On Fri, 10 Jun 2011, Andrzej Bialecki wrote:
I have a feeling that .pfa and .pbf are the fonts themselves, and the
.pfm and .afm files are metadata about them. Can anyone confirm? If so,
we should split this entry into two
The files ending with "m" are font metrics.
Thanks for the info. I've split these out into
* application/x-font-adobe-metric (afm)
* application/x-font-printer-metric (pfm)
Does that look sane? I've added some mime magic detection too, based on
various files I found.
If anyone knows of any small, free sample files for any of these we could
use for testing with, that'd be great. I've knocked up a test .afm file,
as it's fairly simple, but we could use .pfm, .pfb and .pfa ones
Be aware that at least some of these files are already handled by
FontBox (part of the PDFBox).
Great. I was hoping to try using this today, but ran out of time just
getting the detection and mimetype stuff sorted :/
The relevant issue is TIKA-507 if anyone wants to jump in and start on a
FontBox parser for any of these formats!
Nick
