I am experiencing the ghostscript-esp bug reported here: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=381386 In short, all versions currently in portage omit accents and umlauts (and possibly other kinds of diacritics) on some, but not all, characters of some, but not all, fonts. (Which is kind of evil, since you might not notice that until too late -- almost happened to me with an application letter ... =8^O) Since a fix seems to be available according to the report mentioned, could you please bump ghostscript-esp to a version incorporating that fix? Switching to ghostscript-gpl isn't really an option for me, since it has other kinds of issues. (This one, for instance, just in case you should ask: http://forums1.itrc.hp.com/service/forums/parseCurl.do?CURL=/cm/QuestionAnswer/1,,0x514a1aca6930d611abd50090277a778c,00.html&admit=-682735245%201156166222016%2028353475 But that's another story...) Thank you!
does the patch truly help for you? I hae uploaded it at http://genstef.homelinux.org/08_revert_cjk.patch for easier trying
(In reply to comment #1) > does the patch truly help for you? > > I hae uploaded it at http://genstef.homelinux.org/08_revert_cjk.patch for > easier trying Thanks a lot, Stefan. I will try it and report back.
Ooops. No, you're right. The patch doesn't help. Like how could it, if it looks like it has already been applied. (I tried it against ghostscript-esp-8.15.2_p20060520.) Or rather, is it that the CJK-patch which is meant to be reverted is not even included in the gentoo sources!? Sorry, seems like I was too quick to assume that that patch would be the solution. But the symptoms are the same for me as in the debian bug report... Maybe there is another solution? I will be happy to test if you run across something. Thank you, anyway, so far.
Looks like we need the current-svn patch too: genstef.homelinux.org/00_r171.patch
Created attachment 94919 [details, diff] Stripped-down 00_r171.patch Thanks. I needed to make some changes to the latter patch so that it would apply. I took out hunks that were already applied and made some changes to unixinst.mak. I hope I haven't thrown out anything vital -- on my part, this is just (dilettantic) trial and (mostly) error. I'll also attach the version of that 08_revert_cjk.patch I used (I just cut off the topmost couple of lines) and a (trivial, though) patch to the ebuild just so you can see what I did. With these changes, the ebuild compiles and installs. I'll check in a minute if the umlaut problem is gone as well ... Keep your fingers crossed.
Created attachment 94920 [details, diff] Stripped-down 08_revert_cjk.patch
Created attachment 94921 [details, diff] Patch to ghostscript-esp-8.15.2_p20060520.ebuild
No, it does not. Umlauts are still gone. :-(((( I'm afraid I have to give up here. Again, thanks for your input. Maybe it's easier to wait for the next 'standard' release of gs-esp.
If you still want to try around a bit I have made a new svn ebuild at http://genstef.homelinux.org/local/app-text/ghostscript-esp/ghostscript-esp-8.15.2_p20099999.ebuild it checks out the latest code from svn. http://genstef.homelinux.org/08_revert_cjk.patch applies cleanly save one file. But if it is really needed I guess a bug should be filed upstream so that they can apply it to the ghostscript-esp svn.
Okay, I will try it asap, which will be no sooner than 20th September. I am (literally) on a lonely island now and have only a dialup connection and no printers to try this out with. I'll report back what your new ebuild does for me when I'm back from my vacation. Thanks!
Looks good. I made a quick check with a Brother HL 1470 and an HP Laserjet 2100TN. On both, the umlauts come out as they should. I didn't bother trying http://genstef.homelinux.org/08_revert_cjk.patch, since everything is fine without. For me, this issue is resolved now. Thank you very much for your help!
