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

RED-9103: Fixed save of document viewer file

Browse Source
This commit is contained in:
Dominique Eifländer 2024-05-02 12:53:55 +02:00
parent 07da43f2d9
commit 49139ee603
1 changed files with 11 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/visualization/ViewerDocumentService.java
View File

@ -4,11 +4,9 @@ import java.awt.geom.AffineTransform;
import java.awt.geom.Rectangle2D;
import java.io.IOException;
import java.io.OutputStream;
import java.util.HashSet;
import java.util.Set;
import org.apache.pdfbox.cos.COSDictionary;
import org.apache.pdfbox.cos.COSName;
import org.apache.pdfbox.pdfwriter.compress.CompressParameters;
import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.pdmodel.PDDocumentCatalog;
import org.apache.pdfbox.pdmodel.PDPage;
@ -40,7 +38,6 @@ import lombok.extern.slf4j.Slf4j;
@RequiredArgsConstructor
public class ViewerDocumentService {
private static final String LAYER_NAME = "Layout grid";
private static final int FONT_SIZE = 10;
public static final float LINE_WIDTH = 1f;
@ -54,8 +51,7 @@ public class ViewerDocumentService {
LayoutGrid layoutGrid = layoutGridService.createLayoutGrid(document);
// PDDocument.save() is very slow, since it actually traverses the entire pdf and writes a new one.
// If we collect all COSDictionaries we changed and tell it explicitly to only add the changed ones by using saveIncremental it's very fast.
Set<COSDictionary> dictionariesToUpdate = new HashSet<>();
PDOptionalContentGroup layer = addLayerToDocument(pdDocument, dictionariesToUpdate, layerVisibilityDefaultValue);
PDOptionalContentGroup layer = addLayerToDocument(pdDocument, layerVisibilityDefaultValue);
PDFont font = new PDType1Font(Standard14Fonts.FontName.HELVETICA);
for (int pageNumber = 0; pageNumber < pdDocument.getNumberOfPages(); pageNumber++) {
@ -68,7 +64,8 @@ public class ViewerDocumentService {
// e.g. not escaped matrix transformations.
escapePreviousContents(pdDocument, pdPage);
VisualizationsOnPage visualizationsOnPage = layoutGrid.getVisualizationsPerPages().get(pageNumber);
VisualizationsOnPage visualizationsOnPage = layoutGrid.getVisualizationsPerPages()
.get(pageNumber);
assert pageNumber == visualizationsOnPage.getPageNumber();
// We need to append to the content stream, otherwise the content could be overlapped by following content.
try (var contentStream = new PDPageContentStream(pdDocument, pdPage, PDPageContentStream.AppendMode.APPEND, true)) {
@ -102,11 +99,11 @@ public class ViewerDocumentService {
contentStream.setFont(font, FONT_SIZE);
contentStream.beginText();
Matrix textMatrix = new Matrix((float) textDeRotationMatrix.getScaleX(),
(float) textDeRotationMatrix.getShearX(),
(float) textDeRotationMatrix.getShearY(),
(float) textDeRotationMatrix.getScaleY(),
(float) placedText.lineStart().getX(),
(float) placedText.lineStart().getY());
(float) textDeRotationMatrix.getShearX(),
(float) textDeRotationMatrix.getShearY(),
(float) textDeRotationMatrix.getScaleY(),
(float) placedText.lineStart().getX(),
(float) placedText.lineStart().getY());
textMatrix.translate(-((font.getStringWidth(placedText.text()) / 1000) * FONT_SIZE + (2 * LINE_WIDTH) + 4), -FONT_SIZE);
contentStream.setTextMatrix(textMatrix);
contentStream.showText(placedText.text());
@ -115,12 +112,9 @@ public class ViewerDocumentService {
contentStream.restoreGraphicsState();
contentStream.endMarkedContent();
}
dictionariesToUpdate.add(pdPage.getCOSObject());
dictionariesToUpdate.add(pdPage.getResources().getCOSObject());
}
dictionariesToUpdate.add(pdDocument.getDocumentInformation().getCOSObject());
// dictionariesToUpdate.add(pdDocument.getDocument().getTrailer());
pdDocument.saveIncremental(outputStream, dictionariesToUpdate);
pdDocument.save(outputStream, CompressParameters.NO_COMPRESSION);
}
@ -145,7 +139,7 @@ public class ViewerDocumentService {
}
private static PDOptionalContentGroup addLayerToDocument(PDDocument pdDocument, Set<COSDictionary> dictionariesToUpdate, boolean layerVisibilityDefaultValue) {
private static PDOptionalContentGroup addLayerToDocument(PDDocument pdDocument, boolean layerVisibilityDefaultValue) {
PDDocumentCatalog catalog = pdDocument.getDocumentCatalog();
PDOptionalContentProperties ocprops = catalog.getOCProperties();
@ -161,7 +155,6 @@ public class ViewerDocumentService {
ocprops.addGroup(layer);
}
ocprops.setGroupEnabled(layer, layerVisibilityDefaultValue);
dictionariesToUpdate.add(catalog.getCOSObject());
return layer;
}