Which of those
"several open source packages" includes a scriptable
(ie, command-line) tool to extract the embedded images from a PDF?
With the xpdf
program, which runs under X, and which IS open-source
(I've certainly never run a binary of it that I didn't compile
locally from source) you can use 'print to file' to convert a PDF, or
selected pages from a PDF, to a postscript file.
Did you not see where I said "scriptable"? "Print to file" from a
GUI
interface (like xpdf) doesn't count.
You also appear to have missed the part where I want to *extract* the
image, not convert it to yet another wrapper format (such as
PostScript).
There *is* pdfimages, but that's not what you appeared to be talking
about. (I also don't actually know whether it works, though I would
tend to assume it does - vide infra.)
There is also a pdf2ps utility bundled with
ghostscript that does the
conversion.
...from PDF to PostScript, which is not what I wanted.
Then you can use all the nice free tools to mess with
the postscript
(or dig into it with vi or emacs).
You clearly have never tried to do anything useful with the PostScript
resulting from pdf2ps; it is not suitable for doing anything with at
finer than page granualrity.
I don't know what kind of stuff xpdf generates; I haven't yet got
around to building any of that. (I was made aware of it only recently,
and even more recently did I get hints on how to build it without all
the GUI goop.)
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B