From af993e8e3b3c6c656456e4fd6107eb84d81baae1 Mon Sep 17 00:00:00 2001 From: Kilian Schuettler Date: Mon, 17 Jul 2023 12:53:40 +0200 Subject: [PATCH] RED-7012: OCR error occurs for 512 pages BASF document * write all pathElements whose BBox is null --- .../pdftronlogic/commons/InvisibleElementRemovalService.java | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/src/main/java/com/iqser/red/pdftronlogic/commons/InvisibleElementRemovalService.java b/src/main/java/com/iqser/red/pdftronlogic/commons/InvisibleElementRemovalService.java index d9b7057..8841398 100644 --- a/src/main/java/com/iqser/red/pdftronlogic/commons/InvisibleElementRemovalService.java +++ b/src/main/java/com/iqser/red/pdftronlogic/commons/InvisibleElementRemovalService.java @@ -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(); }