CFIndex PDF files stored in the DB?

[Guest]

Hi all,

Is it possible to index PDF content stored as BLOB in the database, using the CF Search mechanism?

We are using the Solr engine right now, and for query based fulltext search it works like a charm, but I couldn't figure out how to do full text search for PDFs stored in the DB.

We don't have too many documents yet, under 1000, so I guess if all else fails, I *could* still write them out to a temp dir and do the indexing, and then do some custom linking to the DB code instead of the temp file, but that looks really dirty to me.

Any tips and advices will be appreciated!

Thanks in advance!

ZHU, Jia

[tclaremont]

I don't run into this problem because I don't store PDF documents in the databse. I just store the path in the database, and leave the document in a directory.