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

* treat TablePageBlock.OTHER like PARAGRAPH (no special treatment)
This commit is contained in:
Kilian Schuettler 2024-04-17 17:40:21 +02:00
parent c4d9c5df02
commit f256f9b30f

View File

@ -44,7 +44,6 @@ public class SectionsBuilderService {
for (ClassificationPage page : document.getPages()) {
List<TextPageBlock> header = new ArrayList<>();
List<TextPageBlock> footer = new ArrayList<>();
List<TextPageBlock> unclassifiedText = new ArrayList<>();
for (AbstractPageBlock current : page.getTextBlocks()) {
if (current.getClassification() == null) {
@ -63,11 +62,6 @@ public class SectionsBuilderService {
continue;
}
if (current.getClassification().equals(PageBlockType.OTHER)) {
unclassifiedText.add((TextPageBlock) current);
continue;
}
if (prev != null && current.getClassification().isHeadline() && !prev.getClassification().isHeadline() || !document.isHeadlines()) {
ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline);
chunkBlock.setHeadline(lastHeadline);
@ -95,9 +89,6 @@ public class SectionsBuilderService {
if (!footer.isEmpty()) {
footers.add(new ClassificationFooter(footer));
}
if (!unclassifiedText.isEmpty()) {
unclassifiedTexts.add(new UnclassifiedText(unclassifiedText));
}
}
ClassificationSection chunkBlock = buildTextBlock(chunkWords, lastHeadline);