FreeType 2.2 RC1 has been released with a call for testing: http://lists.gnu.org/archive/html/freetype-devel/2006-02/msg00094.html I'd like an ebuild in the tree to follow up on that request ;-). While recently a lot of effort has been undertaken to make the new release ABI compatible[1][2] a few API apparently issues remain, detailed here: http://freetype.sourceforge.net/freetype2/freetype-2.2.0.html 1 = http://lists.gnu.org/archive/html/freetype-devel/2006-02/msg00037.html 2 = http://lists.gnu.org/archive/html/freetype-devel/2006-02/msg00039.html
Reopen when it's not RC.
Freetype 2.2.1 has been released. This is the new stable version of the library. It's available from: http://savannah.nongnu.org/download/freetype/ Or: http://sourceforge.net/project/showfiles.php?group_id=3157 (2.2.0 was skipped because a 2.2.0 tarball was mistakingly uploaded about two weeks ago, although never announced. This release has been officially announced by Werner Lemberg on the Freetype mailing list.)
Since the web archive has finally been updated, here's a link to the release announcement on the freetype-devel list, quoting the CHANGES file: http://lists.gnu.org/archive/html/freetype-devel/2006-05/msg00026.html
Maybe the patches at http://lists.nongnu.org/archive/html/freetype-devel/2006-05/msg00044.html should be considered when Freetype 2.2 is added to the tree?
Yes, i saw those some time ago already. With the lingering patent issues I don't think it's wise to add them at this point in time (subpixel), the fontconfig one.. . does it really matter?
Considering it's in Debian now it looks like the patch situation has become bearable - perhaps an effort should be started to get the necessary patches into the tree to finally move forward on this one: http://freetype.sourceforge.net/freetype2/patches/rogue-patches.html
foser do you need some help testing with this? I'd like to give it a try in the next days, especially for the cjk fixes (lots of), if I can be of help just tell me :)
Comparision between CJK hinting in 2.1.10 and 2.2.1: http://www.eikehein.com/ftview-2_1_10-sazanami-2.png http://www.eikehein.com/ftview-current-sazanami-2.png
2.2.1 is in p.mask, sorry for the delay. still need to add blockers for all known rogue apps
Thanks for this ebuild. Works great on my ~x86 Gentoo system. I unmasked this to see if it fixes a problem with CJK fonts I have (for unknown reason bold typeface is way too bold for CJK fonts), and it turns out this version improves the issue a lot. Here are two screenshots of the same page, one made with freetype-2.1.10-r2, one with freetype-2.2.1. The difference is obvious even if you can't read Japanese. http://www.metamorpher.de/files/asahi_2.1.10-r2.png http://www.metamorpher.de/files/asahi_2.2.1.png
Hi, it's time to push transition (no p.mask) a bit :) I've tested freetype-2.2.1 on two ~x86 and three x86 machines for nearly 7 weeks now. I applied patches to the following packages that they compile with new freetype: media-libs/fontconfig-2.3.2-r1 media-libs/sdl-ttf-2.0.8 x11-libs/libXfont-1.1.0-r1 (libXfont-1.2.0 doesn't need to be patched) I took all patches for these packages from http://www.freetype.org/freetype2/patches/rogue-patches.html and they all worked out-of-the-box(tm) :) Fonts rendered by freetype-2.2 do really look MUCH better than with old freetype. Poly-C
Not sure if this problem is related to freetype 2.2.1 specifically, but I'm getting random X.Org crashes that are related to fonts (happens in emacs and xfontsel), see http://bugs.gentoo.org/show_bug.cgi?id=99089#c23
what is holding back this one?
in modules.cfg, 'AUX_MODULES += otvalid' is needed to support East Asian vertical writing for xdvi.
Created attachment 99278 [details, diff] freetype-2.2.1-verticalwriting.patch
This bug is really annoying. It seems that the only thing holding this back is kde-3.5.3 not being stable, although kde-3.5.5 is already availble. Could kde people do something about it, so this could be finally closed ?
This has nothing to do with KDE. The current ~arch version is fine, and the 3.5.5 release will be stable soon. There is another bug for this version of freetype though, and it's been a while since I heard from Foser.
Please, be so kind and name that bug, cause I don't see it here and neither in that tracer bug #14233. I'm not saying "bigfoot does not exist", just "I have not seen a proof of its existance, that's 100% not forged". In other words, I suspect that you're talking about an app that was broken in the first place, freetype page states clearly "apps should not have been using private freetype headers in the first place".
That was #142333 of course, not #14233.
Because it does not fit into that tracker, I'd say (the application builds, it's a bug in freetype is hitting, it's bug #149723 you could have just seen by running a search ...
First thing, it's a games bug, so it should not be a stopper. The second, in that bug there's already a patch to freetype solving the issue.
(In reply to comment #21) > First thing, it's a games bug, so it should not be a stopper. The second, in > that bug there's already a patch to freetype solving the issue. Please, stop producing pointless noise on this bugs. It just annoys developers and won't help anything. Why don't you just unmask this version for yourself if you can't live without it?
Yes, this noise would be pointless, if it would be two, well even three months since 2.2.1 release, but it's nearly six, and the whole thing is starting to look just like those slang-2 "problems" (I'm not saying there are none, but that many of them are artificial). Nearly all of those problems are with apps depending on freetype, not freetype itself and by now nearly all apps that are still being developed were fixed to work with freetype 2.2.
(In reply to comment #23) If you have nothing useful to add, then please stop commenting on this bug. Thanks.
Wouldn't it be an idea to remove the hard mark and just mark this package for testing (and then shock horror, making it available)... that way the devs won't get more "pointless noise"; I won't even point out that it probably takes more effort bitching about "pointless noise" than actually solving the "problem".
If you have issues, then file a *new* bug; this is for version bump that has been done ages ago. Someone unmask it and close this dumpspace for rants.
Unmasked.
