Remove StructParent from annotation that are artifacts

LonelyMidoriya · MaximPlusov · commit 4a75d5a2e878 · 2026-05-06T17:16:43.000+03:00
diff --git a/java/opendataloader-pdf-core/src/main/java/org/opendataloader/pdf/processors/AutoTaggingProcessor.java b/java/opendataloader-pdf-core/src/main/java/org/opendataloader/pdf/processors/AutoTaggingProcessor.java
@@ -419,6 +419,12 @@ private static void processAnnotations(PDDocument document, COSDocument cosDocum
                 annotObj.setKey(ASAtom.STRUCT_PARENT, COSInteger.construct(structParentInt));
                 cosDocument.addChangedObject(annotObj);
                 pageChanged = true;
+            } else {
+                if (annotation.knownKey(ASAtom.STRUCT_PARENT)) {
+                    annotation.removeKey(ASAtom.STRUCT_PARENT);
+                    cosDocument.addChangedObject(annotObj);
+                    pageChanged = true;
+                }
             }
         }
         // Flush the Annots array (may be an indirect object separate from the page)

Original file line number	Diff line number	Diff line change
`@@ -419,6 +419,12 @@ private static void processAnnotations(PDDocument document, COSDocument cosDocum`
`419`	`419`	`annotObj.setKey(ASAtom.STRUCT_PARENT, COSInteger.construct(structParentInt));`
`420`	`420`	`cosDocument.addChangedObject(annotObj);`
`421`	`421`	`pageChanged = true;`
	`422`	`+ } else {`
	`423`	`+ if (annotation.knownKey(ASAtom.STRUCT_PARENT)) {`
	`424`	`+ annotation.removeKey(ASAtom.STRUCT_PARENT);`
	`425`	`+ cosDocument.addChangedObject(annotObj);`
	`426`	`+ pageChanged = true;`
	`427`	`+ }`
`422`	`428`	`}`
`423`	`429`	`}`
`424`	`430`	`// Flush the Annots array (may be an indirect object separate from the page)`