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

RED-8932 Fixed not merged headline with identifier

Browse Source
This commit is contained in:
Dominique Eifländer 2024-04-24 11:38:26 +02:00
parent 59745a916c
commit 9eaecdf378
2 changed files with 2 additions and 2 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/services/blockification/DocuMineBlockificationService.java
View File

@ -61,8 +61,8 @@ public class DocuMineBlockificationService {
boolean newLineAfterSplit = prev != null && word.getMinYDirAdj() != prev.getMinYDirAdj() && wasSplitted && splitX1 != word.getMinXDirAdj(); boolean newLineAfterSplit = prev != null && word.getMinYDirAdj() != prev.getMinYDirAdj() && wasSplitted && splitX1 != word.getMinXDirAdj();
boolean isSplitByRuling = isSplitByRuling(minX, minY, maxX, maxY, word, horizontalRulingLines, verticalRulingLines); boolean isSplitByRuling = isSplitByRuling(minX, minY, maxX, maxY, word, horizontalRulingLines, verticalRulingLines);
boolean splitByDir = prev != null && !prev.getDir().equals(word.getDir()); boolean splitByDir = prev != null && !prev.getDir().equals(word.getDir());
boolean splitByOtherFontAndOtherY = prev != null && prev.getMaxYDirAdj() != word.getMaxYDirAdj() && (word.getFontStyle().contains("bold") && !prev.getFontStyle() boolean splitByOtherFontAndOtherY = prev != null && Math.abs(prev.getMaxYDirAdj() - word.getMaxYDirAdj()) > word.getTextHeight() * 0.2 && (word.getFontStyle()
.contains("bold") || prev.getFontStyle().contains("bold") && !word.getFontStyle().contains("bold")); .contains("bold") && !prev.getFontStyle().contains("bold") || prev.getFontStyle().contains("bold") && !word.getFontStyle().contains("bold"));
Matcher matcher = pattern.matcher(chunkWords.stream().collect(Collectors.joining(" ")).toString()); Matcher matcher = pattern.matcher(chunkWords.stream().collect(Collectors.joining(" ")).toString());
boolean startsOnSameX = Math.abs(minX - word.getMinXDirAdj()) < 5 && matcher.matches(); boolean startsOnSameX = Math.abs(minX - word.getMinXDirAdj()) < 5 && matcher.matches();

BIN
layoutparser-service/layoutparser-service-server/src/test/resources/files/SinglePages/DocumineIdentifierProblem.pdf Normal file
View File

Binary file not shown.