Marmelator/redaction-service
1
0
Fork 0
Code Issues Pull Requests 14 Actions Packages Projects Releases 622 Wiki Activity

RSS-123: Improved sorting algorithm for TextPositionSequences

Browse Source
This commit is contained in:
deiflaender 2022-10-27 12:08:25 +02:00
parent ec8e21d904
commit 8e74615a1e
2 changed files with 42 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/model/SearchableText.java
View File

@ -6,6 +6,7 @@ import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSeque
import com.iqser.red.service.redaction.v1.server.redaction.utils.IdBuilder;
import com.iqser.red.service.redaction.v1.server.redaction.utils.SeparatorUtils;
import com.iqser.red.service.redaction.v1.server.redaction.utils.TextNormalizationUtilities;
import com.iqser.red.service.redaction.v1.server.redaction.utils.TextPositionSequenceComparator;
import java.util.ArrayList;
import java.util.Collections;
@ -217,8 +218,7 @@ public class SearchableText {
public String getAsStringWithLinebreaksSorted(List<TextPositionSequence> sequences) {
var sorted = sequences.stream()
.sorted(Comparator.comparing(a -> a.getTextPositions().get(0).getXDirAdj()))
.sorted(Comparator.comparing(a -> a.getTextPositions().get(0).getYDirAdj()))
.sorted(new TextPositionSequenceComparator())
.sorted(Comparator.comparing(a -> a.getPage()))
.collect(Collectors.toList());

40
redaction-service-v1/redaction-service-server-v1/src/main/java/com/iqser/red/service/redaction/v1/server/redaction/utils/TextPositionSequenceComparator.java Normal file
View File

@ -0,0 +1,40 @@
package com.iqser.red.service.redaction.v1.server.redaction.utils;
import java.util.Comparator;
import com.iqser.red.service.redaction.v1.server.parsing.model.TextPositionSequence;
public class TextPositionSequenceComparator implements Comparator<TextPositionSequence> {
@Override
public int compare(TextPositionSequence pos1, TextPositionSequence pos2) {
// only compare text that is in the same direction
int cmp1 = Float.compare(pos1.getDir().getDegrees(), pos2.getDir().getDegrees());
if (cmp1 != 0) {
return cmp1;
}
// get the text direction adjusted coordinates
float x1 = pos1.getMinXDirAdj();
float x2 = pos2.getMinXDirAdj();
float pos1YBottom = pos1.getMaxYDirAdj();
float pos2YBottom = pos2.getMaxYDirAdj();
// note that the coordinates have been adjusted so 0,0 is in upper left
float pos1YTop = pos1YBottom - pos1.getHeight();
float pos2YTop = pos2YBottom - pos2.getHeight();
float yDifference = Math.abs(pos1YBottom - pos2YBottom);
// we will do a simple tolerance comparison
if (yDifference < .1 || pos2YBottom >= pos1YTop && pos2YBottom <= pos1YBottom || pos1YBottom >= pos2YTop && pos1YBottom <= pos2YBottom) {
return Float.compare(x1, x2);
} else if (pos1YBottom < pos2YBottom) {
return -1;
} else {
return 1;
}
}
}