solr-cell search works for some pdfs not others

I have been searching for two days and have not been able to find an answer.

I have solr installed from the repo开发者_JAVA技巧s on an Ubuntu server running on tomcat 6. I have added the solr-cell jar and tika libraries.

I can run a curl command that works for some pdf files and indexes them fine, but it does not not work for others. At first i thought that some files were corrupted but that does not appear to be be the case. There does not appear to me to be any major difference between the ones thaqt work and those that don't.

The error i get is a 500 error - see example here

The curl request i make is:

$ curl 'http://mysolrserver.com:port/solr/update/extract?map.content=text&map.stream_name=id&extractOnly=true&commit=true' -F "file=@/absolute/path/to/file.pdf"

This does work for some PDFs fine, just not others.

I believe I have solr 1.4.0 installed.

Any help would be appreciated - thank you

--EDIT-- I am using Ubuntu 10.04.1 if that helps at all.

A NullPointerException is probably a bug. Report it to PDFBox and/or Tika.

OK the nightly snapshot of solr uses PDFBox 1.3.1 as opposed to the current stable which uses 0.7.* which is a fair amount of revision changes.

I can index all the pdfs using this snapshot version of solr. This seems to me something that will be fixed in the next stable version.