On 6/11/2016 2:15 AM, Ross Moore wrote:
Hi Karl,
On Jun 11, 2016, at 8:57 AM, Karl Berry
mailto:karl@freefriends.org> wrote: Hi Ross - I don't have the DavidCLM-Medium font, and had never heard of the culmus package; I see now it is a pretty big effort on sf.
At any rate, I (or Thanh or whomever) need an example demonstrating the failure that can be run. Maybe you can make a self-contained set of files (the font(s) in needed formats, the tex source, etc.)?
Or: it doesn't matter if there is a workaround for it, as long as it shows the failure. So if you have an example that uses fonts already in TL, so much the better.
Certainly. The texlive-de.tex documentation uses a font that is affected this way.
Attached is a cut-down example source, test-berafont.tex with PDFs: test-berafont-pdftex.pdf — PDF/A using pdfTeX and test-berafont-gs.pdf — plain PDF using dvips+GS
You can see the difference in the attached PDFs:
SCI:TL-DE16 ross$ grep -a CharSet test-berafont-gs.pdf /CharSet(/A/E/L/T/V/X/a/aacute/acute/e/g/i/l/n/o/one/r/s/six/t/two/u/v/z/zero)/FontFile3 12 0 R>> SCI:TL-DE16 ross$ mv test-berafont.pdf test-berafont-pdftex.pdf SCI:TL-DE16 ross$ grep -a CharSet test-berafont-pdftex.pdf /CharSet (/A/E/L/T/V/X/aacute/e/g/i/l/n/o/one/r/s/six/t/two/u/v/z/zero)
where pdfTeX just puts /aacute GS includes /a/aacute/acute .
so what does acrobat distiller do ... maybe the font assembles the aacute glyph from two (using subroutines) and gs sees that as a reason to also add those glyphs (if i remember right pdftex tries to merge these things subr thingies in order to make smaller files) .. or does the ps file have a reference to the acute? if so, then this is imo a harmless difference and one cannot expect all postprocessors to behave the same Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------