Marmelator/layout-parser
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 289 Wiki Activity

RED-8481: Use visual layout parsing to detect signatures

Browse Source 
working on failing tests
This commit is contained in:
yhampe 2024-02-15 10:16:07 +01:00
parent 903b1c1fd4
commit b4a225144d
4 changed files with 39 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/python_api/adapter/VisualLayoutParsingAdapter.java
View File

@ -35,12 +35,9 @@ public class VisualLayoutParsingAdapter {
public Map<Integer, List<ClassifiedImage>> buildExtractedSignaturesPerPage(VisualLayoutParsingResponse visualLayoutParsingResponse) {
Map<Integer, List<ClassifiedImage>> signatures = new HashMap<>();
if(visualLayoutParsingResponse.getData() != null ) {
visualLayoutParsingResponse.getData().forEach(tableData -> signatures.computeIfAbsent(tableData.getPage_idx(), tableCell -> new ArrayList<>()).addAll(convertSignatures(tableData.getPage_idx(), tableData.getBoxes())));
return signatures;
}
return null;
}

5
layoutparser-service/layoutparser-service-server/src/test/java/com/knecon/fforesight/service/layoutparser/server/VisualLayoutParsingServiceTests.java Normal file
View File

@ -0,0 +1,5 @@
package com.knecon.fforesight.service.layoutparser.server;
public class VisualLayoutParsingServiceTests {
}

27
layoutparser-service/layoutparser-service-server/src/test/java/com/knecon/fforesight/service/layoutparser/server/utils/AbstractTest.java
View File

@ -20,6 +20,7 @@ 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 org.xmlunit.builder.Input;
import com.iqser.red.commons.jackson.ObjectMapperFactory;
import com.iqser.red.storage.commons.service.StorageService;
@ -51,6 +52,8 @@ public abstract class AbstractTest {
private RabbitTemplate rabbitTemplate;
protected final static String ORIGIN_FILE_ID = "origin";
protected final static String VISUAL_LAYOUT_FILE = "visual";
protected final static String TABLE_FILE_ID = "table";
protected final static String IMAGE_FILE_ID = "image";
protected final static String STRUCTURE_FILE_ID = "structure";
@ -96,7 +99,7 @@ public abstract class AbstractTest {
@SneakyThrows
protected LayoutParsingRequest prepareStorage(String file) {
return prepareStorage(file, "cv_table_parsing_response/empty.json", "image_service_response/empty.json");
return prepareStorage(file, "cv_table_parsing_response/empty.json", "image_service_response/empty.json","visual_layout_parsing_response/empty.json");
}
@ -116,6 +119,7 @@ public abstract class AbstractTest {
.originFileStorageId(ORIGIN_FILE_ID)
.tablesFileStorageId(Optional.of(TABLE_FILE_ID))
.imagesFileStorageId(Optional.of(IMAGE_FILE_ID))
.visualLayoutParsingFileId(Optional.of(VISUAL_LAYOUT_FILE))
.structureFileStorageId(STRUCTURE_FILE_ID)
.textBlockFileStorageId(TEXT_FILE_ID)
.positionBlockFileStorageId(POSITION_FILE_ID)
@ -136,6 +140,16 @@ public abstract class AbstractTest {
return prepareStorage(pdfFileResource.getInputStream(), cvServiceResponseFileResource.getInputStream(), imageInfoFileResource.getInputStream());
}
@SneakyThrows
protected LayoutParsingRequest prepareStorage(String file, String cvServiceResponseFile, String imageInfoFile, String visualLayoutParsingResponseFile) {
ClassPathResource pdfFileResource = new ClassPathResource(file);
ClassPathResource cvServiceResponseFileResource = new ClassPathResource(cvServiceResponseFile);
ClassPathResource imageInfoFileResource = new ClassPathResource(imageInfoFile);
ClassPathResource visualLayoutParsingResponseResource = new ClassPathResource(visualLayoutParsingResponseFile);
return prepareStorage(pdfFileResource.getInputStream(), cvServiceResponseFileResource.getInputStream(), imageInfoFileResource.getInputStream(), visualLayoutParsingResponseResource.getInputStream());
}
@SneakyThrows
protected LayoutParsingRequest prepareStorage(InputStream fileStream, InputStream cvServiceResponseFileStream, InputStream imageInfoStream) {
@ -147,6 +161,17 @@ public abstract class AbstractTest {
return buildDefaultLayoutParsingRequest(LayoutParsingType.REDACT_MANAGER);
}
@SneakyThrows
protected LayoutParsingRequest prepareStorage(InputStream fileStream, InputStream cvServiceResponseFileStream, InputStream imageInfoStream, InputStream visualLayoutParsingResponseFileStream) {
storageService.storeObject(TenantContext.getTenantId(), IMAGE_FILE_ID, imageInfoStream);
storageService.storeObject(TenantContext.getTenantId(), TABLE_FILE_ID, cvServiceResponseFileStream);
storageService.storeObject(TenantContext.getTenantId(), ORIGIN_FILE_ID, fileStream);
storageService.storeObject(TenantContext.getTenantId(),VISUAL_LAYOUT_FILE,visualLayoutParsingResponseFileStream );
return buildDefaultLayoutParsingRequest(LayoutParsingType.REDACT_MANAGER);
}
@AfterEach
public void cleanupStorage() {

8
layoutparser-service/layoutparser-service-server/src/test/resources/visual_layout_parsing_response/empty.json Normal file
View File

@ -0,0 +1,8 @@
{
"dossierId": "123",
"fileId": "123",
"targetFileExtension": "ORIGIN.pdf.gz",
"responseFileExtension": "EXTRACTED_TABLES.json.gz",
"data": [],
"X_TENANT_ID": ""
}