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

Merge branch 'TAAS-103-hotfix' into 'main'

Browse Source 
* added back in if statement

See merge request fforesight/layout-parser!82
This commit is contained in:
Dominique Eifländer 2023-11-16 12:48:48 +01:00
commit 144a9591a2
2 changed files with 2 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/LayoutParsingPipeline.java
View File

@ -244,9 +244,9 @@ public class LayoutParsingPipeline {
private void increaseDocumentStatistics(ClassificationPage classificationPage, ClassificationDocument document) { private void increaseDocumentStatistics(ClassificationPage classificationPage, ClassificationDocument document) {
// if (!classificationPage.isLandscape()) { if (!classificationPage.isLandscape()) {
document.getFontSizeCounter().addAll(classificationPage.getFontSizeCounter().getCountPerValue()); document.getFontSizeCounter().addAll(classificationPage.getFontSizeCounter().getCountPerValue());
// } }
document.getFontCounter().addAll(classificationPage.getFontCounter().getCountPerValue()); document.getFontCounter().addAll(classificationPage.getFontCounter().getCountPerValue());
document.getTextHeightCounter().addAll(classificationPage.getTextHeightCounter().getCountPerValue()); document.getTextHeightCounter().addAll(classificationPage.getTextHeightCounter().getCountPerValue());
document.getFontStyleCounter().addAll(classificationPage.getFontStyleCounter().getCountPerValue()); document.getFontStyleCounter().addAll(classificationPage.getFontStyleCounter().getCountPerValue());

5
layoutparser-service/layoutparser-service-processor/src/main/java/com/knecon/fforesight/service/layoutparser/processor/services/parsing/PDFLinesTextStripper.java
View File

@ -329,11 +329,6 @@ public class PDFLinesTextStripper extends PDFTextStripper {
.getXDirAdj() - (previous.getXDirAdj() + previous.getWidthDirAdj()) < maximumGapSize; .getXDirAdj() - (previous.getXDirAdj() + previous.getWidthDirAdj()) < maximumGapSize;
} }
// !(sublist.isEmpty() || sublist.size() == 1 && (sublist.get(0).getUnicode().equals(" ") || sublist.get(0).getUnicode().equals("\u00A0") || sublist.get(0)
// .getUnicode()
// .equals("\t")))
@Override @Override
public String getText(PDDocument doc) throws IOException { public String getText(PDDocument doc) throws IOException {