OK, we finally have a fully functional solution that is now working on both a 3.6 site and a 3.8 site, both of which have tons of binary files. You can read the full article here:
Is there any more up to date information on how to debug binary file searching/indexing, especially indexing of PDF files?
I'm using eZ Publish 4.2 and PDF files are not being indexing, but I can't get any error information. How can I do some debugging and find out what's going on?