As already discussed on GitHub (https://github.com/hibernate/hibernate-search/pull/1634)
While updating to a more recent version of Tika (1.17), we cannot rely on the default AutoDetectParser anymore, as we need to define a custom Tika instance with a special tika-config.xml to drop a specific parser (GrobIdParser, https://wiki.apache.org/tika/GrobidJournalParser) which, at least after my lib update, is seems now
the default for indexing PDFs. This is effectively suggested in
https://issues.apache.org/jira/browse/TIKA-2243
In my case, I had to provide a custom tika-config.xml:
{code:xml}
<?xml version="1.0" encoding="UTF-8"?>
<properties>
<parsers>
<parser class="org.apache.tika.parser.DefaultParser">
<mime-exclude>application/pdf</mime-exclude>
<parser-exclude class="org.apache.tika.parser.journal.JournalParser"/>
</parser>
<parser class="org.apache.tika.parser.pdf.PDFParser">
<mime>application/pdf</mime>
</parser>
</parsers>
</properties>
{code}
To reflect this customization in the TikaBridge as well, some additional handle for this is required as the AutoDetectParser is hard-wired in TikaBridge.
|
|
