RED-5567: Upgraded to lastest pdftron version

Pull request #14 : RED-6321: OCR not working correctly with 3.6.0
Merge in RED/ocr-service from RED-6321 to master * commit '2d503c74a6fba0b02ac4c92af82a493165e45761': RED-6321: OCR not working correctly with 3.6.0 *added end() statements to formWriters to write their changes to the PDF contentStream *moved replaceOriginalPageWithOcrPage outside the main OCR Loop since it caused exponential RAM Usage in some cases
2023-05-26 09:20:05 +02:00 · 2023-03-06 10:16:02 +01:00 · 2023-03-06 10:08:32 +01:00 · 2023-03-02 09:28:11 +01:00 · 2023-02-28 14:56:37 +01:00 · 2023-02-28 14:07:44 +01:00
19 changed files with 299 additions and 272 deletions
@@ -95,7 +95,7 @@ public class PlanSpec {
                                        .enabled(true),
                                new InjectVariablesTask().description("Inject git Tag").path("git.tag").namespace("g").scope(InjectVariablesScope.LOCAL),
                                new VcsTagTask().description("${bamboo.g.gitTag}").tagName("${bamboo.g.gitTag}").defaultRepository())
-                        .dockerConfiguration(new DockerConfiguration().image("nexus.iqser.com:5001/infra/maven:3.8.4-openjdk-17-slim")
+                        .dockerConfiguration(new DockerConfiguration().image("nexus.knecon.com:5001/infra/maven:3.8.4-openjdk-17-slim")
                                .volume("/etc/maven/settings.xml", "/usr/share/maven/conf/settings.xml")
                                .volume("/var/run/docker.sock", "/var/run/docker.sock"))))
                .linkedRepositories("RED / " + SERVICE_NAME)
@@ -51,7 +51,7 @@ mvn -f ${bamboo_build_working_directory}/$SERVICE_NAME-v1/pom.xml \
    -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
+    -DaltDeploymentRepository=iqser_release::default::https://nexus.knecon.com/repository/red-platform-releases

 mvn --no-transfer-progress \
    -f ${bamboo_build_working_directory}/$SERVICE_NAME-image-v1/pom.xml \
@@ -21,7 +21,7 @@ then
        -f ${bamboo_build_working_directory}/$SERVICE_NAME-v1/pom.xml \
        sonar:sonar \
        -Dsonar.projectKey=RED_$SERVICE_NAME \
-        -Dsonar.host.url=https://sonarqube.iqser.com \
+        -Dsonar.host.url=https://sonarqube.knecon.com \
        -Dsonar.login=${bamboo_sonarqube_api_token_secret} \
        -Dsonar.branch.name=${bamboo_planRepository_1_branch} \
        -Dsonar.dependencyCheck.jsonReportPath=target/dependency-check-report.json \
@@ -33,7 +33,7 @@ else
        -f ${bamboo_build_working_directory}/$SERVICE_NAME-v1/pom.xml \
        sonar:sonar \
        -Dsonar.projectKey=RED_$SERVICE_NAME \
-        -Dsonar.host.url=https://sonarqube.iqser.com \
+        -Dsonar.host.url=https://sonarqube.knecon.com \
        -Dsonar.login=${bamboo_sonarqube_api_token_secret} \
        -Dsonar.pullrequest.key=${bamboo_repository_pr_key} \
        -Dsonar.pullrequest.branch=${bamboo_repository_pr_sourceBranch} \
@@ -3,9 +3,9 @@
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <parent>
-        <groupId>com.iqser.red</groupId>
+        <groupId>com.knecon.fforesight</groupId>
        <artifactId>platform-docker-dependency</artifactId>
-        <version>1.2.0</version>
+        <version>0.1.0</version>
        <relativePath/>
    </parent>
    <modelVersion>4.0.0</modelVersion>
@@ -1,4 +1,4 @@
-FROM red/base-image:2.0.0
+FROM red/base-image:2.0.2

 COPY "libs/pdftron/OCRModuleLinux.tar.gz" .
 RUN tar xvzf OCRModuleLinux.tar.gz
@@ -39,7 +39,7 @@
        <dependency>
            <groupId>com.pdftron</groupId>
            <artifactId>PDFNet</artifactId>
-            <version>9.4.0</version>
+            <version>10.1.0</version>
        </dependency>
        <dependency>
            <groupId>org.springframework.boot</groupId>
@@ -1,14 +1,12 @@
 package com.iqser.red.service.ocr.v1.server.initializer;

-import javax.annotation.PostConstruct;
-
+import com.pdftron.pdf.PDFNet;
+import lombok.RequiredArgsConstructor;
+import lombok.SneakyThrows;
 import org.springframework.beans.factory.annotation.Value;
 import org.springframework.stereotype.Component;

-import com.pdftron.pdf.PDFNet;
-
-import lombok.RequiredArgsConstructor;
-import lombok.SneakyThrows;
+import javax.annotation.PostConstruct;

@Component
@RequiredArgsConstructor
@@ -29,7 +27,5 @@ public class PDFNetInitializer {
        PDFNet.setTempPath("/tmp/pdftron");
        PDFNet.addResourceSearchPath(ocrModulePath);
        PDFNet.initialize(pdftronLicense);
-
    }
-
 }
@@ -47,16 +47,17 @@ public class ImagePositionRetrievalService {
        ElementReader reader = new ElementReader();
        for (int pageId = 1; pageId <= pdfDoc.getPageCount(); ++pageId) {
            RectCollection imagePositions = new RectCollection();
-            reader.begin(pdfDoc.getPage(pageId));

+            reader.begin(pdfDoc.getPage(pageId));
            findImagePositionsOnPage(reader, imagePositions, pdfDoc.getPage(pageId), mirrorY);
            imagePositions = mergeOverlappingRects(imagePositions);
-
            reader.end();
+
            if (imagePositions.getNumRects() > 0) {
                pageIdToImagePositions.put(pageId, imagePositions);
            }
        }
+        reader.destroy();
        return pageIdToImagePositions;
    }

@@ -93,7 +93,17 @@ public class InvisibleElementRemovalService {

            removeOverlappedElements(page, writer, context);
        }
-        pdfDoc.save(out, SDFDoc.SaveMode.REMOVE_UNUSED, null);
+
+        try {
+            pdfDoc.save(out, SDFDoc.SaveMode.LINEARIZED, null);
+        } catch (Exception e) {
+            log.error("File could not be saved after invisible element removal");
+            throw new RuntimeException(e);
+        }
+
+        writer.destroy();
+        reader.destroy();
+        pdfDoc.close();
    }


@@ -217,12 +227,14 @@ public class InvisibleElementRemovalService {

            processElements(formWriter, context);
            formWriter.end();
+            formWriter.destroy();
            context.reader().end();
        }
    }


    private void processPath(Element pathElement, ElementWriter writer, InvisibleElementRemovalContext context) throws PDFNetException {
+
        PathData pathData = pathElement.getPathData();

        if (pathData.getOperators().length == 0 && pathData.getPoints().length == 0) {
@@ -353,6 +365,7 @@ public class InvisibleElementRemovalService {

            processOverlappedElements(formWriter, context);
            formWriter.end();
+            formWriter.destroy();
            context.reader().end();
        }
    }
@@ -427,6 +440,9 @@ public class InvisibleElementRemovalService {
        rect.getGState().setStrokeColorSpace(ColorSpace.createDeviceRGB());
        rect.getGState().setStrokeColor(colorPt);
        writer.writePlacedElement(rect);
+
+        colorPt.destroy();
+        eb.destroy();
    }


@@ -7,6 +7,7 @@ import java.io.ByteArrayOutputStream;
 import java.io.IOException;
 import java.io.InputStream;
 import java.io.OutputStream;
+import java.util.HashMap;
 import java.util.Map;

 import org.springframework.amqp.rabbit.core.RabbitTemplate;
@@ -67,17 +68,20 @@ public class OCRService {

        try (ByteArrayOutputStream transferOutputStream = new ByteArrayOutputStream()) {
            try (InputStream fileStream = fileStorageService.getOriginalFileAsStream(dossierId, fileId)) {
-                long start = System.currentTimeMillis();
+                long removalStart = System.currentTimeMillis();
                log.debug("Start invisible element removal for file with dossierId {} and fileId {}", dossierId, fileId);
                invisibleElementRemovalService.removeInvisibleElements(fileStream, transferOutputStream, false);
-                long end = System.currentTimeMillis();
-                log.debug("Invisible element removal successful for file with dossierId {} and fileId {}, took {}s", dossierId, fileId, format("%.1f", (end - start) / 1000.0));
+                long removalEnd = System.currentTimeMillis();
+                log.debug("Invisible element removal successful for file with dossierId {} and fileId {}, took {}s",
+                        dossierId,
+                        fileId,
+                        format("%.1f", (removalEnd - removalStart) / 1000.0));
            }
            try (InputStream transferInputStream = new ByteArrayInputStream(transferOutputStream.toByteArray())) {
-                long start = System.currentTimeMillis();
+                long ocrStart = System.currentTimeMillis();
                runOcr(transferInputStream, out, fileId);
-                long end = System.currentTimeMillis();
-                log.info("ocr successful for file with dossierId {} and fileId {}, took {}s", dossierId, fileId, format("%.1f", (end - start) / 1000.0));
+                long ocrEnd = System.currentTimeMillis();
+                log.info("ocr successful for file with dossierId {} and fileId {}, took {}s", dossierId, fileId, format("%.1f", (ocrEnd - ocrStart) / 1000.0));
            }
        }
    }
@@ -89,18 +93,22 @@ public class OCRService {
        PDFDoc pdfDoc = new PDFDoc(fileStream);

        Map<Integer, RectCollection> pageIdToRectCollection = imagePositionRetrievalService.getImagePositionPerPage(pdfDoc, true);
+        rabbitTemplate.convertAndSend(MessagingConfiguration.OCR_STATUS_UPDATE_RESPONSE_QUEUE,
+                objectMapper.writeValueAsString(OCRStatusUpdateResponse.builder().fileId(fileId).numberOfPagesToOCR(pageIdToRectCollection.size()).build()));

        // Optimization:
        // When a page does not have a TextZone, PDFTron whites out the page. But, PDFTron scans it anyway, resulting in a longer runtime.
        // So, we need to remove pages without images.
        // Furthermore, creating a new document is *much* faster than reusing the same document and adding/removing pages one by one.
-        // Therefore, we create a new Document with a single page for every page that contains text.
+        // Therefore, we create a new Document with a single page for every page that contains an image.
+        // For some reason, if we insert the OCR singlePageDoc into the original PDFDoc inside the loop, for some documents, the RAM Usage increases exponentially with every page.
+        // This is why, we replace the OCRed Pages outside the main loop.
        int numProcessedPages = 0;
+        Map<Integer, PDFDoc> pageIdToSinglePagePdfDoc = new HashMap<>();
        for (Integer pageId : pageIdToRectCollection.keySet()) {
            try {
                PDFDoc singlePagePdfDoc = extractSinglePagePdfDoc(pdfDoc, pageId);
                processOcr(pageIdToRectCollection, pageId, singlePagePdfDoc);
-                ++numProcessedPages;

                log.info("{}/{} Page {} done, OCR regions {}",
                        numProcessedPages,
@@ -108,8 +116,8 @@ public class OCRService {
                        pageId,
                        getAllOcrTextZonesAsString(pageIdToRectCollection, pageId));

-                replaceOriginalPageWithOcrPage(pdfDoc, pageId, singlePagePdfDoc);
-                singlePagePdfDoc.close();
+                pageIdToSinglePagePdfDoc.put(pageId, singlePagePdfDoc);
+                ++numProcessedPages;

                rabbitTemplate.convertAndSend(MessagingConfiguration.OCR_STATUS_UPDATE_RESPONSE_QUEUE,
                        objectMapper.writeValueAsString(OCRStatusUpdateResponse.builder()
@@ -123,6 +131,9 @@ public class OCRService {
                throw new RuntimeException(e);
            }
        }
+        log.info("Copying {} OCRed Pages into original Document", pageIdToSinglePagePdfDoc.size());
+        pageIdToSinglePagePdfDoc.forEach((pageId, singlePagePdfDoc) -> replaceOriginalPageWithOcrPage(pdfDoc, pageId, singlePagePdfDoc));
+        Optimizer.optimize(pdfDoc);

        rabbitTemplate.convertAndSend(MessagingConfiguration.OCR_STATUS_UPDATE_RESPONSE_QUEUE,
                objectMapper.writeValueAsString(OCRStatusUpdateResponse.builder()
@@ -132,13 +143,13 @@ public class OCRService {
                        .ocrFinished(true)
                        .build()));

-        Optimizer.optimize(pdfDoc);
        try {
            pdfDoc.save(out, SDFDoc.SaveMode.LINEARIZED, null);
        } catch (Exception e) {
            log.error("Processed File with fileId {} could not be saved", fileId);
            throw new RuntimeException(e);
        }
+        pdfDoc.close();
    }


@@ -163,11 +174,13 @@ public class OCRService {
    }


-    private static void replaceOriginalPageWithOcrPage(PDFDoc pdfDoc, Integer pageId, PDFDoc ocrPageDoc) throws PDFNetException {
+    @SneakyThrows
+    private static void replaceOriginalPageWithOcrPage(PDFDoc pdfDoc, Integer pageId, PDFDoc singlePagePdfDoc) {

-        Page ocrPage = ocrPageDoc.getPage(1);
+        Page ocrPage = singlePagePdfDoc.getPage(1);
        pdfDoc.pageInsert(pdfDoc.getPageIterator(pageId), ocrPage);
        pdfDoc.pageRemove(pdfDoc.getPageIterator(pageId + 1));
+        singlePagePdfDoc.close();
    }


@@ -1,24 +1,23 @@
 package com.iqser.red.service.ocr.v1.server.service;

-import java.io.ByteArrayInputStream;
-import java.io.ByteArrayOutputStream;
-import java.io.IOException;
-
-import org.springframework.amqp.AmqpRejectAndDontRequeueException;
-import org.springframework.amqp.rabbit.annotation.RabbitHandler;
-import org.springframework.amqp.rabbit.annotation.RabbitListener;
-import org.springframework.http.HttpStatus;
-import org.springframework.stereotype.Service;
-
 import com.fasterxml.jackson.core.JsonProcessingException;
 import com.fasterxml.jackson.databind.ObjectMapper;
 import com.iqser.red.service.ocr.v1.api.model.DocumentRequest;
 import com.iqser.red.service.ocr.v1.server.client.FileStatusProcessingUpdateClient;
 import com.iqser.red.service.ocr.v1.server.configuration.MessagingConfiguration;
-
+import com.iqser.red.service.ocr.v1.server.initializer.PDFNetInitializer;
 import feign.FeignException;
 import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;
+import org.springframework.amqp.AmqpRejectAndDontRequeueException;
+import org.springframework.amqp.rabbit.annotation.RabbitHandler;
+import org.springframework.amqp.rabbit.annotation.RabbitListener;
+import org.springframework.http.HttpStatus;
+import org.springframework.stereotype.Service;
+
+import java.io.ByteArrayInputStream;
+import java.io.ByteArrayOutputStream;
+import java.io.IOException;

@Slf4j
@Service
@@ -58,7 +57,6 @@ public class OcrMessageReceiver {
        }

        fileStatusProcessingUpdateClient.ocrSuccessful(ocrRequestMessage.getDossierId(), ocrRequestMessage.getFileId());
-
    }


@@ -0,0 +1,76 @@
+package com.iqser.red.service.ocr.v1.server;
+
+import static org.assertj.core.api.Assertions.assertThat;
+
+import org.junit.jupiter.api.AfterAll;
+import org.junit.jupiter.api.AfterEach;
+import org.junit.jupiter.api.extension.ExtendWith;
+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.Bean;
+import org.springframework.context.annotation.Configuration;
+import org.springframework.context.annotation.Import;
+import org.springframework.context.annotation.Primary;
+import org.springframework.test.context.junit.jupiter.SpringExtension;
+
+import com.iqser.red.service.ocr.v1.server.utils.FileSystemBackedStorageService;
+import com.iqser.red.storage.commons.StorageAutoConfiguration;
+import com.iqser.red.storage.commons.service.StorageService;
+import com.pdftron.pdf.PDFNet;
+
+import lombok.SneakyThrows;
+
+@ExtendWith(SpringExtension.class)
+@SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT)
+@Import(AbstractTest.TestConfiguration.class)
+public class AbstractTest {
+
+    @Autowired
+    protected StorageService storageService;
+
+    @MockBean
+    protected RabbitTemplate rabbitTemplate;
+
+
+    @AfterAll
+    public static void terminatePDFNet() {
+        PDFNet.terminate();
+        System.out.println("PDFNet Terminated");
+    }
+
+
+    @SneakyThrows
+    public void dummyTest() {
+
+        // Build needs one test to not fail.
+        assertThat(1).isEqualTo(1);
+    }
+
+
+    @AfterEach
+    public void cleanupStorage() {
+
+        if (this.storageService instanceof FileSystemBackedStorageService) {
+            ((FileSystemBackedStorageService) this.storageService).clearStorage();
+        }
+    }
+
+
+
+    @Configuration
+    @EnableAutoConfiguration(exclude = {StorageAutoConfiguration.class, RabbitAutoConfiguration.class})
+    public static class TestConfiguration {
+
+        @Bean
+        @Primary
+        public StorageService inMemoryStorage() {
+
+            return new FileSystemBackedStorageService();
+        }
+
+    }
+}
@@ -1,86 +1,49 @@
 package com.iqser.red.service.ocr.v1.server;

 import static com.iqser.red.service.ocr.v1.server.utils.OsUtils.getTemporaryDirectory;
+import static com.iqser.red.service.ocr.v1.server.utils.PdfTextExtraction.extractAllTextFromDocument;
 import static org.assertj.core.api.Assertions.assertThat;

 import java.io.FileInputStream;
 import java.io.FileOutputStream;
-import java.io.IOException;
-import java.io.InputStream;
-import java.util.ArrayList;
-import java.util.List;
 import java.util.concurrent.TimeUnit;

-import org.junit.jupiter.api.AfterEach;
 import org.junit.jupiter.api.BeforeEach;
 import org.junit.jupiter.api.Disabled;
 import org.junit.jupiter.api.Test;
-import org.junit.jupiter.api.extension.ExtendWith;
-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.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.junit.jupiter.SpringExtension;

 import com.fasterxml.jackson.databind.ObjectMapper;
 import com.iqser.red.service.ocr.v1.server.service.FileStorageService;
 import com.iqser.red.service.ocr.v1.server.service.OCRService;
-import com.iqser.red.service.ocr.v1.server.utils.FileSystemBackedStorageService;
 import com.iqser.red.service.persistence.service.v1.api.model.dossiertemplate.dossier.file.FileType;
-import com.iqser.red.storage.commons.StorageAutoConfiguration;
-import com.iqser.red.storage.commons.service.StorageService;
-import com.pdftron.common.PDFNetException;
 import com.pdftron.pdf.OCRModule;
-import com.pdftron.pdf.PDFDoc;
-import com.pdftron.pdf.Page;
-import com.pdftron.pdf.PageIterator;
-import com.pdftron.pdf.TextExtractor;

 import io.micrometer.prometheus.PrometheusMeterRegistry;
 import io.micrometer.prometheus.PrometheusTimer;
 import lombok.SneakyThrows;

-@ExtendWith(SpringExtension.class)
-@SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT //
-        , properties = {"pdftron.ocrmodule.path=/YourOCRModulePath"})
-@Import(OcrServiceIntegrationTest.TestConfiguration.class)
-public class OcrServiceIntegrationTest {
-
-    @Autowired
-    protected StorageService storageService;
-
-    @Autowired
-    protected FileStorageService fileStorageService;
+@SpringBootTest(properties = {"pdftron.ocrmodule.path=/YourOCRModulePath"})
+public class OcrServiceIntegrationTest extends AbstractTest {

    @Autowired
    protected ObjectMapper objectMapper;

-    @MockBean
-    protected RabbitTemplate rabbitTemplate;
-
    @Autowired
    private OCRService ocrService;

    @Autowired
    private PrometheusMeterRegistry registry;

-
    @BeforeEach
    @SneakyThrows
-    @Disabled // OCRModule is not available on build server. If you want to run the test set the property at the top.
+    @Disabled
    public void assertOCRModuleIsLoaded() {
-
        assert OCRModule.isModuleAvailable();
    }

-
    @Test
    @Disabled // OCRModule is not available on build server. If you want to run the test set the property at the top.
    public void testOCRMetrics() {
@@ -163,60 +126,11 @@ public class OcrServiceIntegrationTest {
        }
        try (var out = new FileOutputStream(getTemporaryDirectory() + "/" + fileName + ".pdf")) {
            ocrService.runOcrOnDocument("dossier", "file", out);
+            System.out.println("File:" + getTemporaryDirectory() + "/" + fileName + ".pdf");
        }

-        System.out.println("File:" + getTemporaryDirectory() + "/" + fileName + ".pdf");
        try (var fileStream = new FileInputStream(getTemporaryDirectory() + "/" + fileName + ".pdf")) {
            return extractAllTextFromDocument(fileStream);
        }
    }
-
-
-    private static String extractAllTextFromDocument(InputStream fileStream) throws IOException, PDFNetException {
-
-        PDFDoc pdfDoc = new PDFDoc(fileStream);
-        TextExtractor extractor = new TextExtractor();
-        List<String> texts = new ArrayList<>();
-
-        PageIterator iterator = pdfDoc.getPageIterator();
-        while (iterator.hasNext()) {
-            Page page = iterator.next();
-            extractor.begin(page);
-            texts.add(extractor.getAsText());
-        }
-
-        return String.join("\n", texts);
-    }
-
-
-    @SneakyThrows
-    public void dummyTest() {
-
-        // Build needs one test to not fail.
-        assertThat(1).isEqualTo(1);
-    }
-
-
-    @AfterEach
-    public void cleanupStorage() {
-
-        if (this.storageService instanceof FileSystemBackedStorageService) {
-            ((FileSystemBackedStorageService) this.storageService).clearStorage();
-        }
-    }
-
-
-    @Configuration
-    @EnableAutoConfiguration(exclude = {StorageAutoConfiguration.class, RabbitAutoConfiguration.class})
-    public static class TestConfiguration {
-
-        @Bean
-        @Primary
-        public StorageService inMemoryStorage() {
-
-            return new FileSystemBackedStorageService();
-        }
-
-    }
-
 }
@@ -1,6 +1,8 @@
 package com.iqser.red.service.ocr.v1.server.service;

 import static com.iqser.red.service.ocr.v1.server.utils.OsUtils.getTemporaryDirectory;
+import static com.iqser.red.service.ocr.v1.server.utils.PdfDraw.drawGrid;
+import static com.iqser.red.service.ocr.v1.server.utils.PdfDraw.drawRectCollection;
 import static org.assertj.core.api.Assertions.assertThat;

 import java.io.FileInputStream;
@@ -13,43 +15,24 @@ import java.util.Map;
 import java.util.stream.Collectors;

 import org.junit.jupiter.api.Test;
-import org.junit.jupiter.api.extension.ExtendWith;
-import org.springframework.amqp.rabbit.core.RabbitTemplate;
 import org.springframework.beans.factory.annotation.Autowired;
-import org.springframework.boot.test.context.SpringBootTest;
-import org.springframework.boot.test.mock.mockito.MockBean;
-import org.springframework.context.annotation.Import;
 import org.springframework.core.io.ClassPathResource;
-import org.springframework.test.context.junit.jupiter.SpringExtension;

-import com.iqser.red.service.ocr.v1.server.Application;
-import com.iqser.red.service.ocr.v1.server.OcrServiceIntegrationTest;
+import com.iqser.red.service.ocr.v1.server.AbstractTest;
 import com.pdftron.common.PDFNetException;
-import com.pdftron.pdf.ColorPt;
-import com.pdftron.pdf.ColorSpace;
-import com.pdftron.pdf.Element;
-import com.pdftron.pdf.ElementBuilder;
 import com.pdftron.pdf.ElementWriter;
 import com.pdftron.pdf.PDFDoc;
-import com.pdftron.pdf.Page;
-import com.pdftron.pdf.Rect;
 import com.pdftron.pdf.RectCollection;
 import com.pdftron.sdf.SDFDoc;

 import lombok.SneakyThrows;

-@ExtendWith(SpringExtension.class)
-@SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT)
-@Import(OcrServiceIntegrationTest.TestConfiguration.class)
-class ImagePositionRetrievalServiceTest {
+
+class ImagePositionRetrievalServiceTest extends AbstractTest {

    @Autowired
    private ImagePositionRetrievalService imagePositionRetrievalService;

-    @MockBean
-    protected RabbitTemplate rabbitTemplate;
-
-
    @Test
    @SneakyThrows
    public void testImagePositionRetrievalForRotateTestFileWithImages() {
@@ -133,37 +116,38 @@ class ImagePositionRetrievalServiceTest {

    private List<int[]> testImagePositionDetection(String fileName) throws IOException, PDFNetException {

-        InputStream fileStream = new FileInputStream(new ClassPathResource("files/" + fileName + ".pdf").getFile().getAbsolutePath());
-        PDFDoc pdfDoc = new PDFDoc(fileStream);
+        try (InputStream fileStream = new FileInputStream(new ClassPathResource("files/" + fileName + ".pdf").getFile().getAbsolutePath())) {
+            PDFDoc pdfDoc = new PDFDoc(fileStream);

-        Map<Integer, RectCollection> pageIdToRectCollection = imagePositionRetrievalService.getImagePositionPerPage(pdfDoc, false);
+            Map<Integer, RectCollection> pageIdToRectCollection = imagePositionRetrievalService.getImagePositionPerPage(pdfDoc, false);

-        ElementWriter writer = new ElementWriter();
-        pageIdToRectCollection.forEach((pageId, rectCollection) -> {
-            try {
-                writer.begin(pdfDoc.getPage(pageId));
-                drawRectCollection(writer, rectCollection, pdfDoc.getPage(pageId));
-                drawGrid(writer, pdfDoc.getPage(pageId));
-                writer.end();
-                StringBuilder zonesString = new StringBuilder();
-                for (int j = 0; j < rectCollection.getNumRects(); ++j) {
-                    var r = rectCollection.getRectAt(j);
-                    zonesString.append(String.format("%d:[lower left (%.1f|%.1f) upper right (%.1f|%.1f)]; ", j, r.getX1(), r.getY1(), r.getX2(), r.getY2()));
+            ElementWriter writer = new ElementWriter();
+            pageIdToRectCollection.forEach((pageId, rectCollection) -> {
+                try {
+                    writer.begin(pdfDoc.getPage(pageId));
+                    drawRectCollection(writer, rectCollection);
+                    drawGrid(writer, pdfDoc.getPage(pageId));
+                    writer.end();
+                    StringBuilder zonesString = new StringBuilder();
+                    for (int j = 0; j < rectCollection.getNumRects(); ++j) {
+                        var r = rectCollection.getRectAt(j);
+                        zonesString.append(String.format("%d:[lower left (%.1f|%.1f) upper right (%.1f|%.1f)]; ", j, r.getX1(), r.getY1(), r.getX2(), r.getY2()));
+                    }
+                    System.out.printf("Page %d: rotation: %d OCR regions %s \n", pageId, pdfDoc.getPage(pageId).getRotation(), zonesString);
+                } catch (PDFNetException e) {
+                    throw new RuntimeException(e);
                }
-                System.out.printf("Page %d: rotation: %d OCR regions %s \n", pageId, pdfDoc.getPage(pageId).getRotation(), zonesString);
-            } catch (PDFNetException e) {
-                throw new RuntimeException(e);
-            }
-        });
+            });

-        // Check visually for red Rectangles to match images in the saved pdf file
-        try (var out = new FileOutputStream(getTemporaryDirectory() + "/" + fileName + "_IMAGE_BBOX.pdf")) {
-            out.write(pdfDoc.save(SDFDoc.SaveMode.LINEARIZED, null));
+            // Check visually for red Rectangles to match images in the saved pdf file
+            try (var out = new FileOutputStream(getTemporaryDirectory() + "/" + fileName + "_IMAGE_BBOX.pdf")) {
+                out.write(pdfDoc.save(SDFDoc.SaveMode.LINEARIZED, null));
+            }
+            pdfDoc.close();
+            System.out.println("result file: " + getTemporaryDirectory() + "/" + fileName + "_IMAGE_BBOX.pdf");
+            // round all coords to nearest int to account for inconsistencies with the calculation of the bounding box
+            return pageIdToRectCollection.values().stream().map(this::toRoundedCoordinateArrayList).flatMap(List::stream).collect(Collectors.toList());
        }
-        System.out.println("result file: " + getTemporaryDirectory() + "/" + fileName + "_IMAGE_BBOX.pdf");
-        fileStream.close();
-        // round all coords to nearest int to account for inconsistencies with the calculation of the bounding box
-        return pageIdToRectCollection.values().stream().map(this::toRoundedCoordinateArrayList).flatMap(List::stream).collect(Collectors.toList());
    }


@@ -178,59 +162,4 @@ class ImagePositionRetrievalServiceTest {
        return coords;
    }

-
-    @SneakyThrows
-    private void drawGrid(ElementWriter writer, Page page) {
-
-        ElementBuilder eb = new ElementBuilder();
-        double dX = 15;
-        double dY = 15;
-        int nRows = (int) (page.getPageHeight() / dY) + 1;
-        int nCols = (int) (page.getPageWidth() / dX) + 1;
-        for (int row = 0; row < nRows; ++row) {
-            for (int col = 0; col < nCols; ++col) {
-                Element cell = eb.createRect(col * dX, row * dY, dX, dY);
-                cell.setPathStroke(true);
-                cell.getGState().setLineWidth(1);
-                cell.getGState().setStrokeOpacity(0.1);
-                cell.getGState().setStrokeColorSpace(ColorSpace.createDeviceRGB());
-                if (row == 0 && col == 0) {
-                    cell.getGState().setStrokeColor(new ColorPt(0, 0, 1));
-                    cell.setPathFill(true);
-                    cell.getGState().setFillOpacity(0.8);
-                    cell.getGState().setFillColorSpace(ColorSpace.createDeviceRGB());
-                    cell.getGState().setFillColor(new ColorPt(0, 0, 1));
-                } else {
-                    cell.setPathFill(false);
-                    cell.getGState().setStrokeColor(new ColorPt(0.1, 0.1, 0.1));
-                }
-                writer.writePlacedElement(cell);
-            }
-        }
-    }
-
-
-    @SneakyThrows
-    public static void drawRectCollection(ElementWriter writer, RectCollection rectCollection, Page page) {
-
-        ColorPt colorPt = new ColorPt(1, 0, 0);
-        ElementBuilder eb = new ElementBuilder();
-        for (int i = 0; i < rectCollection.getNumRects(); ++i) {
-            Rect r = rectCollection.getRectAt(i);
-            Element rect = eb.createRect(r.getX1(), r.getY1(), r.getWidth(), r.getHeight());
-
-            rect.setPathStroke(true);
-            rect.getGState().setLineWidth(5);
-            rect.getGState().setStrokeColorSpace(ColorSpace.createDeviceRGB());
-            rect.getGState().setStrokeColor(colorPt);
-
-            rect.setPathFill(true);
-            rect.getGState().setFillColorSpace(ColorSpace.createDeviceRGB());
-            rect.getGState().setFillColor(colorPt);
-            rect.getGState().setFillOpacity(0.5);
-
-            writer.writePlacedElement(rect);
-        }
-    }
-
 }
@@ -1,41 +1,25 @@
 package com.iqser.red.service.ocr.v1.server.service;

 import static com.iqser.red.service.ocr.v1.server.utils.OsUtils.getTemporaryDirectory;
+import static com.iqser.red.service.ocr.v1.server.utils.PdfTextExtraction.extractAllTextFromDocument;
 import static org.assertj.core.api.AssertionsForClassTypes.assertThat;

 import java.io.FileInputStream;
 import java.io.FileOutputStream;

 import org.junit.jupiter.api.Test;
-import org.junit.jupiter.api.extension.ExtendWith;
-import org.springframework.amqp.rabbit.core.RabbitTemplate;
 import org.springframework.beans.factory.annotation.Autowired;
-import org.springframework.boot.test.context.SpringBootTest;
-import org.springframework.boot.test.mock.mockito.MockBean;
-import org.springframework.context.annotation.Import;
 import org.springframework.core.io.ClassPathResource;
-import org.springframework.test.context.junit.jupiter.SpringExtension;

-import com.iqser.red.service.ocr.v1.server.Application;
-import com.iqser.red.service.ocr.v1.server.OcrServiceIntegrationTest;
-import com.pdftron.pdf.PDFDoc;
-import com.pdftron.pdf.Page;
-import com.pdftron.pdf.PageIterator;
-import com.pdftron.pdf.TextExtractor;
+import com.iqser.red.service.ocr.v1.server.AbstractTest;

 import lombok.SneakyThrows;

-@ExtendWith(SpringExtension.class)
-@SpringBootTest(classes = Application.class, webEnvironment = SpringBootTest.WebEnvironment.RANDOM_PORT)
-@Import(OcrServiceIntegrationTest.TestConfiguration.class)
-public class InvisibleElementRemovalServiceTest {
+public class InvisibleElementRemovalServiceTest extends AbstractTest {

    @Autowired
    private InvisibleElementRemovalService invisibleElementRemovalService;

-    @MockBean
-    protected RabbitTemplate rabbitTemplate;
-

    @Test
    @SneakyThrows
@@ -46,27 +30,18 @@ public class InvisibleElementRemovalServiceTest {
        ClassPathResource pdfFileResource = new ClassPathResource("files/" + fileName + ".pdf");

        try (var initialFileStream = pdfFileResource.getInputStream(); var out = new FileOutputStream(getTemporaryDirectory() + "/" + fileName + ".pdf")) {
-            invisibleElementRemovalService.removeInvisibleElements(initialFileStream, out,false);
+            invisibleElementRemovalService.removeInvisibleElements(initialFileStream, out, false);
        }

        try (var initialFileStream = pdfFileResource.getInputStream(); var out = new FileOutputStream(getTemporaryDirectory() + "/" + fileName + "_delta.pdf")) {
-            invisibleElementRemovalService.removeInvisibleElements(initialFileStream, out,true);
+            invisibleElementRemovalService.removeInvisibleElements(initialFileStream, out, true);
        }

        System.out.println("Output File without invisible elements: files/" + fileName + ".pdf");
        System.out.println("Output Delta File without invisible elements: files/" + fileName + "_delta.pdf");
-        TextExtractor extractor = new TextExtractor();

-        PDFDoc pdfDoc;
-        try(var fileStream = new FileInputStream(getTemporaryDirectory() + "/" + fileName + ".pdf")) {
-            pdfDoc = new PDFDoc(fileStream);
-        }
-
-        PageIterator iterator = pdfDoc.getPageIterator();
-        while (iterator.hasNext()) {
-            Page page = iterator.next();
-            extractor.begin(page);
-            String[] text = extractor.getAsText().split("\n");
+        try (var fileStream = new FileInputStream(getTemporaryDirectory() + "/" + fileName + ".pdf")) {
+            String[] text = extractAllTextFromDocument(fileStream).split("\n");
            assertThat(text).contains("APPENDIX 16 Pathology Report", "Amendment 1", "Page 255 of 260");
        }
    }
@@ -0,0 +1,74 @@
+package com.iqser.red.service.ocr.v1.server.utils;
+
+import com.pdftron.pdf.ColorPt;
+import com.pdftron.pdf.ColorSpace;
+import com.pdftron.pdf.Element;
+import com.pdftron.pdf.ElementBuilder;
+import com.pdftron.pdf.ElementWriter;
+import com.pdftron.pdf.Page;
+import com.pdftron.pdf.Rect;
+import com.pdftron.pdf.RectCollection;
+
+import lombok.SneakyThrows;
+
+
+public class PdfDraw {
+
+    @SneakyThrows
+    public static void drawGrid(ElementWriter writer, Page page) {
+
+        ElementBuilder eb = new ElementBuilder();
+        double dX = 15;
+        double dY = 15;
+        int nRows = (int) (page.getPageHeight() / dY) + 1;
+        int nCols = (int) (page.getPageWidth() / dX) + 1;
+        for (int row = 0; row < nRows; ++row) {
+            for (int col = 0; col < nCols; ++col) {
+                Element cell = eb.createRect(col * dX, row * dY, dX, dY);
+                cell.setPathStroke(true);
+                cell.getGState().setLineWidth(1);
+                cell.getGState().setStrokeOpacity(0.1);
+                cell.getGState().setStrokeColorSpace(ColorSpace.createDeviceRGB());
+                if (row == 0 && col == 0) {
+                    cell.getGState().setStrokeColor(new ColorPt(0, 0, 1));
+                    cell.setPathFill(true);
+                    cell.getGState().setFillOpacity(0.8);
+                    cell.getGState().setFillColorSpace(ColorSpace.createDeviceRGB());
+                    cell.getGState().setFillColor(new ColorPt(0, 0, 1));
+                } else {
+                    cell.setPathFill(false);
+                    cell.getGState().setStrokeColor(new ColorPt(0.1, 0.1, 0.1));
+                }
+                writer.writePlacedElement(cell);
+            }
+        }
+        eb.destroy();
+    }
+
+
+    @SneakyThrows
+    public static void drawRectCollection(ElementWriter writer, RectCollection rectCollection) {
+
+        ColorPt colorPt = new ColorPt(1, 0, 0);
+        ElementBuilder eb = new ElementBuilder();
+        for (int i = 0; i < rectCollection.getNumRects(); ++i) {
+            Rect r = rectCollection.getRectAt(i);
+            Element rect = eb.createRect(r.getX1(), r.getY1(), r.getWidth(), r.getHeight());
+
+            rect.setPathStroke(true);
+            rect.getGState().setLineWidth(5);
+            rect.getGState().setStrokeColorSpace(ColorSpace.createDeviceRGB());
+            rect.getGState().setStrokeColor(colorPt);
+
+            rect.setPathFill(true);
+            rect.getGState().setFillColorSpace(ColorSpace.createDeviceRGB());
+            rect.getGState().setFillColor(colorPt);
+            rect.getGState().setFillOpacity(0.5);
+
+            writer.writePlacedElement(rect);
+        }
+        colorPt.destroy();
+        eb.destroy();
+    }
+
+}
@@ -0,0 +1,35 @@
+package com.iqser.red.service.ocr.v1.server.utils;
+
+import java.io.IOException;
+import java.io.InputStream;
+import java.util.ArrayList;
+import java.util.List;
+
+import com.pdftron.common.PDFNetException;
+import com.pdftron.pdf.PDFDoc;
+import com.pdftron.pdf.Page;
+import com.pdftron.pdf.PageIterator;
+import com.pdftron.pdf.TextExtractor;
+
+
+public class PdfTextExtraction {
+
+    public static String extractAllTextFromDocument(InputStream fileStream) throws IOException, PDFNetException {
+
+        PDFDoc pdfDoc = new PDFDoc(fileStream);
+        TextExtractor extractor = new TextExtractor();
+        List<String> texts = new ArrayList<>();
+
+        PageIterator iterator = pdfDoc.getPageIterator();
+        while (iterator.hasNext()) {
+            Page page = iterator.next();
+            extractor.begin(page);
+            texts.add(extractor.getAsText());
+        }
+
+        extractor.destroy();
+        pdfDoc.close();
+        return String.join("\n", texts);
+    }
+
+}
@@ -7,7 +7,7 @@
    <parent>
        <groupId>com.iqser.red</groupId>
        <artifactId>platform-dependency</artifactId>
-        <version>1.16.0</version>
+        <version>1.17.0</version>
        <relativePath/>
    </parent>

@@ -27,7 +27,7 @@
            <dependency>
                <groupId>com.iqser.red</groupId>
                <artifactId>platform-commons-dependency</artifactId>
-                <version>1.20.0</version>
+                <version>1.21.0</version>
                <scope>import</scope>
                <type>pom</type>
            </dependency>
Author	SHA1	Message	Date
deiflaender	19f12f0016	RED-5567: Upgraded to lastest pdftron version	2023-05-26 09:20:05 +02:00
Kilian Schuettler	d91fb737cb	Pull request #14 : RED-6321: OCR not working correctly with 3.6.0 Merge in RED/ocr-service from RED-6321 to master * commit '2d503c74a6fba0b02ac4c92af82a493165e45761': RED-6321: OCR not working correctly with 3.6.0 added end() statements to formWriters to write their changes to the PDF contentStream moved replaceOriginalPageWithOcrPage outside the main OCR Loop since it caused exponential RAM Usage in some cases	2023-03-06 10:16:02 +01:00
Kilian Schuettler	2d503c74a6	RED-6321: OCR not working correctly with 3.6.0 added end() statements to formWriters to write their changes to the PDF contentStream moved replaceOriginalPageWithOcrPage outside the main OCR Loop since it caused exponential RAM Usage in some cases	2023-03-06 10:08:32 +01:00
Corina Olariu	bdcba7cb6d	Pull request #11 : RED-4988 Check jacoco version in poms and update to a current compatible version Merge in RED/ocr-service from RED-4988 to master * commit '334409e4db326945a1eeb86004d79e0b0faeb99c': RED-4988 Check jacoco version in poms and update to a current compatible version - update platform-dependency, platform-commons-dependency	2023-03-02 09:28:11 +01:00
Shamel Hussain	23622d4a85	Pull request #13 : RED-5718: Update base image version to allow using random user ids Merge in RED/ocr-service from shussain/Dockerfile-1677589647499 to master * commit '4266184456ea4d97fa19b52ff37d5d9fc09dd170': RED-5718: Update base image version to allow using random user ids	2023-02-28 14:56:37 +01:00
Shamel Hussain	4266184456	RED-5718: Update base image version to allow using random user ids	2023-02-28 14:07:44 +01:00
Kilian Schuettler	a4ca2db37d	Pull request #12 : RED-6280: Performance Test Issue with OCR-Service Merge in RED/ocr-service from RED-6280 to master * commit '742725834933ad74ad582366b2b62015524bedb3': RED-6280: Performance Test Issue with OCR-Service removed init/terminate calls again manual memory cleanup at every opportunity	2023-02-28 10:30:58 +01:00
Kilian Schuettler	7427258349	RED-6280: Performance Test Issue with OCR-Service removed init/terminate calls again manual memory cleanup at every opportunity	2023-02-28 10:21:21 +01:00
devplant	334409e4db	RED-4988 Check jacoco version in poms and update to a current compatible version - update platform-dependency, platform-commons-dependency	2023-02-27 17:41:15 +02:00
Kilian Schuettler	bc661b7ea4	Pull request #9 : RED-6126: In the OCRService, OCR Text is not applied to Document Merge in RED/ocr-service from RED-6126 to master * commit 'caff5580dda644451433921a6a66dafe1cfa5dca': RED-6126: In the OCRService, OCR Text is not applied to Document refactored Tests with inheritance called PDFNet init/terminate in tests don't call init on startup RED-6126: In the OCRService, OCR Text is not applied to Document called PDFNet.initialize and terminate before and after message receive updated comments renamed some variables	2023-02-22 13:33:49 +01:00
Kilian Schuettler	caff5580dd	RED-6126: In the OCRService, OCR Text is not applied to Document refactored Tests with inheritance called PDFNet init/terminate in tests *don't call init on startup	2023-02-22 11:58:07 +01:00
Kilian Schuettler	430ad45a67	RED-6126: In the OCRService, OCR Text is not applied to Document called PDFNet.initialize and terminate before and after message receive updated comments *renamed some variables	2023-02-22 10:32:55 +01:00