IndexOutOfBoundsException
clorenz opened this issue · 1 comments
clorenz commented
The highlighter runs (probably in a very special case) in an IndexOutOfBoundsException, unfortunately, no further info (like the document ID and the minocr file name) is provided:
null:java.lang.IndexOutOfBoundsException: 6131276
at java.base/java.nio.DirectByteBuffer.get(DirectByteBuffer.java:270)
at org.mdz.search.solrocr.util.FileBytesCharIterator.adjustOffset(FileBytesCharIterator.java:81)
at org.mdz.search.solrocr.util.FileBytesCharIterator.subSequence(FileBytesCharIterator.java:136)
at org.mdz.search.solrocr.formats.OcrPassageFormatter.format(OcrPassageFormatter.java:41)
at org.mdz.search.solrocr.lucene.OcrFieldHighlighter.highlightFieldForDoc(OcrFieldHighlighter.java:74)
at org.mdz.search.solrocr.lucene.OcrHighlighter.highlightOcrFields(OcrHighlighter.java:195)
at org.mdz.search.solrocr.solr.SolrOcrHighlighter.doHighlighting(SolrOcrHighlighter.java:82)
at org.mdz.search.solrocr.solr.HighlightComponent.process(HighlightComponent.java:95)
at org.apache.solr.handler.component.SearchHandler.handleRequestBody(SearchHandler.java:298)
jbaiter commented
Closing this for now, since the issue didn't reappear after the latest updates. We now also provide more context in the case of OOB reads, so if the issue should re-appear we will have more information to work with.