This may not be specifically geared towards beginners, but it
shouldn't be too hard a task to throw me some ideas or code snippets, so here
goes:
Adobe has a new application and Palm-side reader which allows me to
"convert" and install PDF files onto my Palm device. This is a Windows-only
tool, and is required to "convert" these files for use on the Palm.
I did some digging, and found out that they're simply taking the pdf
file you specify, converting it to PDF version 1.4, and splitting it into 64k
records (64k is the maximum record size a Palm can handle). They prepend a
header of 89 bytes to the front of the file (to satisfy the Palm at install
time), and tack on a .pdb extension to it (so foo.pdf becomes foo.pdf.pdb),
and send it to the Palm.
Pulling one of these off of the Palm with pilot-xfer (linux tools I
maintain) and renaming it back to .pdf (stripping the .pdb extension) allows
me to read it with xpdf and acroread. This is a clear sign that the whole file
is still intact (and actually, the file looks identical to the original file
in both readers, except images appear to be quantized down to 4bpp). Acroread
can deal with "corrupt" files, by ignoring the junk it can't dither, and I
assume this 89-byte header is this junk.
My question is: How do I read in this original pdf, and at 64k
boundaries, "chunk" the file into records that I can then pass back to the
Palm itself? Likewise, in perl, how do I output these 89 bytes in the header?
I can cut/paste them from the file in a hex editor or robust editor, and stick
them at the top of the new pdf, but I can't seem to figure out a way to do
this with perl at the moment.
I have the PDF and PDF::Create modules installed, but they don't seem
to document the format well enough for me to grok.
I'm SOOOOOO close...
Thanks.
/d
