Baslinux mailing list

[sindi keesan <keesan AT sdf.lonestar.org>]

On Wed, 17 May 2006 3aoo-cvfd AT dea.spamcon.org wrote:

> sindi keesan wrote:
> > Have you got Slackware 4.0 xpdf working in BL3?
>
> Yes.  I needed to add the xfnts package and another
> package -- don't remember which -- maybe elflibs?
> ldd will tell you what's missing.
>
> Cheers,
> Steven

I did not need to add X fonts.  After some experimenting, I was able to 
edit xpdfrc to point both Times (serif) and Helvetica (sans-serif), 
proportionally spaced fonts, to a set of Helvetica fonts that was not 
originally used by xpdf (the one they use does not work as a substitute 
for Times, it lacks Palatine), which is smaller than the Times fonts 
because it is not decorated.  It is also easier to read onscreen.  I left 
bold and italic and bolditalic, and the monospaced Courier, and symbols 
and dingbats, and have 347K of fonts (rather than about 4MB of ghostscript 
fonts).  Pointing everything at Helvetica regular instead of italic, bold, 
or bolditalic would cut this to about 100K.  I don't foresee a use for 
Dingbats but I left it anyway to avoid causing problems, but maybe you can 
substitute Courier if Dingbat is monospaced.

You can delete /doc to save more space.

I have packaged these fonts, my xpdfrc, and the CLI programs, doc and man 
pages (in man format) into pdf-noX.tgz and posted about 2.5MB .tgz at
http://keesan.freeshell.org/blfiles/pdf3-noX.tgz

Unpack on / (or use pkg).  No install.sh or libraries.

I don't think you need to any anything else to use this package in BL3 or 
BL2.  Please test.

pdftoppm -r 300 -f 1 -l 1 -mono file.pdf file

This should produce one page of .pbm file (first and last pages = 1) in 
300 dpi.  I tested on some sheet music that had both serif and sans fonts 
in it (but not Courier, Dingbats, or italic or bold).  I don't have a way 
to produce pdf files to test with.  I think WORD should produce them.

If someone wants to convert the man pages to text I will substitute those.
I put them in /usr/man not their normal locations.

The binaries include pdftotext, pdfimages (extract only text or only 
images, the latter useful for scanned documents) and the slower pdftoppm 
which converts rather than extracts and needs the fonts even if there is 
no text in the page - it complains without them.  I have not tested any of 
these with 16MB RAM and pdftoppm may not work with that on large files.

pdfinfo and pdffonts give information on the pdf file and can be deleted.
pdftops can be deleted if you have no ps printer or ps viewer 
(ImageMagick).  ghostcript has a pdf2ps.

The six binaries are about 300K each except pdftoppm which is 600K.

Please let me know if this works for you.

Sindi Keesan