re: [htdig] interpreting pdf Files

[Tim C. <cl...@de...>]

Doug:

My understanding is that the command line magic that you need to pass to
pdftotext directly is a bit too cumbersome to fit into htdig.conf.  Also,
from a memory standpoint, my experience with this was it was better for the
pdftotext program to write to a text/html temp file and then have htdig read
off of this vs. an "in-line" conversion and passthrough which caused
repeated failures for me.
If you would just like to try something different than doc2html.ph, I can
recommend a script which was written by Stefan Nehlsen (I think, I believe
that Martin Allert gave me a copy which he may well have made a lot of
changes to) called parsepdf.pl which I have found effective.  I think that
it is in the contributed works section although I should probably make sure
that is true.
Good Luck,
Tim