RED-7012: OCR error occurs for 512 pages BASF document

* write all pathElements whose BBox is null
This commit is contained in:
Kilian Schuettler 2023-07-17 12:53:40 +02:00
parent 5811cf76d4
commit af993e8e3b

View File

@ -254,7 +254,7 @@ public class InvisibleElementRemovalService {
PathData pathData = pathElement.getPathData();
if (pathData.getOperators().length == 0 && pathData.getPoints().length == 0 || pathElement.getBBox() == null) {
writer.writeGStateChanges(pathElement);
writer.writeElement(pathElement);
return;
}
@ -432,6 +432,7 @@ public class InvisibleElementRemovalService {
.stream()
.filter(elementFeatures -> elementFeatures.getElementType() == Element.e_path)
.map(elementFeatures -> (ElementFeatures.Path) elementFeatures)
.filter(elementFeatures -> !elementFeatures.getFillColor().equals(Color.WHITE))
.filter(element -> element.isBackground(textBBox))
.toList();
}