
 
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            pdfVisualisationService.visualizeClassifications(classifiedDoc, pdDocument);
+                pdfVisualisationService.visualizeClassifications(classifiedDoc, pdDocument);
 
-            return convert(pdDocument, classifiedDoc.getPages().size());
+                return convert(pdDocument, classifiedDoc.getPages().size());
+
+            } catch (IOException e) {
+                throw new RedactionException(e);
+            }
 
         } catch (IOException e) {
             throw new RedactionException(e);
         }
 
+
     }
 
-
     @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);
 
-        try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
-            pdDocument.setAllSecurityToBeRemoved(true);
+            storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getDossierId(), redactionRequest.getFileId(), FileType.ORIGIN));
+            try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
+                pdDocument.setAllSecurityToBeRemoved(true);
 
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            pdfVisualisationService.visualizeParagraphs(classifiedDoc, pdDocument);
+                pdfVisualisationService.visualizeParagraphs(classifiedDoc, pdDocument);
+                return convert(pdDocument, classifiedDoc.getPages().size());
 
-            return convert(pdDocument, classifiedDoc.getPages().size());
+            } catch (IOException e) {
+                throw new RedactionException(e);
+            }
 
         } catch (IOException e) {
             throw new RedactionException(e);
         }
 
+
     }
 
 
     @Override
     public RedactionResult htmlTables(@RequestBody RedactionRequest redactionRequest) {
 
-        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(redactionRequest.getProjectId(), redactionRequest.getFileId(), FileType.ORIGIN));
-
         Document classifiedDoc;
-        try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
-            pdDocument.setAllSecurityToBeRemoved(true);
-            classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+
+        try {
+            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);
         }
@@ -141,9 +143,9 @@ public class RedactionController implements RedactionResource {
 
 
     @Override
-    public void updateRules(@PathVariable(RULE_SET_PARAMETER_NAME) String ruleSetId) {
+    public void updateRules(@PathVariable(RULE_SET_PARAMETER_NAME) String dossierTemplateId) {
 
-        droolsExecutionService.updateRules(ruleSetId);
+        droolsExecutionService.updateRules(dossierTemplateId);
     }
 
 
@@ -166,4 +168,5 @@ public class RedactionController implements RedactionResource {
 
     }
 
+
 }

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

@@ -0,0 +1,52 @@
+package com.iqser.red.service.redaction.v1.server.memory;
+
+import lombok.extern.slf4j.Slf4j;
+
+import java.text.CharacterIterator;
+import java.text.StringCharacterIterator;
+
+@Slf4j
+public class MemoryStats {
+
+
+    public static void printMemoryStats() {
+        log.info("\n\n ------------------------------ \n" +
+                " Used Memory: " + humanReadableByteCountBin(getUsedMemory()) + "\n" +
+                " Free Memory: " + humanReadableByteCountBin(getFreeMemory()) + "\n" +
+                " Total Memory: " + humanReadableByteCountBin(getTotalMemory()) + "\n" +
+                " Max Memory: " + humanReadableByteCountBin(getMaxMemory()) + "\n" +
+                "\n ------------------------------ \n");
+    }
+
+
+    public static String humanReadableByteCountBin(long bytes) {
+        long absB = bytes == Long.MIN_VALUE ? Long.MAX_VALUE : Math.abs(bytes);
+        if (absB < 1024) {
+            return bytes + " B";
+        }
+        long value = absB;
+        CharacterIterator ci = new StringCharacterIterator("KMGTPE");
+        for (int i = 40; i >= 0 && absB > 0xfffccccccccccccL >> i; i -= 10) {
+            value >>= 10;
+            ci.next();
+        }
+        value *= Long.signum(bytes);
+        return String.format("%.1f %ciB", value / 1024.0, ci.current());
+    }
+
+    private static long getMaxMemory() {
+        return Runtime.getRuntime().maxMemory();
+    }
+
+    private static long getUsedMemory() {
+        return getMaxMemory() - getFreeMemory();
+    }
+
+    private static long getTotalMemory() {
+        return Runtime.getRuntime().totalMemory();
+    }
+
+    private static long getFreeMemory() {
+        return Runtime.getRuntime().freeMemory();
+    }
+}

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

@@ -6,6 +6,7 @@ import com.iqser.red.service.redaction.v1.server.tableextraction.model.Ruling;
 import lombok.Getter;
 import lombok.Setter;
 import lombok.extern.slf4j.Slf4j;
+import org.apache.commons.lang3.reflect.FieldUtils;
 import org.apache.pdfbox.contentstream.operator.Operator;
 import org.apache.pdfbox.contentstream.operator.OperatorName;
 import org.apache.pdfbox.contentstream.operator.color.*;
@@ -195,6 +196,9 @@ public class PDFLinesTextStripper extends PDFTextStripper {
                 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);
+
                 if (rect.getHeight() > 2 && rect.getWidth() > 2) {
                     this.images.add(new PdfImage(pdfImage.getImage(), rect, pageNumber));
                 }
@@ -296,6 +300,18 @@ 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/parsing/model/RedTextPosition.java

@@ -0,0 +1,52 @@
+package com.iqser.red.service.redaction.v1.server.parsing.model;
+
+import com.fasterxml.jackson.annotation.JsonIgnore;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+import lombok.SneakyThrows;
+import org.apache.pdfbox.text.TextPosition;
+import org.springframework.beans.BeanUtils;
+
+@Data
+@NoArgsConstructor
+public class RedTextPosition {
+
+    private String textMatrix;
+    private int rotation;
+    private float y;
+    private float pageHeight;
+    private float pageWidth;
+    private String unicode;
+    private float XDirAdj;
+    private float YDirAdj;
+    private float width;
+    private float heightDir;
+
+    // not used in reanalysis
+    @JsonIgnore
+    private float widthOfSpace;
+
+    // not used in reanalysis
+    @JsonIgnore
+    private float fontSizeInPt;
+
+    // not used in reanalysis
+    @JsonIgnore
+    private String fontName;
+
+
+    @SneakyThrows
+    public static RedTextPosition fromTextPosition(TextPosition textPosition) {
+        var pos = new RedTextPosition();
+        BeanUtils.copyProperties(textPosition, pos);
+        pos.setFontName(textPosition.getFont().getName());
+
+        pos.setFontSizeInPt(textPosition.getFontSizeInPt());
+
+        pos.setTextMatrix(textPosition.getTextMatrix().toString());
+
+        return pos;
+    }
+
+
+}

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

@@ -1,29 +1,52 @@
 package com.iqser.red.service.redaction.v1.server.parsing.model;
 
+import com.fasterxml.jackson.annotation.JsonIgnore;
+import com.fasterxml.jackson.annotation.JsonIgnoreProperties;
 import com.iqser.red.service.redaction.v1.model.Point;
 import com.iqser.red.service.redaction.v1.model.Rectangle;
 import lombok.Data;
-import lombok.RequiredArgsConstructor;
+import lombok.NoArgsConstructor;
 import org.apache.pdfbox.text.TextPosition;
 
 import java.util.ArrayList;
 import java.util.List;
+import java.util.stream.Collectors;
 
 @Data
-@RequiredArgsConstructor
+@NoArgsConstructor
+@JsonIgnoreProperties({ "empty" })
 public class TextPositionSequence implements CharSequence {
 
-    private final int page;
-    private List<TextPosition> textPositions = new ArrayList<>();
+    private int page;
+    private List<RedTextPosition> textPositions = new ArrayList<>();
+
+    private float x1;
+    private float x2;
+
+    public TextPositionSequence(int page) {
+        this.page = page;
+    }
+
+
+    public static TextPositionSequence fromData(List<RedTextPosition> textPositions, int page) {
+        var textPositionSequence = new TextPositionSequence();
+        textPositionSequence.textPositions = textPositions;
+        textPositionSequence.page = page;
+
+        return textPositionSequence;
+    }
 
 
     public TextPositionSequence(List<TextPosition> textPositions, int page) {
 
-        this.textPositions = textPositions;
+        this.textPositions = textPositions.stream().map(RedTextPosition::fromTextPosition).collect(Collectors.toList());
         this.page = page;
     }
 
 
+
+
+
     @Override
     public int length() {
 
@@ -34,7 +57,7 @@ public class TextPositionSequence implements CharSequence {
     @Override
     public char charAt(int index) {
 
-        TextPosition textPosition = textPositionAt(index);
+        RedTextPosition textPosition = textPositionAt(index);
         String text = textPosition.getUnicode();
         return text.charAt(0);
     }
@@ -42,7 +65,7 @@ public class TextPositionSequence implements CharSequence {
 
     public char charAt(int index, boolean caseInSensitive) {
 
-        TextPosition textPosition = textPositionAt(index);
+        RedTextPosition textPosition = textPositionAt(index);
         String text = textPosition.getUnicode();
         return caseInSensitive ? text.toLowerCase().charAt(0) : text.charAt(0);
     }
@@ -51,7 +74,7 @@ public class TextPositionSequence implements CharSequence {
     @Override
     public TextPositionSequence subSequence(int start, int end) {
 
-        return new TextPositionSequence(textPositions.subList(start, end), page);
+        return fromData(textPositions.subList(start, end), page);
     }
 
 
@@ -66,18 +89,25 @@ public class TextPositionSequence implements CharSequence {
     }
 
 
-    public TextPosition textPositionAt(int index) {
+    public RedTextPosition textPositionAt(int index) {
 
         return textPositions.get(index);
     }
 
 
-    public void add(TextPosition textPosition) {
+    public void add(RedTextPosition textPosition) {
 
         this.textPositions.add(textPosition);
     }
 
 
+    public void add(TextPosition textPosition) {
+
+        this.textPositions.add(RedTextPosition.fromTextPosition(textPosition));
+    }
+
+
+    @JsonIgnore
     public float getX1() {
 
         if (textPositions.get(0).getRotation() == 90) {
@@ -88,6 +118,7 @@ public class TextPositionSequence implements CharSequence {
     }
 
 
+    @JsonIgnore
     public float getX2() {
 
         if (textPositions.get(0).getRotation() == 90) {
@@ -98,13 +129,14 @@ public class TextPositionSequence implements CharSequence {
         }
     }
 
-
+    @JsonIgnore
     public float getRotationAdjustedY() {
 
         return textPositions.get(0).getY();
     }
 
 
+    @JsonIgnore
     public float getY1() {
 
         if (textPositions.get(0).getRotation() == 90) {
@@ -115,6 +147,7 @@ public class TextPositionSequence implements CharSequence {
     }
 
 
+    @JsonIgnore
     public float getY2() {
 
         if (textPositions.get(0).getRotation() == 90) {
@@ -125,38 +158,40 @@ public class TextPositionSequence implements CharSequence {
     }
 
 
+    @JsonIgnore
     public float getTextHeight() {
 
         return textPositions.get(0).getHeightDir() + 2;
     }
 
 
+    @JsonIgnore
     public float getHeight() {
 
         return getY2() - getY1();
     }
 
 
+    @JsonIgnore
     public float getWidth() {
 
         return getX2() - getX1();
     }
 
 
+    @JsonIgnore
     public String getFont() {
-
-        return textPositions.get(0)
-                .getFont()
-                .toString()
+        return textPositions.get(0).getFontName()
                 .toLowerCase()
                 .replaceAll(",bold", "")
                 .replaceAll(",italic", "");
     }
 
 
+    @JsonIgnore
     public String getFontStyle() {
 
-        String lowercaseFontName = textPositions.get(0).getFont().toString().toLowerCase();
+        String lowercaseFontName = textPositions.get(0).getFontName().toLowerCase();
 
         if (lowercaseFontName.contains("bold") && lowercaseFontName.contains("italic")) {
             return "bold, italic";
@@ -170,25 +205,25 @@ public class TextPositionSequence implements CharSequence {
 
     }
 
-
+    @JsonIgnore
     public float getFontSize() {
 
         return textPositions.get(0).getFontSizeInPt();
     }
 
-
+    @JsonIgnore
     public float getSpaceWidth() {
 
         return textPositions.get(0).getWidthOfSpace();
     }
 
-
+    @JsonIgnore
     public int getRotation() {
 
         return textPositions.get(0).getRotation();
     }
 
-
+    @JsonIgnore
     public Rectangle getRectangle() {
 
         float height = getTextHeight();

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

@@ -0,0 +1,35 @@
+package com.iqser.red.service.redaction.v1.server.queue;
+
+import lombok.RequiredArgsConstructor;
+import org.springframework.amqp.core.Queue;
+import org.springframework.amqp.core.QueueBuilder;
+import org.springframework.context.annotation.Bean;
+import org.springframework.context.annotation.Configuration;
+
+@Configuration
+@RequiredArgsConstructor
+public class MessagingConfiguration {
+
+
+    public static final String REDACTION_QUEUE = "redactionQueue";
+
+    public static final String REDACTION_DQL = "redactionDQL";
+
+
+    @Bean
+    public Queue redactionQueue() {
+
+        return QueueBuilder.durable(REDACTION_QUEUE)
+                .withArgument("x-dead-letter-exchange", "")
+                .withArgument("x-dead-letter-routing-key", REDACTION_DQL)
+                .maxPriority(2)
+                .build();
+    }
+
+
+    @Bean
+    public Queue redactionDeadLetterQueue() {
+
+        return QueueBuilder.durable(REDACTION_DQL).build();
+    }
+}

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

@@ -0,0 +1,54 @@
+package com.iqser.red.service.redaction.v1.server.queue;
+
+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.server.client.FileStatusProcessingUpdateClient;
+import com.iqser.red.service.redaction.v1.server.redaction.service.ReanalyzeService;
+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;
+
+import static com.iqser.red.service.redaction.v1.server.queue.MessagingConfiguration.REDACTION_DQL;
+import static com.iqser.red.service.redaction.v1.server.queue.MessagingConfiguration.REDACTION_QUEUE;
+
+@Slf4j
+@Service
+@RequiredArgsConstructor
+public class RedactionMessageReceiver {
+
+    private final ObjectMapper objectMapper;
+    private final ReanalyzeService reanalyzeService;
+    private final FileStatusProcessingUpdateClient fileStatusProcessingUpdateClient;
+
+    @RabbitHandler
+    @RabbitListener(queues = REDACTION_QUEUE)
+    public void receiveAnalyzeRequest(String in) throws JsonProcessingException {
+
+        var analyzeRequest = objectMapper.readValue(in, AnalyzeRequest.class);
+        log.info("Processing analyze request: {}", analyzeRequest);
+        AnalyzeResult result;
+        if (analyzeRequest.isReanalyseOnlyIfPossible()) {
+            result = reanalyzeService.reanalyze(analyzeRequest);
+        } else {
+            result = reanalyzeService.analyze(analyzeRequest);
+        }
+        log.info("Successfully analyzed {}", analyzeRequest);
+
+        fileStatusProcessingUpdateClient.analysisSuccessful(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), result);
+    }
+
+    @RabbitHandler
+    @RabbitListener(queues = REDACTION_DQL)
+    public void receiveAnalyzeRequestDQL(String in) throws JsonProcessingException {
+
+        var analyzeRequest = objectMapper.readValue(in, AnalyzeRequest.class);
+        log.info("Failed to process analyze request: {}", analyzeRequest);
+
+        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/CellValue.java

@@ -3,19 +3,23 @@ package com.iqser.red.service.redaction.v1.server.redaction.model;
 import com.iqser.red.service.redaction.v1.server.classification.model.TextBlock;
 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
 import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
-import lombok.Value;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.NoArgsConstructor;
 
+import java.util.ArrayList;
 import java.util.Iterator;
 import java.util.List;
 
-@Value
+@Data
+@NoArgsConstructor
+@AllArgsConstructor
 public class CellValue {
 
-    private List<TextBlock> textBlocks;
+    private List<TextBlock> textBlocks = new ArrayList<>();
 
     private int rowSpanStart;
 
-
     @Override
     public String toString() {
 

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

@@ -18,13 +18,13 @@ public class Dictionary {
     private Map<String, DictionaryModel> localAccessMap = new HashMap<>();
 
     @Getter
-    private long version;
+    private DictionaryVersion version;
 
 
-    public Dictionary(List<DictionaryModel> dictionaryModels, long dictionaryVersion) {
+    public Dictionary(List<DictionaryModel> dictionaryModels, DictionaryVersion version) {
         this.dictionaryModels = dictionaryModels;
         this.dictionaryModels.forEach(dm -> localAccessMap.put(dm.getType(), dm));
-        this.version = dictionaryVersion;
+        this.version = version;
     }
 
 

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

@@ -10,6 +10,6 @@ import java.util.Set;
 public class DictionaryIncrement {
 
     private Set<DictionaryIncrementValue> values;
-    private long dictionaryVersion;
+    private DictionaryVersion dictionaryVersion;
 
 }

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

@@ -21,6 +21,7 @@ public class DictionaryModel implements Serializable {
     private boolean recommendation;
     private Set<DictionaryEntry> entries;
     private Set<String> localEntries;
+    private boolean isDossierDictionary;
 
     public Set<String> getValues(boolean local) {
         return local ? localEntries : entries.stream().filter(e -> !e.isDeleted()).map(e -> e.getValue()).collect(Collectors

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

@@ -0,0 +1,16 @@
+package com.iqser.red.service.redaction.v1.server.redaction.model;
+
+import lombok.AllArgsConstructor;
+import lombok.Builder;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@Builder
+@NoArgsConstructor
+@AllArgsConstructor
+public class DictionaryVersion {
+
+    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

@@ -37,8 +37,10 @@ public class Entity {
     private String textBefore;
     private String textAfter;
 
+    private boolean isDossierDictionaryEntry;
 
-    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) {
+
+    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) {
 
         this.word = word;
         this.type = type;
@@ -54,10 +56,11 @@ public class Entity {
         this.textAfter = textAfter;
         this.start = start;
         this.end = end;
+        this.isDossierDictionaryEntry = isDossierDictionaryEntry;
     }
 
 
-    public Entity(String word, String type, Integer start, Integer end, String headline, int sectionNumber, boolean isDictionaryEntry) {
+    public Entity(String word, String type, Integer start, Integer end, String headline, int sectionNumber, boolean isDictionaryEntry, boolean isDossierDictionaryEntry) {
 
         this.word = word;
         this.type = type;
@@ -66,6 +69,7 @@ public class Entity {
         this.headline = headline;
         this.sectionNumber = sectionNumber;
         this.isDictionaryEntry = isDictionaryEntry;
+        this.isDossierDictionaryEntry = isDossierDictionaryEntry;
     }
 
 }

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

@@ -5,8 +5,6 @@ import lombok.Builder;
 import lombok.Data;
 import lombok.NoArgsConstructor;
 
-import java.awt.geom.Rectangle2D;
-
 @Data
 @Builder
 @NoArgsConstructor
@@ -14,7 +12,7 @@ import java.awt.geom.Rectangle2D;
 public class Image {
 
     private String type;
-    private Rectangle2D position;
+    private RedRectangle2D position;
     private boolean redaction;
     private String redactionReason;
     private String legalBasis;

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

@@ -1,8 +1,7 @@
 package com.iqser.red.service.redaction.v1.server.redaction.model;
 
-import lombok.AllArgsConstructor;
+import com.fasterxml.jackson.annotation.JsonIgnore;
 import lombok.Data;
-import lombok.NoArgsConstructor;
 import lombok.NonNull;
 import lombok.RequiredArgsConstructor;
 
@@ -10,19 +9,23 @@ import java.awt.geom.Rectangle2D;
 import java.awt.image.BufferedImage;
 
 @Data
-@NoArgsConstructor
-@AllArgsConstructor
 @RequiredArgsConstructor
 public class PdfImage {
 
-    @NonNull
+    @JsonIgnore
     private BufferedImage image;
     @NonNull
-    private Rectangle2D position;
+    private RedRectangle2D position;
     private ImageType imageType;
     private boolean isAppendedToParagraph;
 
     @NonNull
     private int page;
 
+    public PdfImage(BufferedImage image, Rectangle2D position, int page) {
+        this.image = image;
+        this.position = new RedRectangle2D(position.getX(), position.getY(), position.getWidth(), position.getHeight());
+        this.page = page;
+    }
+
 }

redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/model/RedRectangle2D.java

@@ -0,0 +1,42 @@
+package com.iqser.red.service.redaction.v1.server.redaction.model;
+
+
+import com.fasterxml.jackson.annotation.JsonIgnore;
+import lombok.AllArgsConstructor;
+import lombok.Data;
+import lombok.NoArgsConstructor;
+
+@Data
+@NoArgsConstructor
+@AllArgsConstructor
+public class RedRectangle2D {
+
+    public static final double THRESHOLD = 0.01;
+
+    private double x;
+    private double y;
+    private double width;
+    private double height;
+
+    @JsonIgnore
+    public boolean isEmpty() {
+        return width <= 0.0f || height <= 0.0f;
+    }
+
+    public boolean contains(double x, double y, double w, double h) {
+        if (isEmpty() || w <= 0 || h <= 0) {
+            return false;
+        }
+        double x0 = getX();
+        double y0 = getY();
+        return round(x) >= round(x0) &&
+                round(y) >= round(y0) &&
+                (x + w) - (x0 + getWidth()) <= THRESHOLD &&
+                (y + h) - (y0 + getHeight()) <= THRESHOLD;
+    }
+
+    private double round(double value) {
+        double d = Math.pow(10, 2);
+        return Math.round(value * d) / d;
+    }
+}

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

@@ -411,7 +411,7 @@ public class Section {
         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);
+        Set<Entity> found = EntitySearchUtils.find(text, Set.of(searchValue), asType, headline, sectionNumber, true, false);
 
         found.forEach(entity -> {
             if (redacted) {
@@ -437,7 +437,7 @@ public class Section {
         } else {
             String word = value.toString();
 
-            Entity entity = new Entity(word, type, value.getRowSpanStart(), value.getRowSpanStart() + word.length(), headline, sectionNumber, false);
+            Entity entity = new Entity(word, type, value.getRowSpanStart(), value.getRowSpanStart() + word.length(), headline, sectionNumber, false, false);
             entity.setRedaction(redact);
             entity.setMatchedRule(ruleNumber);
             entity.setRedactionReason(reason);

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

@@ -1,7 +1,7 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;
 
-import com.iqser.red.service.file.management.v1.api.model.RedactionChangeLog;
 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;
@@ -9,7 +9,7 @@ import org.springframework.stereotype.Service;
 @Service
 public class AnalyzeResponseService {
 
-    public AnalyzeResult createAnalyzeResponse(int pageCount, RedactionLog redactionLog, RedactionChangeLog redactionChangeLog) {
+    public AnalyzeResult createAnalyzeResponse(String dossierId, String fileId, long duration, int pageCount, RedactionLog redactionLog, RedactionChangeLog redactionChangeLog) {
         boolean hasHints = redactionLog.getRedactionLogEntry().stream().anyMatch(RedactionLogEntry::isHint);
 
         boolean hasRequests = redactionLog.getRedactionLogEntry()
@@ -31,6 +31,9 @@ public class AnalyzeResponseService {
                 .isEmpty() && redactionChangeLog.getRedactionLogEntry().stream().anyMatch(entry -> !entry.getType().equals("false_positive"));
 
         return AnalyzeResult.builder()
+                .dossierId(dossierId)
+                .fileId(fileId)
+                .duration(duration)
                 .numberOfPages(pageCount)
                 .hasHints(hasHints)
                 .hasRedactions(hasRedactions)
@@ -39,6 +42,8 @@ public class AnalyzeResponseService {
                 .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/DictionaryService.java

@@ -6,10 +6,7 @@ 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.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.*;
 import feign.FeignException;
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
@@ -21,6 +18,8 @@ import java.awt.Color;
 import java.util.*;
 import java.util.stream.Collectors;
 
+import static com.iqser.red.service.configuration.v1.api.resource.DictionaryResource.GLOBAL_DOSSIER;
+
 @Slf4j
 @Service
 @RequiredArgsConstructor
@@ -28,69 +27,89 @@ 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 long updateDictionary(String ruleSetId) {
+    public DictionaryVersion updateDictionary(String dossierTemplateId, String dossierId) {
 
-        long version = dictionaryClient.getVersion(ruleSetId);
-
-        var foundDictionary = dictionariesByRuleSets.get(ruleSetId);
-
-        if (foundDictionary == null || version > foundDictionary.getDictionaryVersion()) {
-            updateDictionaryEntry(ruleSetId, version);
+        long dossierTemplateDictionaryVersion = dictionaryClient.getVersion(dossierTemplateId, GLOBAL_DOSSIER);
+        var dossierTemplateDictionary = dictionariesByDossierTemplate.get(dossierTemplateId);
+        if (dossierTemplateDictionary == null || dossierTemplateDictionaryVersion > dossierTemplateDictionary.getDictionaryVersion()) {
+            updateDictionaryEntry(dossierTemplateId, dossierTemplateDictionaryVersion, GLOBAL_DOSSIER);
         }
 
-        return version;
+        long dossierDictionaryVersion = dictionaryClient.getVersion(dossierTemplateId, dossierId);
+        var dossierDictionary = dictionariesByDossier.get(dossierId);
+        if (dossierDictionary == null || dossierDictionaryVersion > dossierDictionary.getDictionaryVersion()) {
+            updateDictionaryEntry(dossierTemplateId, dossierDictionaryVersion, dossierId);
+        }
+
+        return DictionaryVersion.builder().dossierTemplateVersion(dossierTemplateDictionaryVersion).dossierVersion(dossierDictionaryVersion).build();
     }
 
 
-    public DictionaryIncrement getDictionaryIncrements(String ruleSetId, long fromVersion) {
+    public DictionaryIncrement getDictionaryIncrements(String dossierTemplateId, DictionaryVersion fromVersion, String dossierId) {
 
-        long version = updateDictionary(ruleSetId);
+        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) {
+                if (dictionaryEntry.getVersion() > fromVersion.getDossierTemplateVersion()) {
                     newValues.add(new DictionaryIncrementValue(dictionaryEntry.getValue(), dictionaryModel.isCaseInsensitive()));
                 }
             });
         });
 
+        if(dictionariesByDossier.containsKey(dossierId)) {
+            dictionaryModels = dictionariesByDossier.get(dossierId).getDictionary();
+            dictionaryModels.forEach(dictionaryModel -> {
+                dictionaryModel.getEntries().forEach(dictionaryEntry -> {
+                    if (dictionaryEntry.getVersion() > fromVersion.getDossierVersion()) {
+                        newValues.add(new DictionaryIncrementValue(dictionaryEntry.getValue(), dictionaryModel.isCaseInsensitive()));
+                    }
+                });
+            });
+        }
+
         return new DictionaryIncrement(newValues, version);
     }
 
 
-    private void updateDictionaryEntry(String ruleSetId, long version) {
+    private void updateDictionaryEntry(String dossierTemplateId, long version, String dossierId) {
 
         try {
             DictionaryRepresentation dictionaryRepresentation = new DictionaryRepresentation();
 
-            TypeResponse typeResponse = dictionaryClient.getAllTypes(ruleSetId);
+            TypeResponse typeResponse = dictionaryClient.getAllTypes(dossierTemplateId, dossierId);
             if (typeResponse != null && CollectionUtils.isNotEmpty(typeResponse.getTypes())) {
 
                 List<DictionaryModel> dictionary = typeResponse.getTypes()
                         .stream()
                         .map(t -> new DictionaryModel(t.getType(), t.getRank(), convertColor(t.getHexColor()), t.isCaseInsensitive(), t
-                                .isHint(), t.isRecommendation(), convertEntries(t), new HashSet<>()))
+                                .isHint(), t.isRecommendation(), convertEntries(t, dossierId), new HashSet<>(), !dossierId.equals(GLOBAL_DOSSIER)))
                         .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);
 
-                dictionariesByRuleSets.put(ruleSetId, dictionaryRepresentation);
+                if(dossierId.equals(GLOBAL_DOSSIER)) {
+                    dictionariesByDossierTemplate.put(dossierTemplateId, dictionaryRepresentation);
+                } else {
+                    dictionariesByDossier.put(dossierId, dictionaryRepresentation);
+                }
             }
         } catch (FeignException e) {
             log.warn("Got some unknown feignException", e);
@@ -99,23 +118,23 @@ 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);
-                long externalVersion = dictionaryClient.getVersion(ruleSetId);
-                if (externalVersion == dictionary.getVersion() + 1) {
-                    dictionary.setVersion(externalVersion);
+                dictionaryClient.addEntries(dm.getType(), dossierTemplateId, new ArrayList<>(dm.getLocalEntries()), false, GLOBAL_DOSSIER);
+                long externalVersion = dictionaryClient.getVersion(dossierTemplateId, GLOBAL_DOSSIER);
+                if (externalVersion == dictionary.getVersion().getDossierTemplateVersion() + 1) {
+                    dictionary.getVersion().setDossierTemplateVersion(externalVersion);
                 }
             }
         });
     }
 
 
-    private Set<DictionaryEntry> convertEntries(TypeResult t) {
+    private Set<DictionaryEntry> convertEntries(TypeResult t, String dossierId) {
 
-        Set<DictionaryEntry> entries = new HashSet<>(dictionaryClient.getDictionaryForType(t.getType(), t.getRuleSetId())
+        Set<DictionaryEntry> entries = new HashSet<>(dictionaryClient.getDictionaryForType(t.getType(), t.getDossierTemplateId(), dossierId)
                 .getEntries());
 
         if (t.isCaseInsensitive()) {
@@ -132,9 +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();
         }
@@ -142,19 +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();
         }
@@ -162,9 +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();
         }
@@ -172,35 +191,44 @@ public class DictionaryService {
     }
 
 
-    public Dictionary getDeepCopyDictionary(String ruleSetId) {
+    public Dictionary getDeepCopyDictionary(String dossierTemplateId, String dossierId) {
 
         List<DictionaryModel> copy = new ArrayList<>();
 
-        var representation = dictionariesByRuleSets.get(ruleSetId);
-        var dictionary = dictionariesByRuleSets.get(ruleSetId).getDictionary();
-        dictionary.forEach(dm -> {
+        var dossierTemplateRepresentation = dictionariesByDossierTemplate.get(dossierTemplateId);
+        dossierTemplateRepresentation.getDictionary().forEach(dm -> {
             copy.add(SerializationUtils.clone(dm));
         });
 
-        return new Dictionary(copy, representation.getDictionaryVersion());
+        //TODO merge dictionaries if they have same names
+        long dossierDictionaryVersion = -1;
+        if(dictionariesByDossier.containsKey(dossierId)) {
+            var dossierRepresentation = dictionariesByDossier.get(dossierId);
+            dossierRepresentation.getDictionary().forEach(dm -> {
+                copy.add(SerializationUtils.clone(dm));
+            });
+            dossierDictionaryVersion = dossierRepresentation.getDictionaryVersion();
+        }
+
+        return new Dictionary(copy, 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).getRequestAddColor();
     }
 
 
-    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

@@ -28,14 +28,14 @@ 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 +55,43 @@ public class DroolsExecutionService {
     }
 
 
-    public KieContainer updateRules(String ruleSetId) {
+    public KieContainer updateRules(String dossierTemplateId) {
 
-        long version = rulesClient.getVersion(ruleSetId);
-        Long rulesVersion = rulesVersionPerRuleSetId.get(ruleSetId);
+        long version = rulesClient.getVersion(dossierTemplateId);
+        Long rulesVersion = rulesVersionPerDossierTemplateId.get(dossierTemplateId);
         if (rulesVersion == null) {
             rulesVersion = -1L;
         }
 
         if (version > rulesVersion.longValue()) {
-            rulesVersionPerRuleSetId.put(ruleSetId, version);
-            return createOrUpdateKieContainer(ruleSetId);
+            rulesVersionPerDossierTemplateId.put(dossierTemplateId, version);
+            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);
+            RulesResponse rules = rulesClient.getRules(dossierTemplateId);
             if (rules == null || StringUtils.isEmpty(rules.getRules())) {
                 throw new RuntimeException("Rules cannot be empty.");
             }
 
             KieServices kieServices = KieServices.Factory.get();
-            KieModule kieModule = getKieModule(ruleSetId, rules.getRules(), kieServices);
+            KieModule kieModule = getKieModule(dossierTemplateId, rules.getRules(), 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 +100,11 @@ 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();
@@ -122,9 +122,9 @@ 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;
         }

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

@@ -33,13 +33,13 @@ public class EntityRedactionService {
     private final SurroundingWordsService surroundingWordsService;
 
 
-    public void processDocument(Document classifiedDoc, String ruleSetId, ManualRedactions manualRedactions) {
+    public void processDocument(Document classifiedDoc, String dossierTemplateId, ManualRedactions manualRedactions, String dossierId) {
 
-        dictionaryService.updateDictionary(ruleSetId);
-        KieContainer container = droolsExecutionService.updateRules(ruleSetId);
-        long rulesVersion = droolsExecutionService.getRulesVersion(ruleSetId);
+        dictionaryService.updateDictionary(dossierTemplateId, dossierId);
+        KieContainer container = droolsExecutionService.updateRules(dossierTemplateId);
+        long rulesVersion = droolsExecutionService.getRulesVersion(dossierTemplateId);
 
-        Dictionary dictionary = dictionaryService.getDeepCopyDictionary(ruleSetId);
+        Dictionary dictionary = dictionaryService.getDeepCopyDictionary(dossierTemplateId, dossierId);
 
         Set<Entity> documentEntities = new HashSet<>(findEntities(classifiedDoc, container, manualRedactions, dictionary, false, null));
 
@@ -71,11 +71,11 @@ public class EntityRedactionService {
                         .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()));
+                                .getStart(), entity.getEnd(), entity.isDossierDictionaryEntry()));
             }
         }
 
-        dictionaryService.updateExternalDictionary(dictionary, ruleSetId);
+        dictionaryService.updateExternalDictionary(dictionary, dossierTemplateId);
 
         classifiedDoc.setDictionaryVersion(dictionary.getVersion());
         classifiedDoc.setRulesVersion(rulesVersion);
@@ -187,6 +187,7 @@ public class EntityRedactionService {
                         .get(0)
                         .getPage());
                 sectionText.getSectionAreas().add(sectionArea);
+                sectionText.getTextBlocks().addAll(cell.getTextBlocks());
 
                 addSectionToManualRedactions(cell.getTextBlocks(), manualRedactions, table.getHeadline(), sectionNumber.intValue());
                 int cellStart = start;
@@ -235,6 +236,8 @@ public class EntityRedactionService {
                 sectionText.setHeadline(table.getHeadline());
                 sectionText.setSectionNumber(sectionNumber.intValue());
                 sectionText.setTable(true);
+                sectionText.setTabularData(tabularData);
+                sectionText.setCellStarts(cellStarts);
                 classifiedDoc.getSectionText().add(sectionText);
             }
 
@@ -267,6 +270,7 @@ public class EntityRedactionService {
                             .getSequences()
                             .get(0)
                             .getPage());
+                    sectionText.getTextBlocks().addAll(cell.getTextBlocks());
                     sectionText.getSectionAreas().add(sectionArea);
                 }
 
@@ -325,6 +329,10 @@ public class EntityRedactionService {
             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);
         }
 
@@ -363,9 +371,9 @@ 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));
+                found.addAll(EntitySearchUtils.find(lowercaseInputString, model.getValues(local), model.getType(), headline, sectionNumber, local, model.isDossierDictionary()));
             } else {
-                found.addAll(EntitySearchUtils.find(searchableString, model.getValues(local), model.getType(), headline, sectionNumber, local));
+                found.addAll(EntitySearchUtils.find(searchableString, model.getValues(local), model.getType(), headline, sectionNumber, local, model.isDossierDictionary()));
             }
         }
 

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

@@ -1,10 +1,11 @@
 package com.iqser.red.service.redaction.v1.server.redaction.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.client.ImageClassificationClient;
 import com.iqser.red.service.redaction.v1.server.client.ImageClassificationResponse;
 import com.iqser.red.service.redaction.v1.server.client.MockMultipartFile;
 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.settings.RedactionServiceSettings;
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
@@ -12,7 +13,6 @@ import org.springframework.stereotype.Service;
 
 import javax.imageio.ImageIO;
 import java.io.ByteArrayOutputStream;
-import java.io.IOException;
 
 @Slf4j
 @Service
@@ -23,37 +23,49 @@ public class ImageClassificationService {
     private final RedactionServiceSettings settings;
 
 
-    public void classifyImages(Document classifiedDoc) {
+    public void classifyImages(Page page) {
 
-        long start = System.currentTimeMillis();
-        classifiedDoc.getPages().forEach(page -> {
-            page.getImages().forEach(image -> {
+        page.getImages().forEach(image -> {
 
-                if (settings.isEnableImageClassification()) {
-                    try (ByteArrayOutputStream baos = new ByteArrayOutputStream()) {
-                        ImageIO.write(image.getImage(), "png", baos);
-                        ImageClassificationResponse response = imageClassificationClient.classify(new MockMultipartFile("file", "Image.png", "image/png", baos
-                                .toByteArray()));
-                        image.setImageType(ImageType.valueOf(response.getCategory()));
+            if (settings.isEnableImageClassification() && !isEntirePageImage(image, page)) {
 
-                    } catch (IOException e) {
-                        log.error("Could not classify image", e);
-                    }
-                } else {
+                long start = System.currentTimeMillis();
+                try (ByteArrayOutputStream baos = new ByteArrayOutputStream()) {
+                    ImageIO.write(image.getImage(), "png", baos);
+                    var mockFile = new MockMultipartFile("file", "Image.png", "image/png", baos.toByteArray());
+                    ImageClassificationResponse response = imageClassificationClient.classify(mockFile);
+                    image.setImageType(ImageType.valueOf(response.getCategory()));
+                } catch (Exception e) {
+                    log.error("Could not classify image", e);
                     image.setImageType(ImageType.OTHER);
                 }
+                log.info("Image classification took: " + (System.currentTimeMillis() - start));
+            } else {
+                image.setImageType(ImageType.OTHER);
+            }
 
-                if (image.getImageType().equals(ImageType.OTHER)) {
-                    page.getTextBlocks().forEach(textblock -> {
-                        if (image.getPosition()
-                                .contains(textblock.getMinX(), textblock.getMinY(), textblock.getWidth(), textblock.getHeight())) {
-                            image.setImageType(ImageType.OCR);
-                        }
-                    });
-                }
-            });
+            image.getImage().flush();
+            image.setImage(null);
+
+            if (image.getImageType().equals(ImageType.OTHER)) {
+                page.getTextBlocks().forEach(textblock -> {
+                    if (image.getPosition()
+                            .contains(textblock.getMinX(), textblock.getMinY(), textblock.getWidth(), textblock.getHeight())) {
+                        image.setImageType(ImageType.OCR);
+                    }
+                });
+            }
         });
-        log.info("Image classification took: " + (System.currentTimeMillis() - start));
+
+    }
+
+    private boolean isEntirePageImage(PdfImage image, Page page){
+        double imageArea = image.getPosition().getHeight() * image.getPosition().getWidth();
+        if(imageArea / page.getCropBoxArea() >= settings.getMaxImageCropboxRatio()){
+            log.info("Skipping image classification because images is almost as large as the entire page");
+            return true;
+        }
+        return false;
     }
 
 }

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

@@ -5,20 +5,22 @@ 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.apache.pdfbox.pdmodel.PDDocument;
+
 import org.kie.api.runtime.KieContainer;
 import org.springframework.stereotype.Service;
 import org.springframework.web.bind.annotation.RequestBody;
 
-import java.awt.geom.Rectangle2D;
 import java.util.*;
 import java.util.stream.Collectors;
 import java.util.stream.Stream;
@@ -35,70 +37,82 @@ public class ReanalyzeService {
     private final RedactionLogCreatorService redactionLogCreatorService;
     private final RedactionStorageService redactionStorageService;
     private final PdfSegmentationService pdfSegmentationService;
-    private final ImageClassificationService imageClassificationService;
     private final RedactionChangeLogService redactionChangeLogService;
     private final AnalyzeResponseService analyzeResponseService;
+    private final LegalBasisClient legalBasisClient;
 
     public AnalyzeResult analyze(AnalyzeRequest analyzeRequest) {
-        var storedObjectStream = redactionStorageService.getStoredObject(RedactionStorageService.StorageIdUtils.getStorageId(analyzeRequest.getProjectId(), analyzeRequest.getFileId(), FileType.ORIGIN));
+
+        long startTime = System.currentTimeMillis();
 
         var pageCount = 0;
         Document classifiedDoc;
-        try (PDDocument pdDocument = PDDocument.load(storedObjectStream)) {
-            pdDocument.setAllSecurityToBeRemoved(true);
-            pageCount = pdDocument.getNumberOfPages();
-            classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
+
+        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);
         }
         log.info("Document structure analysis successful, starting redaction analysis...");
 
-        imageClassificationService.classifyImages(classifiedDoc);
-        entityRedactionService.processDocument(classifiedDoc, analyzeRequest.getRuleSetId(), analyzeRequest.getManualRedactions());
-        imageClassificationService.classifyImages(classifiedDoc);
+        entityRedactionService.processDocument(classifiedDoc, analyzeRequest.getDossierTemplateId(), analyzeRequest.getManualRedactions(), analyzeRequest
+                .getDossierId());
         redactionLogCreatorService.createRedactionLog(classifiedDoc, pageCount, analyzeRequest.getManualRedactions(), analyzeRequest
-                .getRuleSetId());
+                .getDossierTemplateId());
 
         log.info("Redaction analysis successful...");
 
+        var legalBasis = legalBasisClient.getLegalBasisMapping(analyzeRequest.getDossierTemplateId());
+        var redactionLog = new RedactionLog(classifiedDoc.getRedactionLogEntities(),legalBasis,
+                classifiedDoc.getDictionaryVersion().getDossierTemplateVersion(),
+                classifiedDoc.getDictionaryVersion().getDossierVersion(),
+                classifiedDoc.getRulesVersion(),
+                legalBasisClient.getVersion(analyzeRequest.getDossierTemplateId()));
 
-        var redactionLog = new RedactionLog(classifiedDoc.getRedactionLogEntities(), classifiedDoc.getDictionaryVersion(), classifiedDoc
-                .getRulesVersion(), analyzeRequest.getRuleSetId());
+        log.info("Analyzed with rules {} and dictionary {} for dossierTemplate: {}", classifiedDoc.getRulesVersion(), classifiedDoc
+                .getDictionaryVersion(), analyzeRequest.getDossierTemplateId());
 
         // 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);
+        var changeLog = redactionChangeLogService.createAndStoreChangeLog(analyzeRequest.getDossierId(), 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());
+        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.REDACTION_LOG, redactionLog);
+        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.TEXT, new Text(pageCount, classifiedDoc
+                .getSectionText()));
+        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.SECTION_GRID, classifiedDoc
+                .getSectionGrid());
 
-        return analyzeResponseService.createAnalyzeResponse(pageCount, redactionLog, changeLog);
+        long duration = System.currentTimeMillis() - startTime;
+        return analyzeResponseService.createAnalyzeResponse(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), duration, pageCount, redactionLog, changeLog);
     }
 
-    public AnalyzeResult reanalyze(@RequestBody AnalyzeRequest renalyzeRequest) {
-        var text = redactionStorageService.getText(renalyzeRequest.getProjectId(), renalyzeRequest.getFileId());
-        // new procedure was not applied, we need a complete analysis
-        if (text.getNumberOfPages() == 0) {
-            return analyze(AnalyzeRequest.builder()
-                    .ruleSetId(renalyzeRequest.getRuleSetId())
-                    .manualRedactions(renalyzeRequest.getManualRedactions())
-                    .projectId(renalyzeRequest.getProjectId())
-                    .fileId(renalyzeRequest.getFileId())
-                    .build());
+
+    @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);
         }
-        var redactionLog = redactionStorageService.getRedactionLog(renalyzeRequest.getProjectId(), renalyzeRequest.getFileId());
 
+        DictionaryIncrement dictionaryIncrement = dictionaryService.getDictionaryIncrements(analyzeRequest.getDossierTemplateId(), new DictionaryVersion(redactionLog
+                .getDictionaryVersion(), redactionLog.getDossierDictionaryVersion()), analyzeRequest.getDossierId());
 
-        DictionaryIncrement dictionaryIncrement = dictionaryService.getDictionaryIncrements(renalyzeRequest.getRuleSetId(), redactionLog.getDictionaryVersion());
-
-        Set<String> manualForceAndRemoveIds = getForceAndRemoveIds(renalyzeRequest.getManualRedactions());
+        Set<String> manualForceAndRemoveIds = getForceAndRemoveIds(analyzeRequest.getManualRedactions());
         Map<String, List<Comment>> comments = null;
         Set<ManualRedactionEntry> manualAdds = null;
 
-        if (renalyzeRequest.getManualRedactions() != null) {
+        if (analyzeRequest.getManualRedactions() != null) {
             // TODO comments will be removed from redactionLog, so we ignore this first.
-            comments = renalyzeRequest.getManualRedactions().getComments();
-            manualAdds = renalyzeRequest.getManualRedactions().getEntriesToAdd();
+            comments = analyzeRequest.getManualRedactions().getComments();
+            manualAdds = analyzeRequest.getManualRedactions().getEntriesToAdd();
         }
 
         Set<Integer> sectionsToReanalyse = new HashSet<>();
@@ -132,115 +146,122 @@ public class ReanalyzeService {
             }
         }
 
+        log.info("Should reanalyze {} sections for request: {}", sectionsToReanalyse.size(), analyzeRequest);
+
         if (sectionsToReanalyse.isEmpty() && (manualAdds == null || manualAdds.isEmpty())) {
-            redactionLog.setDictionaryVersion(dictionaryIncrement.getDictionaryVersion());
-            var changeLog = redactionChangeLogService.createAndStoreChangeLog(renalyzeRequest.getProjectId(), renalyzeRequest.getFileId(), redactionLog);
-            redactionStorageService.storeObject(renalyzeRequest.getProjectId(), renalyzeRequest.getFileId(), FileType.REDACTION_LOG, redactionLog);
-            return analyzeResponseService.createAnalyzeResponse(text.getNumberOfPages(), redactionLog, changeLog);
+            return finalizeAnalysis(analyzeRequest, startTime, redactionLog, text, dictionaryIncrement);
         }
 
-        try {
+        List<SectionText> reanalysisSections = new ArrayList<>();
 
-            List<SectionText> reanalysisSections = new ArrayList<>();
-            for (SectionText sectionText : text.getSectionTexts()) {
-
-                if (sectionsToReanalyse.contains(sectionText.getSectionNumber())) {
-                    reanalysisSections.add(sectionText);
-                }
+        for (SectionText sectionText : text.getSectionTexts()) {
 
+            if (sectionsToReanalyse.contains(sectionText.getSectionNumber())) {
+                reanalysisSections.add(sectionText);
             }
-
-            KieContainer kieContainer = droolsExecutionService.updateRules(renalyzeRequest.getRuleSetId());
-
-            Dictionary dictionary = dictionaryService.getDeepCopyDictionary(renalyzeRequest.getRuleSetId());
-
-            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()));
-                }
-            }
-
-            List<RedactionLogEntry> newRedactionLogEntries = new ArrayList<>();
-            for (int page = 1; page <= text.getNumberOfPages(); page++) {
-                if (entitiesPerPage.get(page) != null) {
-                    newRedactionLogEntries.addAll(redactionLogCreatorService.addEntries(entitiesPerPage, renalyzeRequest
-                            .getManualRedactions(), page, renalyzeRequest.getRuleSetId()));
-                }
-
-                if (imagesPerPage.get(page) != null) {
-                    newRedactionLogEntries.addAll(redactionLogCreatorService.addImageEntries(imagesPerPage, renalyzeRequest
-                            .getManualRedactions(), page, renalyzeRequest.getRuleSetId()));
-                }
-
-                newRedactionLogEntries.addAll(redactionLogCreatorService.addManualAddEntries(manualAdds, comments, page, renalyzeRequest
-                        .getRuleSetId()));
-            }
-
-            redactionLog.getRedactionLogEntry().removeIf(entry -> sectionsToReanalyse.contains(entry.getSectionNumber()) && !entry.isImage() || entry.getSectionNumber() == 0 && !entry.isImage());
-            redactionLog.getRedactionLogEntry().addAll(newRedactionLogEntries);
-            redactionLog.setDictionaryVersion(dictionaryIncrement.getDictionaryVersion());
-
-            var changeLog = redactionChangeLogService.createAndStoreChangeLog(renalyzeRequest.getProjectId(), renalyzeRequest.getFileId(), redactionLog);
-            redactionStorageService.storeObject(renalyzeRequest.getProjectId(), renalyzeRequest.getFileId(), FileType.REDACTION_LOG, redactionLog);
-            return analyzeResponseService.createAnalyzeResponse(text.getNumberOfPages(), redactionLog, changeLog);
-
-
-        } catch (Exception e) {
-            throw new RedactionException(e);
         }
 
+        //--
+
+        KieContainer kieContainer = droolsExecutionService.updateRules(analyzeRequest.getDossierTemplateId());
+
+        Dictionary dictionary = dictionaryService.getDeepCopyDictionary(analyzeRequest.getDossierTemplateId(), analyzeRequest.getDossierId());
+
+        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
+                        .getDossierTemplateId()));
+            }
+
+            if (imagesPerPage.get(page) != null) {
+                newRedactionLogEntries.addAll(redactionLogCreatorService.addImageEntries(imagesPerPage, analyzeRequest.getManualRedactions(), page, analyzeRequest
+                        .getDossierTemplateId()));
+            }
+
+            newRedactionLogEntries.addAll(redactionLogCreatorService.addManualAddEntries(manualAdds, comments, page, analyzeRequest
+                    .getDossierTemplateId()));
+        }
+
+        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().getDossierTemplateVersion());
+        redactionLog.setDossierDictionaryVersion(dictionaryIncrement.getDictionaryVersion().getDossierVersion());
+
+        var changeLog = redactionChangeLogService.createAndStoreChangeLog(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), redactionLog);
+        redactionStorageService.storeObject(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), FileType.REDACTION_LOG, redactionLog);
+
+        long duration = System.currentTimeMillis() - startTime;
+
+        return analyzeResponseService.createAnalyzeResponse(analyzeRequest.getDossierId(), analyzeRequest.getFileId(), duration, text
+                .getNumberOfPages(), redactionLog, changeLog);
     }
 
 
@@ -263,7 +284,7 @@ public class ReanalyzeService {
 
         return Image.builder()
                 .type(entry.getType())
-                .position(new Rectangle2D.Float(position.getTopLeft().getX(), position.getTopLeft()
+                .position(new RedRectangle2D(position.getTopLeft().getX(), position.getTopLeft()
                         .getY(), position.getWidth(), position.getHeight()))
                 .sectionNumber(entry.getSectionNumber())
                 .section(entry.getSection())

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

@@ -1,10 +1,9 @@
 package com.iqser.red.service.redaction.v1.server.redaction.service;
 
-import com.fasterxml.jackson.databind.ObjectMapper;
-import com.iqser.red.service.file.management.v1.api.model.ChangeType;
 import com.iqser.red.service.file.management.v1.api.model.FileType;
-import com.iqser.red.service.file.management.v1.api.model.RedactionChangeLog;
-import com.iqser.red.service.file.management.v1.api.model.RedactionChangeLogEntry;
+import com.iqser.red.service.redaction.v1.model.ChangeType;
+import com.iqser.red.service.redaction.v1.model.RedactionChangeLog;
+import com.iqser.red.service.redaction.v1.model.RedactionChangeLogEntry;
 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.server.storage.RedactionStorageService;
@@ -21,15 +20,14 @@ import java.util.stream.Collectors;
 @RequiredArgsConstructor
 public class RedactionChangeLogService {
 
-    private final RedactionStorageService storageStorageService;
-    private final ObjectMapper objectMapper;
+    private final RedactionStorageService redactionStorageService;
 
-    public RedactionChangeLog createAndStoreChangeLog(String projectId, String fileId, RedactionLog currentRedactionLog) {
+    public RedactionChangeLog createAndStoreChangeLog(String dossierId, String fileId, RedactionLog currentRedactionLog) {
 
         try {
-            RedactionLog previousRedactionLog = storageStorageService.getRedactionLog(projectId, fileId);
+            RedactionLog previousRedactionLog = redactionStorageService.getRedactionLog(dossierId, fileId);
             var changeLog = createChangeLog(currentRedactionLog, previousRedactionLog);
-            storageStorageService.storeObject(projectId, fileId, FileType.REDACTION_CHANGELOG, objectMapper.writeValueAsBytes(changeLog));
+            redactionStorageService.storeObject(dossierId, fileId, FileType.REDACTION_CHANGELOG, changeLog);
             return changeLog;
         } catch (Exception e) {
             log.debug("Previous redaction log not available");
@@ -59,8 +57,11 @@ public class RedactionChangeLogService {
                 .map(entry -> convert(entry, ChangeType.REMOVED))
                 .collect(Collectors.toList()));
 
-        return new RedactionChangeLog(changeLogEntries, currentRedactionLog.getDictionaryVersion(), currentRedactionLog.getRulesVersion(), currentRedactionLog
-                .getRuleSetId(), currentRedactionLog.getFilename());
+        return new RedactionChangeLog(changeLogEntries,
+                currentRedactionLog.getDictionaryVersion(),
+                currentRedactionLog.getDossierDictionaryVersion(),
+                currentRedactionLog.getRulesVersion(),
+                currentRedactionLog.getLegalBasisVersion());
     }
 
 
@@ -88,6 +89,7 @@ public class RedactionChangeLogService {
                 .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

@@ -4,6 +4,7 @@ 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;
@@ -14,7 +15,6 @@ 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.apache.pdfbox.text.TextPosition;
 import org.springframework.stereotype.Service;
 
 import java.util.ArrayList;
@@ -32,7 +32,7 @@ public class RedactionLogCreatorService {
 
 
     public void createRedactionLog(Document classifiedDoc, int numberOfPages, ManualRedactions manualRedactions,
-                                   String ruleSetId) {
+                                   String dossierTemplateId) {
 
         Set<Integer> manualRedactionPages = getManualRedactionPages(manualRedactions);
 
@@ -42,24 +42,24 @@ public class RedactionLogCreatorService {
 
             if (classifiedDoc.getEntities().get(page) != null) {
                 classifiedDoc.getRedactionLogEntities()
-                        .addAll(addEntries(classifiedDoc.getEntities(), manualRedactions, page, ruleSetId));
+                        .addAll(addEntries(classifiedDoc.getEntities(), manualRedactions, page, dossierTemplateId));
             }
 
             if (manualRedactionPages.contains(page)) {
                 classifiedDoc.getRedactionLogEntities()
-                        .addAll(addManualAddEntries(manualRedactions.getEntriesToAdd(), manualRedactions.getComments(), page, ruleSetId));
+                        .addAll(addManualAddEntries(manualRedactions.getEntriesToAdd(), manualRedactions.getComments(), page, dossierTemplateId));
             }
 
             if (classifiedDoc.getImages().get(page) != null && !classifiedDoc.getImages().get(page).isEmpty()) {
                 classifiedDoc.getRedactionLogEntities()
-                        .addAll(addImageEntries(classifiedDoc.getImages(), manualRedactions, page, ruleSetId));
+                        .addAll(addImageEntries(classifiedDoc.getImages(), manualRedactions, page, dossierTemplateId));
             }
         }
     }
 
 
     public List<RedactionLogEntry> addImageEntries(Map<Integer, Set<Image>> images, ManualRedactions manualRedactions,
-                                                   int pageNumber, String ruleSetId) {
+                                                   int pageNumber, String dossierTemplateId) {
 
         List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
 
@@ -69,14 +69,14 @@ public class RedactionLogCreatorService {
 
             RedactionLogEntry redactionLogEntry = RedactionLogEntry.builder()
                     .id(id)
-                    .color(getColorForImage(image, ruleSetId, false))
+                    .color(getColorForImage(image, dossierTemplateId, false))
                     .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)
@@ -96,11 +96,11 @@ public class RedactionLogCreatorService {
                             redactionLogEntry.setRedacted(false);
                             redactionLogEntry.setStatus(Status.APPROVED);
                             manualOverrideReason = image.getRedactionReason() + ", removed by manual override";
-                            redactionLogEntry.setColor(getColorForImage(image, ruleSetId, false));
+                            redactionLogEntry.setColor(getColorForImage(image, dossierTemplateId, false));
                         } else if (manualRemoval.getStatus().equals(Status.REQUESTED)) {
                             manualOverrideReason = image.getRedactionReason() + ", requested to remove";
                             redactionLogEntry.setStatus(Status.REQUESTED);
-                            redactionLogEntry.setColor(getColorForImage(image, ruleSetId, true));
+                            redactionLogEntry.setColor(getColorForImage(image, dossierTemplateId, true));
                         } else {
                             redactionLogEntry.setStatus(Status.DECLINED);
                         }
@@ -121,13 +121,13 @@ public class RedactionLogCreatorService {
                             image.setRedaction(true);
                             redactionLogEntry.setRedacted(true);
                             redactionLogEntry.setStatus(Status.APPROVED);
-                            redactionLogEntry.setColor(getColorForImage(image, ruleSetId, false));
+                            redactionLogEntry.setColor(getColorForImage(image, dossierTemplateId, 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.setColor(getColorForImage(image, dossierTemplateId, true));
                             redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
                         } else {
                             redactionLogEntry.setStatus(Status.DECLINED);
@@ -166,7 +166,7 @@ public class RedactionLogCreatorService {
 
 
     public List<RedactionLogEntry> addEntries(Map<Integer, List<Entity>> entities, ManualRedactions manualRedactions,
-                                              int page, String ruleSetId) {
+                                              int page, String dossierTemplateId) {
 
         List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
 
@@ -180,7 +180,7 @@ public class RedactionLogCreatorService {
 
             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.
@@ -199,11 +199,11 @@ public class RedactionLogCreatorService {
                                 redactionLogEntry.setRedacted(false);
                                 redactionLogEntry.setStatus(Status.APPROVED);
                                 manualOverrideReason = entity.getRedactionReason() + ", removed by manual override";
-                                redactionLogEntry.setColor(getColor(entity, ruleSetId, false));
+                                redactionLogEntry.setColor(getColor(entity, dossierTemplateId, false));
                             } else if (manualRemoval.getStatus().equals(Status.REQUESTED)) {
                                 manualOverrideReason = entity.getRedactionReason() + ", requested to remove";
                                 redactionLogEntry.setStatus(Status.REQUESTED);
-                                redactionLogEntry.setColor(getColor(entity, ruleSetId, true));
+                                redactionLogEntry.setColor(getColor(entity, dossierTemplateId, true));
                             } else {
                                 redactionLogEntry.setStatus(Status.DECLINED);
                             }
@@ -224,13 +224,13 @@ public class RedactionLogCreatorService {
                                 entity.setRedaction(true);
                                 redactionLogEntry.setRedacted(true);
                                 redactionLogEntry.setStatus(Status.APPROVED);
-                                redactionLogEntry.setColor(getColor(entity, ruleSetId, false));
+                                redactionLogEntry.setColor(getColor(entity, dossierTemplateId, 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.setColor(getColor(entity, dossierTemplateId, true));
                                 redactionLogEntry.setLegalBasis(manualForceRedact.getLegalBasis());
                             } else {
                                 redactionLogEntry.setStatus(Status.DECLINED);
@@ -272,24 +272,24 @@ public class RedactionLogCreatorService {
     }
 
 
-    private List<Rectangle> getRectanglesPerLine(List<TextPosition> textPositions, int page) {
+    private List<Rectangle> getRectanglesPerLine(List<RedTextPosition> textPositions, int page) {
 
         List<Rectangle> rectangles = new ArrayList<>();
         if (textPositions.size() == 1) {
-            rectangles.add(new TextPositionSequence(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(new TextPositionSequence(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(new TextPositionSequence(textPositions.subList(startIndex, textPositions.size()), page).getRectangle());
+                rectangles.add( TextPositionSequence.fromData(textPositions.subList(startIndex, textPositions.size()), page).getRectangle());
             }
         }
 
@@ -299,7 +299,7 @@ public class RedactionLogCreatorService {
 
     public List<RedactionLogEntry> addManualAddEntries(Set<ManualRedactionEntry> manualAdds,
                                                        Map<String, List<Comment>> comments, int page,
-                                                       String ruleSetId) {
+                                                       String dossierTemplateId) {
 
         List<RedactionLogEntry> redactionLogEntities = new ArrayList<>();
 
@@ -311,7 +311,7 @@ public class RedactionLogCreatorService {
 
             String id = manualRedactionEntry.getId();
 
-            RedactionLogEntry redactionLogEntry = createRedactionLogEntry(manualRedactionEntry, id, ruleSetId);
+            RedactionLogEntry redactionLogEntry = createRedactionLogEntry(manualRedactionEntry, id, dossierTemplateId);
 
             List<Rectangle> rectanglesOnPage = new ArrayList<>();
             for (Rectangle rectangle : manualRedactionEntry.getPositions()) {
@@ -338,11 +338,11 @@ public class RedactionLogCreatorService {
 
 
     private RedactionLogEntry createRedactionLogEntry(ManualRedactionEntry manualRedactionEntry, String id,
-                                                      String ruleSetId) {
+                                                      String dossierTemplateId) {
 
         return RedactionLogEntry.builder()
                 .id(id)
-                .color(getColorForManualAdd(manualRedactionEntry.getType(), ruleSetId, manualRedactionEntry.getStatus()))
+                .color(getColorForManualAdd(manualRedactionEntry.getType(), dossierTemplateId, manualRedactionEntry.getStatus()))
                 .reason(manualRedactionEntry.getReason())
                 .legalBasis(manualRedactionEntry.getLegalBasis())
                 .value(manualRedactionEntry.getValue())
@@ -355,21 +355,22 @@ public class RedactionLogCreatorService {
                 .status(manualRedactionEntry.getStatus())
                 .manualRedactionType(ManualRedactionType.ADD)
                 .isDictionaryEntry(false)
+                .isDossierDictionaryEntry(manualRedactionEntry.isAddToDossierDictionary())
                 .build();
     }
 
 
-    private RedactionLogEntry createRedactionLogEntry(Entity entity, String ruleSetId) {
+    private RedactionLogEntry createRedactionLogEntry(Entity entity, String dossierTemplateId) {
 
         return RedactionLogEntry.builder()
-                .color(getColor(entity, ruleSetId, false))
+                .color(getColor(entity, dossierTemplateId, false))
                 .reason(entity.getRedactionReason())
                 .legalBasis(entity.getLegalBasis())
                 .value(entity.getWord())
                 .type(entity.getType())
                 .redacted(entity.isRedaction())
-                .isHint(isHint(entity, ruleSetId))
-                .isRecommendation(isRecommendation(entity, ruleSetId))
+                .isHint(isHint(entity, dossierTemplateId))
+                .isRecommendation(isRecommendation(entity, dossierTemplateId))
                 .section(entity.getHeadline())
                 .sectionNumber(entity.getSectionNumber())
                 .matchedRule(entity.getMatchedRule())
@@ -378,60 +379,61 @@ public class RedactionLogCreatorService {
                 .textBefore(entity.getTextBefore())
                 .startOffset(entity.getStart())
                 .endOffset(entity.getEnd())
+                .isDossierDictionaryEntry(entity.isDossierDictionaryEntry())
                 .build();
     }
 
 
-    private float[] getColor(Entity entity, String ruleSetId, boolean requestedToRemove) {
+    private float[] getColor(Entity entity, String dossierTemplateId, boolean requestedToRemove) {
 
         if (requestedToRemove) {
-            return dictionaryService.getRequestRemoveColor(ruleSetId);
+            return dictionaryService.getRequestRemoveColor(dossierTemplateId);
         }
-        if (!entity.isRedaction() && !isHint(entity, ruleSetId)) {
-            return dictionaryService.getNotRedactedColor(ruleSetId);
+        if (!entity.isRedaction() && !isHint(entity, dossierTemplateId)) {
+            return dictionaryService.getNotRedactedColor(dossierTemplateId);
         }
-        return dictionaryService.getColor(entity.getType(), ruleSetId);
+        return dictionaryService.getColor(entity.getType(), dossierTemplateId);
     }
 
 
-    private float[] getColorForManualAdd(String type, String ruleSetId, Status status) {
+    private float[] getColorForManualAdd(String type, String dossierTemplateId, Status status) {
 
         if (status.equals(Status.REQUESTED)) {
-            return dictionaryService.getRequestAddColor(ruleSetId);
+            return dictionaryService.getRequestAddColor(dossierTemplateId);
         } else if (status.equals(Status.DECLINED)) {
-            return dictionaryService.getNotRedactedColor(ruleSetId);
+            return dictionaryService.getNotRedactedColor(dossierTemplateId);
         }
-        return getColor(type, ruleSetId);
+        return getColor(type, dossierTemplateId);
     }
 
 
-    private float[] getColor(String type, String ruleSetId) {
+    private float[] getColor(String type, String dossierTemplateId) {
 
-        return dictionaryService.getColor(type, ruleSetId);
+        return dictionaryService.getColor(type, dossierTemplateId);
     }
 
 
-    private float[] getColorForImage(Image image, String ruleSetId, boolean requestedToRemove) {
+    private float[] getColorForImage(Image image, String dossierTemplateId, boolean requestedToRemove) {
 
         if (requestedToRemove) {
-            return dictionaryService.getRequestRemoveColor(ruleSetId);
+            return dictionaryService.getRequestRemoveColor(dossierTemplateId);
         }
-        if (!image.isRedaction() && !dictionaryService.isHint(image.getType(), ruleSetId)) {
-            return dictionaryService.getNotRedactedColor(ruleSetId);
+        if (!image.isRedaction() && !dictionaryService.isHint(image.getType(), dossierTemplateId)) {
+            return dictionaryService.getNotRedactedColor(dossierTemplateId);
         }
-        return dictionaryService.getColor(image.getType(), ruleSetId);
+        return dictionaryService.getColor(image.getType(), dossierTemplateId);
     }
 
 
-    private boolean isHint(Entity entity, String ruleSetId) {
+    private boolean isHint(Entity entity, String dossierTemplateId) {
 
-        return dictionaryService.isHint(entity.getType(), ruleSetId);
+        return dictionaryService.isHint(entity.getType(), dossierTemplateId);
     }
 
 
-    private boolean isRecommendation(Entity entity, String ruleSetId) {
+    private boolean isRecommendation(Entity entity, String dossierTemplateId) {
 
-        return dictionaryService.isRecommendation(entity.getType(), ruleSetId);
+        return dictionaryService.isRecommendation(entity.getType(), dossierTemplateId);
     }
 
 

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

@@ -49,7 +49,7 @@ public class EntitySearchUtils {
 
 
     public Set<Entity> find(String inputString, Set<String> values, String type, String headline, int sectionNumber,
-                            boolean local) {
+                            boolean local, boolean isDossierDictionary) {
 
         Set<Entity> found = new HashSet<>();
 
@@ -69,7 +69,7 @@ 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));
+                    found.add(new Entity(inputString.substring(startIndex, stopIndex), type, startIndex, stopIndex, headline, sectionNumber, !local, isDossierDictionary));
                 }
             } while (startIndex > -1);
         }

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

@@ -3,9 +3,9 @@ package com.iqser.red.service.redaction.v1.server.redaction.utils;
 import com.google.common.hash.HashFunction;
 import com.google.common.hash.Hashing;
 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
+import com.iqser.red.service.redaction.v1.server.redaction.model.RedRectangle2D;
 import lombok.experimental.UtilityClass;
 
-import java.awt.geom.Rectangle2D;
 import java.nio.charset.StandardCharsets;
 import java.util.List;
 
@@ -25,12 +25,8 @@ public class IdBuilder {
     }
 
 
-    public String buildId(Rectangle2D rectangle2D, int page) {
-
-        StringBuilder sb = new StringBuilder();
-        sb.append("x").append(rectangle2D.getX()).append("y").append(rectangle2D.getY()).append("h").append(rectangle2D.getHeight()).append("w").append(rectangle2D.getWidth()).append("p").append(page);
-
-        return hashFunction.hashString(sb.toString(), StandardCharsets.UTF_8).toString();
+    public String buildId(RedRectangle2D rectangle2D, int page) {
+        return hashFunction.hashString("x" + rectangle2D.getX() + "y" + rectangle2D.getY() + "h" + rectangle2D.getHeight() + "w" + rectangle2D.getWidth() + "p" + page, StandardCharsets.UTF_8).toString();
     }
 
 

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

@@ -1,175 +1,291 @@
 package com.iqser.red.service.redaction.v1.server.segmentation;
 
-import com.iqser.red.service.redaction.v1.model.SectionArea;
+import com.iqser.red.service.redaction.v1.model.Rectangle;
 import com.iqser.red.service.redaction.v1.server.classification.model.Document;
 import com.iqser.red.service.redaction.v1.server.classification.model.Page;
-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.service.BlockificationService;
 import com.iqser.red.service.redaction.v1.server.classification.service.ClassificationService;
-import com.iqser.red.service.redaction.v1.server.exception.RedactionException;
-import com.iqser.red.service.redaction.v1.server.parsing.PDFAreaTextStripper;
+import com.iqser.red.service.redaction.v1.server.memory.MemoryStats;
 import com.iqser.red.service.redaction.v1.server.parsing.PDFLinesTextStripper;
 import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
-import com.iqser.red.service.redaction.v1.server.redaction.model.CellValue;
-import com.iqser.red.service.redaction.v1.server.storage.RedactionStorageService;
+import com.iqser.red.service.redaction.v1.server.redaction.model.PdfImage;
+import com.iqser.red.service.redaction.v1.server.redaction.service.ImageClassificationService;
 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.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.HashMap;
 import java.util.List;
-import java.util.Map;
 
 @Slf4j
 @Service
 @RequiredArgsConstructor
 public class PdfSegmentationService {
 
+    private final static int MAX_PAGES_BEFORE_GC = 250;
+
     private final RulingCleaningService rulingCleaningService;
     private final TableExtractionService tableExtractionService;
     private final BlockificationService blockificationService;
     private final ClassificationService classificationService;
     private final SectionsBuilderService sectionsBuilderService;
+    private final ImageClassificationService imageClassificationService;
 
 
-    private final RedactionStorageService redactionStorageService;
-
-
-    private void postProcessSections(PDDocument pdDocument, List<SectionText> texts) {
+    public Document parseDocument(InputStream documentInputStream) throws IOException {
+        return parseDocument(documentInputStream, false);
+    }
 
+    public Document parseDocument(InputStream documentInputStream, boolean ignoreImages) throws IOException {
+        PDDocument pdDocument = null;
         try {
-            for (SectionText sectionText : texts) {
+            //create tempFile
+            File tempFile = File.createTempFile("document", ".pdf");
+            IOUtils.copy(documentInputStream, new FileOutputStream(tempFile));
 
-                List<TextBlock> textBlocks = new ArrayList<>();
+            // initialize required variables
+            Document document = new Document();
+            List<Page> pages = new ArrayList<>();
 
-                Map<Integer, List<SectionArea>> sectionAreasPerPage = new HashMap<>();
-                for (SectionArea sectionArea : sectionText.getSectionAreas()) {
-                    sectionAreasPerPage.computeIfAbsent(sectionArea.getPage(), (x) -> new ArrayList<>())
-                            .add(sectionArea);
+
+            pdDocument = reinitializePDDocument(tempFile, null);
+            long pageCount = pdDocument.getNumberOfPages();
+
+            for (int pageNumber = 1; pageNumber <= pageCount; pageNumber++) {
+
+                if (pageNumber % MAX_PAGES_BEFORE_GC == 0) {
+                    pdDocument = reinitializePDDocument(tempFile, pdDocument);
                 }
 
-                Map<String, CellValue> tabularData = new HashMap<>();
-                List<Integer> cellStarts = new ArrayList<>();
-                for (Integer page : sectionAreasPerPage.keySet()) {
-                    List<SectionArea> areasOnPage = sectionAreasPerPage.get(page);
+                PDFLinesTextStripper stripper = new PDFLinesTextStripper();
+                PDPage pdPage = pdDocument.getPage(pageNumber - 1);
+                stripper.setPageNumber(pageNumber);
+                stripper.setStartPage(pageNumber);
+                stripper.setEndPage(pageNumber);
+                stripper.setPdpage(pdPage);
+                stripper.getText(pdDocument);
 
-                    PDPage pdPage = pdDocument.getPage(page - 1);
-                    PDRectangle cropBox = pdPage.getCropBox();
-                    PDFAreaTextStripper textStripper = new PDFAreaTextStripper();
-                    textStripper.setPageNumber(page);
+                PDRectangle pdr = pdPage.getMediaBox();
+                boolean isLandscape = pdr.getWidth() > pdr.getHeight();
 
-                    int cellStart = 0;
-                    for (SectionArea sectionArea : areasOnPage) {
+                int rotation = pdPage.getRotation();
+                boolean isRotated = rotation != 0 && rotation != 360;
 
-                        Rectangle2D rect = null;
-                        if (pdPage.getRotation() == 90) {
-                            rect = new Rectangle2D.Float(sectionArea.getTopLeft().getY(), sectionArea.getTopLeft()
-                                    .getX(), sectionArea.getHeight(), sectionArea.getWidth() + 0.001f);
-                        } else {
-                            rect = new Rectangle2D.Float(sectionArea.getTopLeft().getX(), -sectionArea.getTopLeft()
-                                    .getY() + cropBox.getUpperRightY() - sectionArea.getHeight(), sectionArea.getWidth(), sectionArea
-                                    .getHeight() + 0.001f);
+                CleanRulings cleanRulings = rulingCleaningService.getCleanRulings(stripper.getRulings(), stripper.getMinCharWidth(), stripper
+                        .getMaxCharHeight());
+
+                Page page = blockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings
+                        .getVertical());
+
+                PDRectangle cropbox = pdPage.getCropBox();
+                float cropboxArea = cropbox.getHeight() * cropbox.getWidth();
+                page.setCropBoxArea(cropboxArea);
+
+                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);
                         }
-
-                        textStripper.addRegion(String.valueOf(1), rect);
-                        textStripper.extractRegions(pdPage);
-                        textStripper.getTextForRegion(String.valueOf(1));
-                        List<TextPositionSequence> positions = textStripper.getTextPositionSequences();
-
-                        TextBlock textBlock = new TextBlock(sectionArea.getTopLeft().getX(), sectionArea.getTopLeft()
-                                .getX() + sectionArea.getWidth(), sectionArea.getTopLeft()
-                                .getY(), sectionArea.getTopLeft().getY() + sectionArea.getHeight(), positions, 0);
-
-                        if (sectionText.isTable()) {
-                            Cell cell = new Cell();
-                            cell.addTextBlock(textBlock);
-                            tabularData.put(sectionArea.getHeader(), new CellValue(cell.getTextBlocks(), cellStart));
-                            cellStarts.add(cellStart);
-                            cellStart = cellStart + cell.toString().trim().length() + 1;
-                        }
-
-                        textBlocks.add(textBlock);
-                        textStripper.clearPositions();
                     }
+                }
+                mergedList.removeAll(imagesInImage);
 
+                page.setImages(mergedList);
+
+                tableExtractionService.extractTables(cleanRulings, page);
+                buildPageStatistics(page);
+                increaseDocumentStatistics(page, document);
+
+
+                if (!ignoreImages) {
+                    imageClassificationService.classifyImages(page);
                 }
-                sectionText.setTextBlocks(textBlocks);
-                sectionText.setTabularData(tabularData);
-                if (sectionText.isTable()) {
-                    sectionText.setCellStarts(cellStarts);
-                }
+
+                pages.add(page);
+
+
             }
 
+            document.setPages(pages);
 
-        } catch (Exception e) {
-            throw new RedactionException(e);
+            classificationService.classifyDocument(document);
+            sectionsBuilderService.buildSections(document);
+            sectionsBuilderService.addImagesToSections(document);
+
+            pdDocument = reinitializePDDocument(tempFile, pdDocument);
+
+            IOUtils.close(pdDocument);
+
+            tempFile.delete();
+
+            return document;
+        } finally {
+            if (pdDocument != null) {
+                pdDocument.close();
+            }
         }
+    }
+
+    private PDDocument reinitializePDDocument(File tempFile, PDDocument pdDocument) throws IOException {
+        if (pdDocument != null) {
+            pdDocument.close();
+        }
+        System.runFinalization();
+        System.gc();
+
+        MemoryStats.printMemoryStats();
+
+        var newPDDocument = PDDocument.load(tempFile, MemoryUsageSetting.setupTempFileOnly());
+        newPDDocument.setAllSecurityToBeRemoved(true);
+
+        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;
+        }
+
 
     }
 
-
-    public Document parseDocument(PDDocument pdDocument) throws IOException {
-
-        Document document = new Document();
-
-        List<Page> pages = new ArrayList<>();
-        PDFLinesTextStripper stripper = new PDFLinesTextStripper();
-        for (int pageNumber = 1; pageNumber <= pdDocument.getNumberOfPages(); pageNumber++) {
-            PDPage pdPage = pdDocument.getPage(pageNumber - 1);
-            stripper.setPageNumber(pageNumber);
-            stripper.setStartPage(pageNumber);
-            stripper.setEndPage(pageNumber);
-            stripper.setPdpage(pdPage);
-            stripper.getText(pdDocument);
-
-            PDRectangle pdr = pdPage.getMediaBox();
-            boolean isLandscape = pdr.getWidth() > pdr.getHeight();
-
-            int rotation = pdPage.getRotation();
-            boolean isRotated = rotation != 0 && rotation != 360;
-
-            CleanRulings cleanRulings = rulingCleaningService.getCleanRulings(stripper.getRulings(), stripper.getMinCharWidth(), stripper
-                    .getMaxCharHeight());
-
-            Page page = blockificationService.blockify(stripper.getTextPositionSequences(), cleanRulings.getHorizontal(), cleanRulings
-                    .getVertical());
-            page.setRotation(rotation);
-
-            tableExtractionService.extractTables(cleanRulings, page);
-
-            buildPageStatistics(page);
-
-            page.setLandscape(isLandscape || isRotated);
-
-            page.setPageNumber(pageNumber);
-            increaseDocumentStatistics(page, document);
-
-            page.setImages(stripper.getImages());
-            pages.add(page);
+    //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;
+    }
 
-        document.setPages(pages);
-
-        classificationService.classifyDocument(document);
-        sectionsBuilderService.buildSections(document);
-        sectionsBuilderService.addImagesToSections(document);
-
-        // This can be improved an done in one pass, but it's complicated to do right away
-        postProcessSections(pdDocument, document.getSectionText());
-
-        return document;
+    // 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);
     }
 
 
@@ -203,4 +319,5 @@ public class PdfSegmentationService {
 
     }
 
+
 }

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

@@ -113,6 +113,14 @@ public class SectionsBuilderService {
             paragraphMap.computeIfAbsent(1, x -> new TreeSet<>()).add(paragraph);
         }
 
+        // first page is always a paragraph, else we can't process pages 1..N,
+        // where N is the first found page with a paragraph
+        if (paragraphMap.get(1) == null) {
+            Paragraph paragraph = new Paragraph();
+            document.getParagraphs().add(paragraph);
+            paragraphMap.computeIfAbsent(1, x -> new TreeSet<>()).add(paragraph);
+        }
+
         for (Page page : document.getPages()) {
             for (PdfImage image : page.getImages()) {
                 SortedSet<Paragraph> paragraphsOnPage = paragraphMap.get(page.getPageNumber());

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

@@ -13,4 +13,6 @@ public class RedactionServiceSettings {
 
     private boolean enableImageClassification = true;
 
+    private float maxImageCropboxRatio = 0.9f;
+
 }

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

@@ -32,16 +32,16 @@ public class RedactionStorageService {
 
 
     @SneakyThrows
-    public void storeObject(String projectId, String fileId, FileType fileType, Object any) {
-        storageService.storeObject(StorageIdUtils.getStorageId(projectId, fileId, fileType), objectMapper.writeValueAsBytes(any));
+    public void storeObject(String dossierId, String fileId, FileType fileType, Object 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;
@@ -50,16 +50,16 @@ public class RedactionStorageService {
         try {
             return objectMapper.readValue(inputStreamResource.getInputStream(), RedactionLog.class);
         } catch (IOException e) {
-            throw new RuntimeException("Could not convert Text", e);
+            throw new RuntimeException("Could not convert RedactionLog", e);
         }
     }
 
 
-    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 +73,9 @@ public class RedactionStorageService {
     }
 
 
-    public SectionGrid getSectionGrid(String projectId, String fileId) {
+    public SectionGrid getSectionGrid(String dossierId, String fileId) {
 
-        var sectionGrid = storageService.getObject(StorageIdUtils.getStorageId(projectId, fileId, FileType.SECTION_GRID));
+        var sectionGrid = storageService.getObject(StorageIdUtils.getStorageId(dossierId, fileId, FileType.SECTION_GRID));
         try {
             return objectMapper.readValue(sectionGrid.getInputStream(), SectionGrid.class);
         } catch (IOException e) {
@@ -95,14 +95,8 @@ public class RedactionStorageService {
 
     public static class StorageIdUtils {
 
-        public static String getStorageId(String projectId, String fileId, FileType fileType) {
-            return projectId + "/" + fileId + "." + fileType.name() + fileType.getExtension();
-        }
-
-
-        public static String getStorageId(String userId, String projectId, String filename) {
-
-            return userId + "/" + projectId + "/" + filename;
+        public static String getStorageId(String dossierId, String fileId, FileType fileType) {
+            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

@@ -1,6 +1,9 @@
 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;
@@ -17,6 +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) {
@@ -27,10 +32,12 @@ public abstract class AbstractTextContainer {
         return page == other.getPage() && this.minX <= other.getTopLeft().getX() && this.maxX >= other.getTopLeft().getX() + other.getWidth() && this.minY <= other.getTopLeft().getY() && this.maxY >= other.getTopLeft().getY() + other.getHeight();
     }
 
+    @JsonIgnore
     public float getHeight() {
         return maxY - minY;
     }
-
+    
+    @JsonIgnore
     public float getWidth() {
         return maxX - minX;
     }

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();
-        r.setStart(this.getStart() - amount);
-        r.setEnd(this.getEnd() + amount);
+        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-dev.yaml

@@ -2,7 +2,7 @@ server:
   port: 8083
 
 configuration-service.url: "http://localhost:8081"
-
+file-management-service.url: "http://localhost:8085"
 
 storage:
   bucket-name: 'redaction'

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

@@ -2,6 +2,7 @@ info:
   description: Redaction Service Server V1
 
 configuration-service.url: "http://configuration-service-v1:8080"
+file-management-service.url: "http://file-management-service-v1:8080"
 image-service.url: "http://image-service-v1:8080"
 
 server:
@@ -10,6 +11,20 @@ server:
 spring:
   profiles:
     active: kubernetes
+  rabbitmq:
+    host: ${RABBITMQ_HOST:localhost}
+    port: ${RABBITMQ_PORT:5672}
+    username: ${RABBITMQ_USERNAME:user}
+    password: ${RABBITMQ_PASSWORD:rabbitmq}
+    listener:
+      simple:
+        acknowledge-mode: AUTO
+        concurrency: 2
+        retry:
+          enabled: true
+          max-attempts: 3
+          max-interval: 15000
+        prefetch: 1
 
 management:
   endpoint:

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

@@ -0,0 +1,51 @@
+package com.iqser.red.service.redaction.v1.server;
+
+import com.iqser.red.storage.commons.exception.StorageObjectDoesNotExist;
+import com.iqser.red.storage.commons.service.StorageService;
+import lombok.SneakyThrows;
+import org.apache.commons.io.IOUtils;
+import org.springframework.core.io.InputStreamResource;
+
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.FileOutputStream;
+import java.util.HashMap;
+import java.util.Map;
+
+public class FileSystemBackedStorageService extends StorageService {
+
+    private final Map<String, File> dataMap = new HashMap<>();
+
+    public FileSystemBackedStorageService() {
+        super(null, null);
+    }
+
+    @SneakyThrows
+    @Override
+    public InputStreamResource getObject(String objectId) {
+
+        var res = dataMap.get(objectId);
+        if (res == null) {
+            throw new StorageObjectDoesNotExist(new RuntimeException());
+        }
+        return new InputStreamResource(new FileInputStream(res));
+
+    }
+
+    @SneakyThrows
+    @Override
+    public void storeObject(String objectId, byte[] data) {
+        File tempFile = File.createTempFile("test", ".tmp");
+
+        IOUtils.write(data, new FileOutputStream(tempFile));
+
+        dataMap.put(objectId, tempFile);
+    }
+
+    public void clearStorage() {
+        this.dataMap.forEach((k, v) -> {
+            v.delete();
+        });
+        this.dataMap.clear();
+    }
+}

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

@@ -1,34 +0,0 @@
-package com.iqser.red.service.redaction.v1.server;
-
-import com.iqser.red.storage.commons.exception.StorageObjectDoesNotExist;
-import com.iqser.red.storage.commons.service.StorageService;
-import org.springframework.core.io.InputStreamResource;
-
-import java.io.ByteArrayInputStream;
-import java.util.HashMap;
-import java.util.Map;
-
-public class InMemoryStorageService extends StorageService {
-
-    private Map<String, byte[]> dataMap = new HashMap<>();
-
-    public InMemoryStorageService() {
-        super(null, null);
-    }
-
-    @Override
-    public InputStreamResource getObject(String objectId) {
-
-        var res = dataMap.get(objectId);
-        if (res == null) {
-            throw new StorageObjectDoesNotExist(new RuntimeException());
-        }
-        return new InputStreamResource(new ByteArrayInputStream(res));
-
-    }
-
-    @Override
-    public void storeObject(String objectId, byte[] data) {
-        dataMap.put(objectId, data);
-    }
-}

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

@@ -3,20 +3,26 @@ 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.configuration.v1.api.resource.DictionaryResource;
 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.SectionText;
 import com.iqser.red.service.redaction.v1.server.client.DictionaryClient;
 import com.iqser.red.service.redaction.v1.server.client.ImageClassificationClient;
+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.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;
+import org.junit.Ignore;
 import org.junit.Test;
 import org.junit.runner.RunWith;
 import org.kie.api.KieServices;
@@ -24,11 +30,15 @@ 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.amqp.rabbit.core.RabbitTemplate;
 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.context.TestConfiguration;
 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;
@@ -42,10 +52,10 @@ import java.util.stream.Collectors;
 
 import static org.assertj.core.api.Assertions.assertThat;
 import static org.mockito.Mockito.when;
-import static org.springframework.boot.test.context.SpringBootTest.WebEnvironment.RANDOM_PORT;
 
 @RunWith(SpringRunner.class)
-@SpringBootTest(webEnvironment = RANDOM_PORT)
+@SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT)
+@Import(RedactionIntegrationTest.RedactionIntegrationTestConfiguration.class)
 public class RedactionIntegrationTest {
 
     private static final String RULES = loadFromClassPath("drools/rules.drl");
@@ -65,6 +75,7 @@ public class RedactionIntegrationTest {
     private static final String SIGNATURE = "signature";
     private static final String FORMULA = "formula";
     private static final String OCR = "ocr";
+    private static final String DOSSIER_REDACTIONS = "dossier_redactions";
 
     private static final String RECOMMENDATION_AUTHOR = "recommendation_CBI_author";
     private static final String RECOMMENDATION_ADDRESS = "recommendation_CBI_address";
@@ -73,9 +84,13 @@ public class RedactionIntegrationTest {
 
     private static final String PII = "PII";
 
+
     @Autowired
     private RedactionController redactionController;
 
+    @Autowired
+    private ReanalyzeService reanalyzeService;
+
     @Autowired
     private ObjectMapper objectMapper;
 
@@ -97,7 +112,14 @@ public class RedactionIntegrationTest {
     @MockBean
     private AmazonS3 amazonS3;
 
+    @MockBean
+    private RabbitTemplate rabbitTemplate;
+
+    @MockBean
+    private LegalBasisClient legalBasisClient;
+
     private final Map<String, List<String>> dictionary = new HashMap<>();
+    private final Map<String, List<String>> dossierDictionary = new HashMap<>();
     private final Map<String, String> typeColorMap = new HashMap<>();
     private final Map<String, Boolean> hintTypeMap = new HashMap<>();
     private final Map<String, Boolean> caseInSensitiveMap = new HashMap<>();
@@ -106,11 +128,12 @@ public class RedactionIntegrationTest {
     private final Colors colors = new Colors();
     private final Map<String, Long> reanlysisVersions = new HashMap<>();
 
-    private final static String TEST_RULESET_ID = "123";
-    private final static String TEST_PROJECT_ID = "123";
+    private final static String TEST_DOSSIER_TEMPLATE_ID = "123";
+    private final static String TEST_DOSSIER_ID = "123";
     private final static String TEST_FILE_ID = "123";
 
-    @TestConfiguration
+    @Configuration
+    @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class})
     public static class RedactionIntegrationTestConfiguration {
 
         @Bean
@@ -132,45 +155,69 @@ public class RedactionIntegrationTest {
         @Bean
         @Primary
         public StorageService inmemoryStorage() {
-            return new InMemoryStorageService();
+            return new FileSystemBackedStorageService();
         }
 
+
+    }
+
+
+    @After
+    public void cleanupStorage() {
+        if (this.storageService instanceof FileSystemBackedStorageService) {
+            ((FileSystemBackedStorageService) this.storageService).clearStorage();
+        }
     }
 
 
     @Before
     public void stubClients() {
 
-        when(rulesClient.getVersion(TEST_RULESET_ID)).thenReturn(0L);
-        when(rulesClient.getRules(TEST_RULESET_ID)).thenReturn(new RulesResponse(RULES));
+        when(rulesClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(0L);
+        when(rulesClient.getRules(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(new RulesResponse(RULES));
 
         loadDictionaryForTest();
         loadTypeForTest();
-        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(0L);
-        when(dictionaryClient.getAllTypes(TEST_RULESET_ID)).thenReturn(TypeResponse.builder()
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(0L);
+        when(dictionaryClient.getAllTypes(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(TypeResponse.builder()
                 .types(getTypeResponse())
                 .build());
-        when(dictionaryClient.getDictionaryForType(VERTEBRATE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(VERTEBRATE));
-        when(dictionaryClient.getDictionaryForType(ADDRESS, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(ADDRESS));
-        when(dictionaryClient.getDictionaryForType(AUTHOR, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(AUTHOR));
-        when(dictionaryClient.getDictionaryForType(SPONSOR, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(SPONSOR));
-        when(dictionaryClient.getDictionaryForType(NO_REDACTION_INDICATOR, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(NO_REDACTION_INDICATOR));
-        when(dictionaryClient.getDictionaryForType(REDACTION_INDICATOR, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(REDACTION_INDICATOR));
-        when(dictionaryClient.getDictionaryForType(HINT_ONLY, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(HINT_ONLY));
-        when(dictionaryClient.getDictionaryForType(MUST_REDACT, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(MUST_REDACT));
-        when(dictionaryClient.getDictionaryForType(PUBLISHED_INFORMATION, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(PUBLISHED_INFORMATION));
-        when(dictionaryClient.getDictionaryForType(TEST_METHOD, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(TEST_METHOD));
-        when(dictionaryClient.getDictionaryForType(PII, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(PII));
-        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_AUTHOR, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(RECOMMENDATION_AUTHOR));
-        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_ADDRESS, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(RECOMMENDATION_ADDRESS));
-        when(dictionaryClient.getDictionaryForType(FALSE_POSITIVE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(FALSE_POSITIVE));
-        when(dictionaryClient.getDictionaryForType(PURITY, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(PURITY));
-        when(dictionaryClient.getDictionaryForType(IMAGE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(IMAGE));
-        when(dictionaryClient.getDictionaryForType(OCR, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(OCR));
-        when(dictionaryClient.getDictionaryForType(LOGO, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(LOGO));
-        when(dictionaryClient.getDictionaryForType(SIGNATURE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(SIGNATURE));
-        when(dictionaryClient.getDictionaryForType(FORMULA, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(FORMULA));
-        when(dictionaryClient.getColors(TEST_RULESET_ID)).thenReturn(colors);
+
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, TEST_DOSSIER_ID)).thenReturn(0L);
+        when(dictionaryClient.getAllTypes(TEST_DOSSIER_TEMPLATE_ID, TEST_DOSSIER_ID)).thenReturn(TypeResponse.builder()
+                .types(List.of(TypeResult.builder()
+                        .type(DOSSIER_REDACTIONS)
+                        .dossierTemplateId(TEST_DOSSIER_TEMPLATE_ID)
+                        .hexColor( "#ffe187")
+                        .isHint(hintTypeMap.get(DOSSIER_REDACTIONS))
+                        .isCaseInsensitive(caseInSensitiveMap.get(DOSSIER_REDACTIONS))
+                        .isRecommendation(recommendationTypeMap.get(DOSSIER_REDACTIONS))
+                        .rank(rankTypeMap.get(DOSSIER_REDACTIONS))
+                        .build()))
+                .build());
+
+        when(dictionaryClient.getDictionaryForType(VERTEBRATE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(VERTEBRATE, false));
+        when(dictionaryClient.getDictionaryForType(ADDRESS, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(ADDRESS, false));
+        when(dictionaryClient.getDictionaryForType(AUTHOR, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(AUTHOR, false));
+        when(dictionaryClient.getDictionaryForType(SPONSOR, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(SPONSOR, false));
+        when(dictionaryClient.getDictionaryForType(NO_REDACTION_INDICATOR, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(NO_REDACTION_INDICATOR, false));
+        when(dictionaryClient.getDictionaryForType(REDACTION_INDICATOR, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(REDACTION_INDICATOR, false));
+        when(dictionaryClient.getDictionaryForType(HINT_ONLY, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(HINT_ONLY, false));
+        when(dictionaryClient.getDictionaryForType(MUST_REDACT, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(MUST_REDACT, false));
+        when(dictionaryClient.getDictionaryForType(PUBLISHED_INFORMATION, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(PUBLISHED_INFORMATION, false));
+        when(dictionaryClient.getDictionaryForType(TEST_METHOD, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(TEST_METHOD, false));
+        when(dictionaryClient.getDictionaryForType(PII, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(PII, false));
+        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_AUTHOR, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(RECOMMENDATION_AUTHOR, false));
+        when(dictionaryClient.getDictionaryForType(RECOMMENDATION_ADDRESS, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(RECOMMENDATION_ADDRESS, false));
+        when(dictionaryClient.getDictionaryForType(FALSE_POSITIVE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(FALSE_POSITIVE, false));
+        when(dictionaryClient.getDictionaryForType(PURITY, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(PURITY, false));
+        when(dictionaryClient.getDictionaryForType(IMAGE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(IMAGE, false));
+        when(dictionaryClient.getDictionaryForType(OCR, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(OCR, false));
+        when(dictionaryClient.getDictionaryForType(LOGO, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(LOGO, false));
+        when(dictionaryClient.getDictionaryForType(SIGNATURE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(SIGNATURE, false));
+        when(dictionaryClient.getDictionaryForType(FORMULA, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(FORMULA, false));
+        when(dictionaryClient.getDictionaryForType(DOSSIER_REDACTIONS, TEST_DOSSIER_TEMPLATE_ID, TEST_DOSSIER_ID)).thenReturn(getDictionaryResponse(DOSSIER_REDACTIONS, true));
+        when(dictionaryClient.getColors(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(colors);
     }
 
 
@@ -276,6 +323,11 @@ public class RedactionIntegrationTest {
                         .stream()
                         .map(this::cleanDictionaryEntry)
                         .collect(Collectors.toSet()));
+        dossierDictionary.computeIfAbsent(DOSSIER_REDACTIONS, v -> new ArrayList<>())
+                .addAll(ResourceLoader.load("dictionaries/dossier_redactions.txt")
+                        .stream()
+                        .map(this::cleanDictionaryEntry)
+                        .collect(Collectors.toSet()));
     }
 
 
@@ -328,6 +380,7 @@ public class RedactionIntegrationTest {
         hintTypeMap.put(FORMULA, false);
         hintTypeMap.put(LOGO, false);
         hintTypeMap.put(SIGNATURE, false);
+        hintTypeMap.put(DOSSIER_REDACTIONS, false);
 
         caseInSensitiveMap.put(VERTEBRATE, true);
         caseInSensitiveMap.put(ADDRESS, false);
@@ -349,6 +402,7 @@ public class RedactionIntegrationTest {
         caseInSensitiveMap.put(SIGNATURE, true);
         caseInSensitiveMap.put(LOGO, true);
         caseInSensitiveMap.put(FORMULA, true);
+        caseInSensitiveMap.put(DOSSIER_REDACTIONS, false);
 
         recommendationTypeMap.put(VERTEBRATE, false);
         recommendationTypeMap.put(ADDRESS, false);
@@ -370,6 +424,7 @@ public class RedactionIntegrationTest {
         recommendationTypeMap.put(FORMULA, false);
         recommendationTypeMap.put(SIGNATURE, false);
         recommendationTypeMap.put(LOGO, false);
+        recommendationTypeMap.put(DOSSIER_REDACTIONS, false);
 
         rankTypeMap.put(FALSE_POSITIVE, 160);
         rankTypeMap.put(PURITY, 155);
@@ -391,6 +446,7 @@ public class RedactionIntegrationTest {
         rankTypeMap.put(LOGO, 28);
         rankTypeMap.put(SIGNATURE, 27);
         rankTypeMap.put(FORMULA, 26);
+        rankTypeMap.put(DOSSIER_REDACTIONS, 200);
 
         colors.setDefaultColor("#acfc00");
         colors.setNotRedacted("#cccccc");
@@ -405,7 +461,7 @@ public class RedactionIntegrationTest {
                 .stream()
                 .map(typeColor -> TypeResult.builder()
                         .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()))
@@ -417,11 +473,11 @@ public class RedactionIntegrationTest {
     }
 
 
-    private DictionaryResponse getDictionaryResponse(String type) {
+    private DictionaryResponse getDictionaryResponse(String type, boolean isDossierDictionary) {
 
         return DictionaryResponse.builder()
                 .hexColor(typeColorMap.get(type))
-                .entries(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))
@@ -441,6 +497,71 @@ public class RedactionIntegrationTest {
 
 
     @Test
+    public void test270Rotated() {
+        AnalyzeRequest request = prepareStorage("files/Minimal Examples/270Rotated.pdf");
+        MemoryStats.printMemoryStats();
+        AnalyzeResult result = reanalyzeService.analyze(request);
+        assertThat(result).isNotNull();
+    }
+
+
+    @Test
+    @Ignore
+    public void testLargeScannedFileOOM() {
+        AnalyzeRequest request = prepareStorage("scanned/VV-377031.pdf");
+        MemoryStats.printMemoryStats();
+        AnalyzeResult result = reanalyzeService.analyze(request);
+        assertThat(result).isNotNull();
+    }
+
+    @Test
+    public void testMergedImages() throws IOException {
+
+        long start = System.currentTimeMillis();
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/merge_images.pdf");
+
+        AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
+        AnalyzeResult result = reanalyzeService.analyze(request);
+
+        Map<String, List<RedactionLogEntry>> duplicates = new HashMap<>();
+
+        var redactionLog = redactionStorageService.getRedactionLog(TEST_DOSSIER_ID, TEST_FILE_ID);
+
+        redactionLog.getRedactionLogEntry().forEach(entry -> {
+            duplicates.computeIfAbsent(entry.getId(), v -> new ArrayList<>()).add(entry);
+        });
+
+        duplicates.entrySet().forEach(entry -> {
+            assertThat(entry.getValue().size()).isEqualTo(1);
+        });
+
+        dictionary.get(AUTHOR).add("Drinking water");
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(1L);
+
+        AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
+                .dossierId(TEST_DOSSIER_ID)
+                .fileId(TEST_FILE_ID)
+                .build());
+
+        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Annotated3.pdf")) {
+            fileOutputStream.write(annotateResponse.getDocument());
+        }
+        long rstart = System.currentTimeMillis();
+        reanalyzeService.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 {
 
         long start = System.currentTimeMillis();
@@ -456,11 +577,11 @@ public class RedactionIntegrationTest {
 
             AnalyzeRequest request = prepareStorage(new FileInputStream((path)));
             System.out.println("Redacting file : " + path.getName());
-            AnalyzeResult result = redactionController.analyze(request);
+            AnalyzeResult result = reanalyzeService.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);
@@ -471,10 +592,10 @@ public class RedactionIntegrationTest {
             });
 
             dictionary.get(AUTHOR).add("Drinking water");
-            when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(1L);
+            when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(1L);
 
             long rstart = System.currentTimeMillis();
-            redactionController.reanalyze(request);
+            reanalyzeService.reanalyze(request);
 
             long rend = System.currentTimeMillis();
             System.out.println("reanalysis analysis duration: " + (rend - rstart));
@@ -509,15 +630,14 @@ public class RedactionIntegrationTest {
     @Test
     public void redactionTest() throws IOException {
 
-        System.out.println("redactionTest");
         long start = System.currentTimeMillis();
         ClassPathResource pdfFileResource = new ClassPathResource("files/new/Single Study - Oral (Gavage) Mouse.pdf");
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
-        AnalyzeResult result = redactionController.analyze(request);
+        AnalyzeResult result = reanalyzeService.analyze(request);
 
-        var redactionLog = redactionStorageService.getRedactionLog(TEST_PROJECT_ID, TEST_FILE_ID);
-        var text = redactionStorageService.getText(TEST_PROJECT_ID, TEST_FILE_ID);
+        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()) {
@@ -530,7 +650,7 @@ 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;
@@ -560,21 +680,21 @@ public class RedactionIntegrationTest {
         dictionary.get(AUTHOR).add("physical");
         reanlysisVersions.put("physical", 2L);
 
-        dictionary.get(VERTEBRATE).add("s-metolachlor");
-        reanlysisVersions.put("s-metolachlor", 3L);
+//        dictionary.get(VERTEBRATE).add("s-metolachlor");
+//        reanlysisVersions.put("s-metolachlor", 3L);
 
-        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(3L);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(3L);
 
-        when(dictionaryClient.getDictionaryForType(VERTEBRATE, TEST_RULESET_ID)).thenReturn(getDictionaryResponse(VERTEBRATE));
+        when(dictionaryClient.getDictionaryForType(VERTEBRATE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(getDictionaryResponse(VERTEBRATE, false));
 
         start = System.currentTimeMillis();
-        AnalyzeResult reanalyzeResult = redactionController.reanalyze(request);
+        AnalyzeResult reanalyzeResult = reanalyzeService.reanalyze(request);
 
         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());
 
@@ -592,10 +712,10 @@ public class RedactionIntegrationTest {
         long start = System.currentTimeMillis();
 
         AnalyzeRequest request = prepareStorage("files/Metolachlor/S-Metolachlor_RAR_02_Volume_2_2018-09-06.pdf");
-        AnalyzeResult result = redactionController.analyze(request);
+        AnalyzeResult result = reanalyzeService.analyze(request);
 
         AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
-                .projectId(TEST_PROJECT_ID)
+                .dossierId(TEST_DOSSIER_ID)
                 .fileId(TEST_FILE_ID)
                 .build());
 
@@ -654,7 +774,7 @@ public class RedactionIntegrationTest {
 
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
         request.setManualRedactions(manualRedactions);
-        AnalyzeResult result = redactionController.analyze(request);
+        AnalyzeResult result = reanalyzeService.analyze(request);
 
         manualRedactions.getEntriesToAdd().add(manualRedactionEntry);
         manualRedactions.setIdsToRemove(Set.of(IdRemoval.builder()
@@ -662,11 +782,11 @@ public class RedactionIntegrationTest {
                 .status(Status.APPROVED)
                 .build()));
 
-        redactionController.reanalyze(request);
+        reanalyzeService.reanalyze(request);
 
 
         AnnotateResponse annotateResponse = redactionController.annotate(AnnotateRequest.builder()
-                .projectId(TEST_PROJECT_ID)
+                .dossierId(TEST_DOSSIER_ID)
                 .fileId(TEST_FILE_ID)
                 .build());
 
@@ -685,15 +805,15 @@ 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/new/Single Study - Oral (Gavage) Mouse.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);
@@ -713,9 +833,9 @@ 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.sections(redactionRequest);
@@ -735,9 +855,9 @@ 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);
@@ -757,9 +877,9 @@ 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);
@@ -777,9 +897,9 @@ public class RedactionIntegrationTest {
 
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
-        AnalyzeResult result = redactionController.analyze(request);
+        AnalyzeResult result = reanalyzeService.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()) {
@@ -800,15 +920,15 @@ public class RedactionIntegrationTest {
     private AnalyzeRequest prepareStorage(InputStream stream) {
 
         AnalyzeRequest request = AnalyzeRequest.builder()
-                .ruleSetId(TEST_RULESET_ID)
-                .projectId(TEST_PROJECT_ID)
+                .dossierTemplateId(TEST_DOSSIER_TEMPLATE_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;
 
@@ -824,10 +944,10 @@ public class RedactionIntegrationTest {
 
         AnalyzeRequest request = prepareStorage(pdfFileResource.getInputStream());
 
-        AnalyzeResult result = redactionController.analyze(request);
+        AnalyzeResult result = reanalyzeService.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/service/EntityRedactionServiceTest.java

@@ -2,16 +2,18 @@ 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.redaction.v1.server.InMemoryStorageService;
+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.apache.pdfbox.pdmodel.PDDocument;
 import org.junit.Before;
 import org.junit.Ignore;
 import org.junit.Test;
@@ -22,10 +24,13 @@ 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.context.TestConfiguration;
 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;
@@ -43,8 +48,9 @@ import java.util.concurrent.atomic.AtomicLong;
 import static org.assertj.core.api.Assertions.assertThat;
 import static org.mockito.Mockito.when;
 
-@SpringBootTest
 @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");
@@ -73,9 +79,13 @@ public class EntityRedactionServiceTest {
     @MockBean
     private AmazonS3 amazonS3;
 
-    private final static String TEST_RULESET_ID = "123";
+    @MockBean
+    private LegalBasisClient legalBasisClient;
 
-    @TestConfiguration
+    private final static String TEST_DOSSIER_TEMPLATE_ID = "123";
+
+    @Configuration
+    @EnableAutoConfiguration(exclude = {RabbitAutoConfiguration.class})
     public static class RedactionIntegrationTestConfiguration {
 
         @Bean
@@ -98,7 +108,7 @@ public class EntityRedactionServiceTest {
         @Bean
         @Primary
         public StorageService inmemoryStorage() {
-            return new InMemoryStorageService();
+            return new FileSystemBackedStorageService();
         }
 
     }
@@ -108,8 +118,8 @@ public class EntityRedactionServiceTest {
     public void testNestedEntitiesRemoval() {
 
         Set<Entity> entities = new HashSet<>();
-        Entity nested = new Entity("nested", "fake type", 10, 16, "fake headline", 0, false);
-        Entity nesting = new Entity("nesting nested", "fake type", 2, 16, "fake headline", 0, false);
+        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);
@@ -128,24 +138,22 @@ public class EntityRedactionServiceTest {
         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)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
 
         DictionaryResponse sponsorResponse = DictionaryResponse.builder()
                 .entries(Collections.emptyList())
                 .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID)).thenReturn(sponsorResponse);
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
 
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            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
-        }
+        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_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
     }
 
 
@@ -157,23 +165,21 @@ public class EntityRedactionServiceTest {
         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)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
         DictionaryResponse sponsorResponse = DictionaryResponse.builder()
                 .entries(Collections.emptyList())
                 .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID)).thenReturn(sponsorResponse);
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
 
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            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
-        }
+        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_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
     }
 
 
@@ -182,64 +188,58 @@ public class EntityRedactionServiceTest {
 
         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)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_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)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
         DictionaryResponse sponsorResponse = DictionaryResponse.builder()
                 .entries(Collections.emptyList())
                 .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID)).thenReturn(sponsorResponse);
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            assertThat(classifiedDoc.getEntities()
-                    .entrySet()
-                    .stream()
-                    .noneMatch(entry -> entry.getValue().stream().anyMatch(e -> e.getMatchedRule() == 9))).isTrue();
-        }
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
+        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_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");
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            assertThat(classifiedDoc.getEntities()
-                    .entrySet()
-                    .stream()
-                    .noneMatch(entry -> entry.getValue().stream().anyMatch(e -> e.getMatchedRule() == 9))).isTrue();
-        }
+        classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_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)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_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)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(sponsorResponse);
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-            assertThat(classifiedDoc.getEntities().get(1).stream()
-                    .filter(entity -> entity.getMatchedRule() == 9)
-                    .count()).isEqualTo(10);
-        }
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
+        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_ID, null, "dossierId");
+        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+        assertThat(classifiedDoc.getEntities().get(1).stream()
+                .filter(entity -> entity.getMatchedRule() == 9)
+                .count()).isEqualTo(10);
 
     }
 
@@ -283,32 +283,30 @@ public class EntityRedactionServiceTest {
                 "\"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);
+        when(rulesClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(RULES_VERSION.incrementAndGet());
+        when(rulesClient.getRules(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(new RulesResponse(tableRules));
+        droolsExecutionService.updateRules(TEST_DOSSIER_TEMPLATE_ID);
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Applicant Producer Table.pdf");
-        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_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)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
         DictionaryResponse sponsorResponse = DictionaryResponse.builder()
                 .entries(Collections.emptyList())
                 .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID)).thenReturn(sponsorResponse);
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-            assertThat(classifiedDoc.getEntities().get(1).stream()
-                    .filter(entity -> entity.getMatchedRule() == 6)
-                    .count()).isEqualTo(13);
-        }
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
+        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_ID, null, "dossierId");
+        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+        assertThat(classifiedDoc.getEntities().get(1).stream()
+                .filter(entity -> entity.getMatchedRule() == 6)
+                .count()).isEqualTo(13);
 
     }
 
@@ -324,32 +322,30 @@ public class EntityRedactionServiceTest {
                 "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);
+        when(rulesClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(RULES_VERSION.incrementAndGet());
+        when(rulesClient.getRules(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(new RulesResponse(tableRules));
+        droolsExecutionService.updateRules(TEST_DOSSIER_TEMPLATE_ID);
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/batches_new_line.pdf");
-        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
         DictionaryResponse addressResponse = DictionaryResponse.builder()
                 .entries(Collections.emptyList())
                 .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
         DictionaryResponse authorResponse = DictionaryResponse.builder()
                 .entries(Collections.emptyList())
                 .build();
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID)).thenReturn(authorResponse);
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(dictionaryResponse);
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-            assertThat(classifiedDoc.getEntities().get(1).stream()
-                    .filter(entity -> entity.getMatchedRule() == 11)
-                    .count()).isEqualTo(1);
-        }
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
+        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_ID, null, "dossierId");
+        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+        assertThat(classifiedDoc.getEntities().get(1).stream()
+                .filter(entity -> entity.getMatchedRule() == 11)
+                .count()).isEqualTo(1);
 
     }
 
@@ -363,24 +359,22 @@ public class EntityRedactionServiceTest {
                 .entries(toDictionaryEntry(Arrays.asList("Bissig R.", "Thanei P.")))
                 .build();
 
-        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
 
         DictionaryResponse sponsorResponse = DictionaryResponse.builder()
                 .entries(Collections.emptyList())
                 .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID)).thenReturn(sponsorResponse);
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            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
-        }
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
+        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_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");
 
@@ -388,20 +382,18 @@ public class EntityRedactionServiceTest {
                 .entries(toDictionaryEntry(Arrays.asList("Tribolet, R.", "Muir, G.", "Kühne-Thu, H.", "Close, C.")))
                 .build();
 
-        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
 
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 9).count()).isEqualTo(3);
-            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(9);
-        }
+        classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_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);
     }
 
 
@@ -415,23 +407,21 @@ public class EntityRedactionServiceTest {
                 .entries(toDictionaryEntry(Collections.singletonList("Aldershof S.")))
                 .build();
 
-        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_RULESET_ID)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_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)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
 
         DictionaryResponse sponsorResponse = DictionaryResponse.builder()
                 .entries(Collections.emptyList())
                 .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID)).thenReturn(sponsorResponse);
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document classifiedDoc = pdfSegmentationService.parseDocument(pdDocument);
-            entityRedactionService.processDocument(classifiedDoc, TEST_RULESET_ID, null);
-            assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
-            assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(6);
-        }
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
+        Document classifiedDoc = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        entityRedactionService.processDocument(classifiedDoc, TEST_DOSSIER_TEMPLATE_ID, null, "dossierId");
+        assertThat(classifiedDoc.getEntities()).hasSize(1); // one page
+        assertThat(classifiedDoc.getEntities().get(1).stream().filter(entity -> entity.getMatchedRule() == 8).count()).isEqualTo(6);
     }
 
 
@@ -456,31 +446,31 @@ public class EntityRedactionServiceTest {
                 "\"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" +
+                "        section.redact(\"address\", 9, \"Redacted because row is a vertebrate sgitudy\", \"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));
+        when(rulesClient.getVersion(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(RULES_VERSION.incrementAndGet());
+        when(rulesClient.getRules(TEST_DOSSIER_TEMPLATE_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()))
+                        TypeResult.builder().dossierTemplateId(TEST_DOSSIER_TEMPLATE_ID).type(AUTHOR_CODE).hexColor("#ffff00").build(),
+                        TypeResult.builder().dossierTemplateId(TEST_DOSSIER_TEMPLATE_ID).type(ADDRESS_CODE).hexColor("#ff00ff").build(),
+                        TypeResult.builder().dossierTemplateId(TEST_DOSSIER_TEMPLATE_ID).type(SPONSOR_CODE).hexColor("#00ffff").build()))
                 .build();
-        when(dictionaryClient.getVersion(TEST_RULESET_ID)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
-        when(dictionaryClient.getAllTypes(TEST_RULESET_ID)).thenReturn(typeResponse);
+        when(dictionaryClient.getVersion(TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(DICTIONARY_VERSION.incrementAndGet());
+        when(dictionaryClient.getAllTypes(TEST_DOSSIER_TEMPLATE_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)).thenReturn(dictionaryResponse);
+        when(dictionaryClient.getDictionaryForType(AUTHOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(dictionaryResponse);
         DictionaryResponse addressResponse = DictionaryResponse.builder()
                 .build();
-        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_RULESET_ID)).thenReturn(addressResponse);
+        when(dictionaryClient.getDictionaryForType(ADDRESS_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(addressResponse);
         DictionaryResponse sponsorResponse = DictionaryResponse.builder()
                 .build();
-        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_RULESET_ID)).thenReturn(sponsorResponse);
+        when(dictionaryClient.getDictionaryForType(SPONSOR_CODE, TEST_DOSSIER_TEMPLATE_ID, DictionaryResource.GLOBAL_DOSSIER)).thenReturn(sponsorResponse);
 
         Colors colors = new Colors();
         colors.setDefaultColor("#acfc00");
@@ -488,7 +478,7 @@ public class EntityRedactionServiceTest {
         colors.setRequestAdd("#04b093");
         colors.setRequestRemove("#04b093");
 
-        when(dictionaryClient.getColors(TEST_RULESET_ID)).thenReturn(colors);
+        when(dictionaryClient.getColors(TEST_DOSSIER_TEMPLATE_ID)).thenReturn(colors);
     }
 
 
@@ -510,7 +500,7 @@ public class EntityRedactionServiceTest {
         }
     }
 
-    private List<DictionaryEntry> toDictionaryEntry(List<String> entries){
+    private List<DictionaryEntry> toDictionaryEntry(List<String> entries) {
         List<DictionaryEntry> dictionaryEntries = new ArrayList<>();
         entries.forEach(entry -> {
             dictionaryEntries.add(new DictionaryEntry(entry, 1L, false));

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

@@ -1,22 +1,28 @@
 package com.iqser.red.service.redaction.v1.server.segmentation;
 
 import com.amazonaws.services.s3.AmazonS3;
+import com.iqser.red.service.redaction.v1.server.Application;
 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.service.BlockificationService;
+import com.iqser.red.service.redaction.v1.server.client.LegalBasisClient;
 import com.iqser.red.service.redaction.v1.server.redaction.model.PdfImage;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Cell;
 import com.iqser.red.service.redaction.v1.server.tableextraction.model.Table;
 import com.iqser.red.service.redaction.v1.server.tableextraction.service.RulingCleaningService;
 import com.iqser.red.service.redaction.v1.server.tableextraction.service.TableExtractionService;
-import org.apache.pdfbox.pdmodel.PDDocument;
 import org.junit.Ignore;
 import org.junit.Test;
 import org.junit.runner.RunWith;
 import org.kie.api.runtime.KieContainer;
+import org.springframework.amqp.rabbit.core.RabbitTemplate;
 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.Configuration;
+import org.springframework.context.annotation.Import;
 import org.springframework.core.io.ClassPathResource;
 import org.springframework.test.context.junit4.SpringRunner;
 
@@ -30,8 +36,10 @@ import java.util.stream.Collectors;
 
 import static org.assertj.core.api.Assertions.assertThat;
 
-@SpringBootTest
+
 @RunWith(SpringRunner.class)
+@SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT)
+@Import(PdfSegmentationServiceTest.TestConfiguration.class)
 public class PdfSegmentationServiceTest {
 
     @Autowired
@@ -52,25 +60,46 @@ public class PdfSegmentationServiceTest {
     @MockBean
     private AmazonS3 amazonS3;
 
+    @MockBean
+    private RabbitTemplate rabbitTemplate;
+
+    @MockBean
+    private LegalBasisClient legalBasisClient;
+
+    @Configuration
+    @EnableAutoConfiguration(exclude = { RabbitAutoConfiguration.class})
+    public static class TestConfiguration {
+
+    }
+
+    @Test
+    public void testMergeImages() throws IOException {
+
+        ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/270Rotated.pdf");
+
+        Document document = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        assertThat(document.getPages().get(0).getImages().size()).isEqualTo(1);
+        assertThat(document.getPages().get(1).getImages().size()).isEqualTo(0);
+
+    }
+
     @Test
     @Ignore
     public void testExtractImages() throws IOException {
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/new/Single Study - Oral (Gavage) Mouse.pdf");
 
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document document = pdfSegmentationService.parseDocument(pdDocument);
-            int i = 0;
-            for (Page page : document.getPages()) {
-                for (PdfImage image : page.getImages()) {
-                    try (ByteArrayOutputStream baos = new ByteArrayOutputStream()) {
-                        ImageIO.write(image.getImage(), "png", baos);
-                        try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Image " + i + ".png")) {
-                            fileOutputStream.write(baos.toByteArray());
-                        }
+        Document document = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        int i = 0;
+        for (Page page : document.getPages()) {
+            for (PdfImage image : page.getImages()) {
+                try (ByteArrayOutputStream baos = new ByteArrayOutputStream()) {
+                    ImageIO.write(image.getImage(), "png", baos);
+                    try (FileOutputStream fileOutputStream = new FileOutputStream("/tmp/Image " + i + ".png")) {
+                        fileOutputStream.write(baos.toByteArray());
                     }
-                    i++;
                 }
+                i++;
             }
         }
     }
@@ -81,21 +110,19 @@ public class PdfSegmentationServiceTest {
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Spanning Cells.pdf");
 
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document document = pdfSegmentationService.parseDocument(pdDocument);
-            assertThat(document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())).isNotEmpty();
-            Table table = document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())
-                    .get(0);
-            assertThat(table.getColCount()).isEqualTo(6);
-            assertThat(table.getRowCount()).isEqualTo(13);
-            assertThat(table.getRows().stream().mapToInt(List::size).sum()).isEqualTo(6 * 13);
-        }
+        Document document = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        assertThat(document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())).isNotEmpty();
+        Table table = document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())
+                .get(0);
+        assertThat(table.getColCount()).isEqualTo(6);
+        assertThat(table.getRowCount()).isEqualTo(13);
+        assertThat(table.getRows().stream().mapToInt(List::size).sum()).isEqualTo(6 * 13);
     }
 
 
@@ -104,38 +131,36 @@ public class PdfSegmentationServiceTest {
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Merge Table.pdf");
 
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document document = pdfSegmentationService.parseDocument(pdDocument);
-            assertThat(document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())).isNotEmpty();
-            Table firstTable = document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())
-                    .get(0);
-            assertThat(firstTable.getColCount()).isEqualTo(8);
-            assertThat(firstTable.getRowCount()).isEqualTo(1);
-            Table secondTable = document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())
-                    .get(1);
-            assertThat(secondTable.getColCount()).isEqualTo(8);
-            assertThat(secondTable.getRowCount()).isEqualTo(2);
-            List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
-                    .get(0)
-                    .stream()
-                    .map(Collections::singletonList)
-                    .collect(Collectors.toList());
-            assertThat(secondTable.getRows()
-                    .stream()
-                    .allMatch(row -> row.stream()
-                            .map(Cell::getHeaderCells)
-                            .collect(Collectors.toList())
-                            .equals(firstTableHeaderCells))).isTrue();
-        }
+        Document document = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        assertThat(document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())).isNotEmpty();
+        Table firstTable = document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())
+                .get(0);
+        assertThat(firstTable.getColCount()).isEqualTo(8);
+        assertThat(firstTable.getRowCount()).isEqualTo(1);
+        Table secondTable = document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())
+                .get(1);
+        assertThat(secondTable.getColCount()).isEqualTo(8);
+        assertThat(secondTable.getRowCount()).isEqualTo(2);
+        List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
+                .get(0)
+                .stream()
+                .map(Collections::singletonList)
+                .collect(Collectors.toList());
+        assertThat(secondTable.getRows()
+                .stream()
+                .allMatch(row -> row.stream()
+                        .map(Cell::getHeaderCells)
+                        .collect(Collectors.toList())
+                        .equals(firstTableHeaderCells))).isTrue();
     }
 
 
@@ -144,38 +169,36 @@ public class PdfSegmentationServiceTest {
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Merge Multi Page Table.pdf");
 
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document document = pdfSegmentationService.parseDocument(pdDocument);
-            assertThat(document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())).isNotEmpty();
-            Table firstTable = document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())
-                    .get(0);
-            assertThat(firstTable.getColCount()).isEqualTo(9);
-            assertThat(firstTable.getRowCount()).isEqualTo(5);
-            Table secondTable = document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())
-                    .get(1);
-            assertThat(secondTable.getColCount()).isEqualTo(9);
-            assertThat(secondTable.getRowCount()).isEqualTo(6);
-            List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
-                    .get(firstTable.getRowCount() - 1)
-                    .stream()
-                    .map(Cell::getHeaderCells)
-                    .collect(Collectors.toList());
-            assertThat(secondTable.getRows()
-                    .stream()
-                    .allMatch(row -> row.stream()
-                            .map(Cell::getHeaderCells)
-                            .collect(Collectors.toList())
-                            .equals(firstTableHeaderCells))).isTrue();
-        }
+        Document document = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        assertThat(document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())).isNotEmpty();
+        Table firstTable = document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())
+                .get(0);
+        assertThat(firstTable.getColCount()).isEqualTo(9);
+        assertThat(firstTable.getRowCount()).isEqualTo(5);
+        Table secondTable = document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())
+                .get(1);
+        assertThat(secondTable.getColCount()).isEqualTo(9);
+        assertThat(secondTable.getRowCount()).isEqualTo(6);
+        List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
+                .get(firstTable.getRowCount() - 1)
+                .stream()
+                .map(Cell::getHeaderCells)
+                .collect(Collectors.toList());
+        assertThat(secondTable.getRows()
+                .stream()
+                .allMatch(row -> row.stream()
+                        .map(Cell::getHeaderCells)
+                        .collect(Collectors.toList())
+                        .equals(firstTableHeaderCells))).isTrue();
     }
 
 
@@ -184,38 +207,36 @@ public class PdfSegmentationServiceTest {
 
         ClassPathResource pdfFileResource = new ClassPathResource("files/Minimal Examples/Rotated Table Headers.pdf");
 
-        try (PDDocument pdDocument = PDDocument.load(pdfFileResource.getInputStream())) {
-            Document document = pdfSegmentationService.parseDocument(pdDocument);
-            assertThat(document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())).isNotEmpty();
-            Table firstTable = document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())
-                    .get(0);
-            assertThat(firstTable.getColCount()).isEqualTo(8);
-            assertThat(firstTable.getRowCount()).isEqualTo(1);
-            Table secondTable = document.getParagraphs()
-                    .stream()
-                    .flatMap(paragraph -> paragraph.getTables().stream())
-                    .collect(Collectors.toList())
-                    .get(1);
-            assertThat(secondTable.getColCount()).isEqualTo(8);
-            assertThat(secondTable.getRowCount()).isEqualTo(6);
-            List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
-                    .get(0)
-                    .stream()
-                    .map(Collections::singletonList)
-                    .collect(Collectors.toList());
-            assertThat(secondTable.getRows()
-                    .stream()
-                    .allMatch(row -> row.stream()
-                            .map(Cell::getHeaderCells)
-                            .collect(Collectors.toList())
-                            .equals(firstTableHeaderCells))).isTrue();
-        }
+        Document document = pdfSegmentationService.parseDocument(pdfFileResource.getInputStream());
+        assertThat(document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())).isNotEmpty();
+        Table firstTable = document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())
+                .get(0);
+        assertThat(firstTable.getColCount()).isEqualTo(8);
+        assertThat(firstTable.getRowCount()).isEqualTo(1);
+        Table secondTable = document.getParagraphs()
+                .stream()
+                .flatMap(paragraph -> paragraph.getTables().stream())
+                .collect(Collectors.toList())
+                .get(1);
+        assertThat(secondTable.getColCount()).isEqualTo(8);
+        assertThat(secondTable.getRowCount()).isEqualTo(6);
+        List<List<Cell>> firstTableHeaderCells = firstTable.getRows()
+                .get(0)
+                .stream()
+                .map(Collections::singletonList)
+                .collect(Collectors.toList());
+        assertThat(secondTable.getRows()
+                .stream()
+                .allMatch(row -> row.stream()
+                        .map(Cell::getHeaderCells)
+                        .collect(Collectors.toList())
+                        .equals(firstTableHeaderCells))).isTrue();
     }
 
 }

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

@@ -1,5 +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"
 
 ribbon:
   ConnectTimeout: 600000

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/dossier_redactions.txt

@@ -0,0 +1 @@
+Difenoconazole

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

@@ -328,4 +328,12 @@ rule "28: Redact Logos"
         Section(matchesImageType("logo"))
     then
         section.redactImage("logo", 28, "Logo found", "Reg (EC) No 1107/2009 Art. 63 (2g)");
-    end
+    end
+
+
+rule "29: Redact Dossier Redactions"
+     when
+         Section(matchesType("dossier_redactions"))
+     then
+         section.redact("dossier_redactions", 29, "Dossier Redaction found", "Article 39(1)(2) of Regulation (EC) No 178/2002");
+     end