Compare commits

...
6 Commits
Author SHA1 Message Date
Corina Olariu 546341ee75 Merge branch 'RED-9177' into 'main'
RED-9177 - Layout parser fails to process file

See merge request fforesight/layout-parser!158
2024-05-22 13:26:10 +02:00
Corina Olariu 0ed1481517 RED-9177 - Layout parser fails to process file
- use originFile as viewerDocumentFile
- return layoutGridOCGName in case the name is found and not check further properties
2024-05-22 13:02:42 +03:00
Andrei Isvoran b2a47f66ae Merge branch 'RED-9149-header' into 'main'
RED-9149 - Remove header detection

See merge request fforesight/layout-parser!157
2024-05-20 14:12:04 +02:00
Andrei Isvoran 3835d03036 RED-9149 - Remove header detection 2024-05-20 14:59:34 +03:00
Dominique Eifländer b867deb9f9 Merge branch 'CLARI-hotfix' into 'main'
hotifx for clarifynd

See merge request fforesight/layout-parser!154
2024-05-15 14:08:07 +02:00
Kilian Schuettler 8648ed0952 hotifx for clarifynd 2024-05-15 14:02:02 +02:00
4 changed files with 14 additions and 11 deletions
@@ -105,16 +105,21 @@ public class LayoutParsingPipeline {
log.info("Starting layout parsing for {}", layoutParsingRequest.identifier());
File originFile = layoutParsingStorageService.getOriginFile(layoutParsingRequest.originFileStorageId());
File viewerDocumentFile = layoutParsingStorageService.getViewerDocFile(layoutParsingRequest.viewerDocumentStorageId()).orElse(originFile);
// File viewerDocumentFile = layoutParsingStorageService.getViewerDocFile(layoutParsingRequest.viewerDocumentStorageId()).orElse(originFile);
File viewerDocumentFile = originFile;
VisualLayoutParsingResponse visualLayoutParsingResponse = new VisualLayoutParsingResponse();
if (layoutParsingRequest.visualLayoutParsingFileId().isPresent()) {
visualLayoutParsingResponse = layoutParsingStorageService.getVisualLayoutParsingFile(layoutParsingRequest.visualLayoutParsingFileId().get());
if (layoutParsingRequest.visualLayoutParsingFileId()
.isPresent()) {
visualLayoutParsingResponse = layoutParsingStorageService.getVisualLayoutParsingFile(layoutParsingRequest.visualLayoutParsingFileId()
.get());
}
ImageServiceResponse imageServiceResponse = new ImageServiceResponse();
if (layoutParsingRequest.imagesFileStorageId().isPresent()) {
imageServiceResponse = layoutParsingStorageService.getImagesFile(layoutParsingRequest.imagesFileStorageId().get());
if (layoutParsingRequest.imagesFileStorageId()
.isPresent()) {
imageServiceResponse = layoutParsingStorageService.getImagesFile(layoutParsingRequest.imagesFileStorageId()
.get());
}
TableServiceResponse tableServiceResponse = new TableServiceResponse();
@@ -63,7 +63,7 @@ public class DocstrumBlockificationService {
}
if (layoutParsingType == LayoutParsingType.CLARIFYND) {
mergeIntersectingBlocks(classificationPage, usedRulings, 0, 6.5f);
mergeIntersectingBlocks(classificationPage, usedRulings, 0, 0);
}
return classificationPage;
@@ -71,8 +71,7 @@ public class DocuMineClassificationService {
== null
|| textBlock.getHighestFontSize()
<= document.getFontSizeCounter()
.getMostPopular()))
|| HeaderFooterDetection.isLikelyHeader(textBlock, document, page)) {
.getMostPopular()))) {
textBlock.setClassification(PageBlockType.HEADER);
} else if (MarkedContentUtils.intersects(textBlock, page.getMarkedContentBboxPerType(), MarkedContentUtils.FOOTER)
@@ -52,17 +52,16 @@ public class ContentStreamUtility {
public static Optional<COSName> findLayoutGridOCGName(PDPage page) {
Optional<COSName> layoutGridOCGName = Optional.empty();
var resourceIterator = page.getResources().getPropertiesNames();
for (COSName cosName : resourceIterator) {
COSBase cosBase = page.getResources().getProperties(cosName).getCOSObject().getDictionaryObject(COSName.NAME);
if (cosBase instanceof COSString string) {
if (ContentStreams.KNECON_LAYOUT.name().equals(string.getString())) {
layoutGridOCGName = Optional.of(cosName);
return Optional.of(cosName);
}
}
}
return layoutGridOCGName;
return Optional.empty();
}