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

RED-8995: unclassified text might be missing from document data

Browse Source 
* treat TablePageBlock.OTHER like PARAGRAPH (no special treatment)
This commit is contained in:
Kilian Schuettler 2024-04-17 17:40:21 +02:00
parent 778bae0f7f
commit 2addf63baf
1 changed files with 0 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/SectionsBuilderService.java
View File

@ -43,7 +43,6 @@ public class SectionsBuilderService {
for (ClassificationPage page : document.getPages()) { for (ClassificationPage page : document.getPages()) {
List<TextPageBlock> header = new ArrayList<>(); List<TextPageBlock> header = new ArrayList<>();
List<TextPageBlock> footer = new ArrayList<>(); List<TextPageBlock> footer = new ArrayList<>();
List<TextPageBlock> unclassifiedText = new ArrayList<>();
for (AbstractPageBlock current : page.getTextBlocks()) { for (AbstractPageBlock current : page.getTextBlocks()) {
if (current.getClassification() == null) { if (current.getClassification() == null) {
@ -62,11 +61,6 @@ public class SectionsBuilderService {
continue; continue;
} }
if (current.getClassification().equals(PageBlockType.OTHER)) {
unclassifiedText.add((TextPageBlock) current);
continue;
}
if (prev != null && current.getClassification().isHeadline() && !prev.getClassification().isHeadline() || !document.isHeadlines()) { if (prev != null && current.getClassification().isHeadline() && !prev.getClassification().isHeadline() || !document.isHeadlines()) {
ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline); ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline);
chunkBlock.setHeadline(lastHeadline); chunkBlock.setHeadline(lastHeadline);
@ -94,9 +88,6 @@ public class SectionsBuilderService {
if (!footer.isEmpty()) { if (!footer.isEmpty()) {
footers.add(new ClassificationFooter(footer)); footers.add(new ClassificationFooter(footer));
} }
if (!unclassifiedText.isEmpty()) {
unclassifiedTexts.add(new UnclassifiedText(unclassifiedText));
}
} }
ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline); ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline);