Pull request #8: RED-6126: In the OCRService, OCR Text is not applied to Document

Merge in RED/ocr-service from RED-6126 to master

* commit '0bc4fea2a52c92efaaaf8cf93c2ae02766168a80':
  RED-6126: In the OCRService, OCR Text is not applied to Document *removed unnecessary getXObject() call, since it fails for inline_images
This commit is contained in:
Kilian Schuettler 2023-02-14 09:57:28 +01:00
commit a6d99f5916

View File

@ -15,7 +15,6 @@ import com.pdftron.common.Matrix2D;
import com.pdftron.common.PDFNetException;
import com.pdftron.pdf.Element;
import com.pdftron.pdf.ElementReader;
import com.pdftron.pdf.Image;
import com.pdftron.pdf.PDFDoc;
import com.pdftron.pdf.Page;
import com.pdftron.pdf.Rect;
@ -68,9 +67,8 @@ public class ImagePositionRetrievalService {
while ((element = reader.next()) != null) {
switch (element.getType()) {
case Element.e_image, Element.e_inline_image -> {
Image image = new Image(element.getXObject());
// see everyPointInDashedLineIsImage.pdf TestFile
if (image.getImageHeight() > PIXEL_THRESHOLD || image.getImageWidth() > PIXEL_THRESHOLD) {
if (element.getImageHeight() > PIXEL_THRESHOLD || element.getImageWidth() > PIXEL_THRESHOLD) {
imagePositions.addRect(toRotationAdjustedRect(element.getBBox(), currentPage, mirrorY));
}
}