glXUseXFont (was: Re: Identifying remaining core font users)

2009-11-14 Thread Richard W.M. Jones
On Fri, Nov 13, 2009 at 05:42:42PM +0100, Nicolas Mailhot wrote: • ocaml-lablgl-0:1.04-2.fc12 — /usr/lib64/ocaml/stublibs/dlltogl.so This one uses a single reference to glXUseXFont to turn an X core font into a GL display list. There is one oblique reference to doing this manually using

Re: Identifying remaining core font users

2009-11-13 Thread Nicolas Mailhot
Le jeudi 12 novembre 2009 à 21:25 -0600, Chris Adams a écrit : Xaw and core fonts are not something new programs should use, but they still work. Are they really a significant maintenance issue? Core fonts are an issue for anyone working on X or Linux fonts. You may think this year's big X11

Re: Identifying remaining core font users

2009-11-13 Thread Andreas Schwab
Nicolas Mailhot nicolas.mail...@laposte.net writes: Le jeudi 12 novembre 2009 à 14:59 +0100, Andreas Schwab a écrit : Nicolas Mailhot nicolas.mail...@laposte.net writes: Therefore, I'd like to identify remaining core font users, and remind them periodically their core font use is not good

Re: Identifying remaining core font users

2009-11-13 Thread Nicolas Mailhot
Le vendredi 13 novembre 2009 à 09:55 +0100, Andreas Schwab a écrit : I don't understand. Emacs uses the fallback for example to display characters from the Korean KSC charset, in my case using -daewoo-minco-medium-*-ksc5601.1987-0 fonts. What's the approved way to do that? You do it like

Re: Identifying remaining core font users

2009-11-13 Thread Andreas Schwab
Nicolas Mailhot nicolas.mail...@laposte.net writes: You do it like everyone else. You pass the codepoints to fontconfig Emacs does that already AFAIK. Andreas. -- Andreas Schwab, sch...@redhat.com GPG Key fingerprint = D4E8 DBE3 3813 BB5D FA84 5EC7 45C6 250E 6F00 984E And now for something

Re: Identifying remaining core font users

2009-11-13 Thread Nicolas Mailhot
Le vendredi 13 novembre 2009 à 10:55 +0100, Andreas Schwab a écrit : Nicolas Mailhot nicolas.mail...@laposte.net writes: You do it like everyone else. You pass the codepoints to fontconfig Emacs does that already AFAIK. Then it has no actual need to the fallback path. It's probably only

Re: Identifying remaining core font users

2009-11-13 Thread Patrice Dumas
On Wed, Nov 11, 2009 at 01:11:03PM +0100, Nicolas Mailhot wrote: Hi, • grads grads-0:1.9b4-28.fc12 — /usr/bin/gradsc — /usr/bin/gradsdods — /usr/bin/gradshdf — /usr/bin/gradsnc — /usr/bin/gxtran The X core fonts are used to draw custom widgets on the graphs. Maybe using xft as

Re: Identifying remaining core font users

2009-11-13 Thread Patrice Dumas
On Fri, Nov 13, 2009 at 11:06:45AM +0100, Nicolas Mailhot wrote: Le vendredi 13 novembre 2009 à 10:55 +0100, Andreas Schwab a écrit : Nicolas Mailhot nicolas.mail...@laposte.net writes: You do it like everyone else. You pass the codepoints to fontconfig Emacs does that already AFAIK.

Re: Identifying remaining core font users

2009-11-13 Thread Nicolas Mailhot
Le vendredi 13 novembre 2009 à 11:09 +0100, Patrice Dumas a écrit : On Fri, Nov 13, 2009 at 11:06:45AM +0100, Nicolas Mailhot wrote: Le vendredi 13 novembre 2009 à 10:55 +0100, Andreas Schwab a écrit : Nicolas Mailhot nicolas.mail...@laposte.net writes: You do it like everyone else.

Re: Identifying remaining core font users

2009-11-13 Thread Patrice Dumas
On Fri, Nov 13, 2009 at 11:22:40AM +0100, Nicolas Mailhot wrote: Well in that case you do not include the code for old unices in the version built for modern unices such as Fedora It is not always possible to distinguish whether the unix is old or new at compile time. Maybe the Xft version

Re: Identifying remaining core font users

2009-11-13 Thread Nicolas Mailhot
Le vendredi 13 novembre 2009 à 17:42 +0100, Nicolas Mailhot a écrit : Here is a new filtered list, based on the suggestions I've received since my first posting. Please tell me if there is still some files that should not belong here, and why I should have written that the test used in this

Re: Identifying remaining core font users

2009-11-13 Thread Jonathan Underwood
2009/11/13 Nicolas Mailhot nicolas.mail...@laposte.net: • xdvik-0:22.84.14-7.fc12  — /usr/bin/pxdvi-xaw3d  — /usr/bin/xdvi-xaw3d A more crusty piece of code than xdvi you never did see. It's very much in bugfix only mode upstream, so I very much doubt upstream, such as it is, will be moving to

Re: Identifying remaining core font users

2009-11-13 Thread Nicolas Mailhot
Le vendredi 13 novembre 2009 à 18:01 +, Jonathan Underwood a écrit : 2009/11/13 Nicolas Mailhot nicolas.mail...@laposte.net: • xdvik-0:22.84.14-7.fc12 — /usr/bin/pxdvi-xaw3d — /usr/bin/xdvi-xaw3d A more crusty piece of code than xdvi you never did see. It's very much in bugfix

Re: Identifying remaining core font users

2009-11-13 Thread Ville Skyttä
On Friday 13 November 2009, Nicolas Mailhot wrote: Le vendredi 13 novembre 2009 à 11:58 +, Richard W.M. Jones a écrit : Nicolas, if possible next time please give the package maintainer (ie. FAS username) next to each package in the list. Otherwise it's harder to tell which packages I

Re: Identifying remaining core font users

2009-11-13 Thread Nicolas Mailhot
Le vendredi 13 novembre 2009 à 21:08 +0200, Ville Skyttä a écrit : On Friday 13 November 2009, Nicolas Mailhot wrote: Le vendredi 13 novembre 2009 à 11:58 +, Richard W.M. Jones a écrit : Nicolas, if possible next time please give the package maintainer (ie. FAS username) next to

Re: Identifying remaining core font users

2009-11-13 Thread Toshio Kuratomi
On Fri, Nov 13, 2009 at 03:36:14PM +, Richard W.M. Jones wrote: I'm no fan of Python either, but there is a fairly simple JSON API to the package database. eg: wget -O ocaml.json 'https://admin.fedoraproject.org/pkgdb/packages/name/ocaml?tg_format=json' which gave me a huge

Re: Identifying remaining core font users

2009-11-12 Thread Gilboa Davara
On Thu, 2009-11-12 at 07:42 +0100, Nicolas Mailhot wrote: Le jeudi 12 novembre 2009 à 06:51 +0200, Gilboa Davara a écrit : I own both icewm and idesk. As far as I know, both icewm and idesk are linked against xft and should not default to core fonts. (Unless I completely misunderstanding

Re: Identifying remaining core font users

2009-11-12 Thread Patrice Dumas
On Wed, Nov 11, 2009 at 02:53:00PM +0100, Nicolas Mailhot wrote: The right person to ask this would be Behdad, as he's the Fedora/Red Hat/upstream maintainer of most core components of our current text stack. IIRC his advice last time I asked the question was to avoid accessing fontconfig

Re: Identifying remaining core font users

2009-11-12 Thread Richard W.M. Jones
On Thu, Nov 12, 2009 at 02:42:12PM +0100, Patrice Dumas wrote: On Wed, Nov 11, 2009 at 02:53:00PM +0100, Nicolas Mailhot wrote: The right person to ask this would be Behdad, as he's the Fedora/Red Hat/upstream maintainer of most core components of our current text stack. IIRC his advice

Re: Identifying remaining core font users

2009-11-12 Thread Andreas Schwab
Nicolas Mailhot nicolas.mail...@laposte.net writes: Therefore, I'd like to identify remaining core font users, and remind them periodically their core font use is not good for their users or for Fedora. What's wrong with proving support for core fonts as a fallback? That's what Emacs is

Re: Identifying remaining core font users

2009-11-12 Thread Adam Jackson
On Thu, 2009-11-12 at 14:59 +0100, Andreas Schwab wrote: Nicolas Mailhot nicolas.mail...@laposte.net writes: Therefore, I'd like to identify remaining core font users, and remind them periodically their core font use is not good for their users or for Fedora. What's wrong with proving

Re: Identifying remaining core font users

2009-11-12 Thread Andreas Schwab
Adam Jackson a...@redhat.com writes: On Thu, 2009-11-12 at 14:59 +0100, Andreas Schwab wrote: Nicolas Mailhot nicolas.mail...@laposte.net writes: Therefore, I'd like to identify remaining core font users, and remind them periodically their core font use is not good for their users or for

Re: Identifying remaining core font users

2009-11-12 Thread Adam Jackson
On Thu, 2009-11-12 at 16:00 +0100, Andreas Schwab wrote: Adam Jackson a...@redhat.com writes: On Thu, 2009-11-12 at 14:59 +0100, Andreas Schwab wrote: Nicolas Mailhot nicolas.mail...@laposte.net writes: Therefore, I'd like to identify remaining core font users, and remind them

Re: Identifying remaining core font users

2009-11-12 Thread Richard W.M. Jones
On Thu, Nov 12, 2009 at 01:49:34PM +, Richard W.M. Jones wrote: Behdad's advice to me was to use Xft to replace raw X*Font calls in the example I gave: http://caml.inria.fr/cgi-bin/viewcvs.cgi/ocaml/trunk/otherlibs/graph/text.c?rev=6171view=markup This is the patch I submitted upstream:

Re: Identifying remaining core font users

2009-11-12 Thread Nicolas Mailhot
Le jeudi 12 novembre 2009 à 13:08 +0200, Gilboa Davara a écrit : On Thu, 2009-11-12 at 07:42 +0100, Nicolas Mailhot wrote: Le jeudi 12 novembre 2009 à 06:51 +0200, Gilboa Davara a écrit : I own both icewm and idesk. As far as I know, both icewm and idesk are linked against xft and

Re: Identifying remaining core font users

2009-11-12 Thread Nicolas Mailhot
Le jeudi 12 novembre 2009 à 14:59 +0100, Andreas Schwab a écrit : Nicolas Mailhot nicolas.mail...@laposte.net writes: Therefore, I'd like to identify remaining core font users, and remind them periodically their core font use is not good for their users or for Fedora. What's wrong with

Re: Identifying remaining core font users

2009-11-12 Thread Nicolas Mailhot
Le jeudi 12 novembre 2009 à 20:34 +0100, Nicolas Mailhot a écrit : However please note that even though using xft is less bad than using core fonts, xft alone is still not a complete text stack. (I should have written, using xft directly. xft2 uses fontconfig but direct xft2 access bypasses

Re: Identifying remaining core font users

2009-11-12 Thread Nicolas Mailhot
Le jeudi 12 novembre 2009 à 17:38 +, Richard W.M. Jones a écrit : On Thu, Nov 12, 2009 at 01:49:34PM +, Richard W.M. Jones wrote: Behdad's advice to me was to use Xft to replace raw X*Font calls in the example I gave:

Re: Identifying remaining core font users

2009-11-12 Thread Patrice Dumas
On Wed, Nov 11, 2009 at 01:11:03PM +0100, Nicolas Mailhot wrote: Hi, • cernlib cernlib-0:2006-34.fc12 — /usr/lib64/cernlib/2006/lib/libgrafX11.so.1_gfortran.2006 — /usr/lib64/cernlib/2006/lib/libpacklib-lesstif.so.1_gfortran.2006 —

Re: Identifying remaining core font users

2009-11-12 Thread Nicolas Mailhot
Le jeudi 12 novembre 2009 à 22:47 +0100, Patrice Dumas a écrit : Hi, Cernlib is already legacy, so it wouldn't be so bad that the graphical stuff in cernlib doesn't work. Also it is not clear that it uses that much directly X, but rather goes through Motif. Maybe Xm* symbols should not be

Re: Identifying remaining core font users

2009-11-12 Thread Patrice Dumas
On Wed, Nov 11, 2009 at 01:11:03PM +0100, Nicolas Mailhot wrote: Hi, encoding standards change. Also, few users means we do not install core font packages by default anymore, so packagers that depend on them but forgot to mark the deps in their packages will deliver broken packages to

Re: Identifying remaining core font users

2009-11-12 Thread Patrice Dumas
On Thu, Nov 12, 2009 at 11:38:26PM +0100, Nicolas Mailhot wrote: Le jeudi 12 novembre 2009 à 22:47 +0100, Patrice Dumas a écrit : Going through Motif or another widget lib does not change the problem. Apps will be affected the same whether they access Core fonts directly or through a proxy.

Re: Identifying remaining core font users

2009-11-12 Thread Kevin Kofler
Patrice Dumas wrote: The second part of what I wanted to say is that maybe it is not really useful to bother apps using legacy Fonts because they use a given widget library that mandates the use of such fonts Then they need to be ported to a modern toolkit (by upstream). E.g. Xaw has long

Re: Identifying remaining core font users

2009-11-12 Thread Chris Adams
Once upon a time, Kevin Kofler kevin.kof...@chello.at said: Then they need to be ported to a modern toolkit (by upstream). E.g. Xaw has long stopped being viable. (In fact it was never supposed to be more than an example.) If there is no upstream, then maybe it's time to retire the package?

Identifying remaining core font users

2009-11-11 Thread Nicolas Mailhot
Hi, It has been plain since 2003¹ our new font access standard would be fontconfig. Since then most users of the old core fonts X11 backend have migrated, but there are still a few stragglers. Unfortunately these stragglers matter. Core fonts were not good in 2003, and they didn't get any better

Re: Identifying remaining core font users

2009-11-11 Thread Hans de Goede
On 11/11/2009 01:11 PM, Nicolas Mailhot wrote: Hi, It has been plain since 2003¹ our new font access standard would be fontconfig. Since then most users of the old core fonts X11 backend have migrated, but there are still a few stragglers. Unfortunately these stragglers matter. Core fonts were

Re: Identifying remaining core font users

2009-11-11 Thread Nicolas Mailhot
Le mercredi 11 novembre 2009 à 13:41 +0100, Hans de Goede a écrit : It would help tremendously to know how you generated this list of files / packages which allegedly use Core Fonts, there are quite a few packages of mine there, but in many cases I have no idea as to why they are here. Sure,

Re: Identifying remaining core font users

2009-11-11 Thread Nicolas Mailhot
Le mercredi 11 novembre 2009 à 13:50 +0100, Nicolas Mailhot a écrit : Le mercredi 11 novembre 2009 à 13:41 +0100, Hans de Goede a écrit : It would help tremendously to know how you generated this list of files / packages which allegedly use Core Fonts, there are quite a few packages of

Re: Identifying remaining core font users

2009-11-11 Thread Richard W.M. Jones
On Wed, Nov 11, 2009 at 01:11:03PM +0100, Nicolas Mailhot wrote: • ocaml ocaml-0:3.11.1-0.rc1.2.fc12.1 — /usr/lib64/ocaml/graphics.cmxs • ocaml ocaml-runtime-0:3.11.1-0.rc1.2.fc12.1 — /usr/lib64/ocaml/stublibs/dllgraphics.so I guess it falls to me (with Debian folk) to do this one, since

Re: Identifying remaining core font users

2009-11-11 Thread Nicolas Mailhot
Le mercredi 11 novembre 2009 à 13:22 +, Richard W.M. Jones a écrit : I guess it falls to me (with Debian folk) to do this one, since upstream are unlikely to care enough to change this old, working code. Here is the code at issue:

Re: Identifying remaining core font users

2009-11-11 Thread Richard W.M. Jones
On Wed, Nov 11, 2009 at 02:53:00PM +0100, Nicolas Mailhot wrote: If depending on this level of libraries is out of the question for you I'd advise ripping the text parts from those modules. Text is much more complex than just drawing a simple form like a triangle, it is getting more complex

Re: Identifying remaining core font users

2009-11-11 Thread Bruno Wolff III
On Wed, Nov 11, 2009 at 13:11:03 +0100, Nicolas Mailhot nicolas.mail...@laposte.net wrote: Therefore, I'd like to identify remaining core font users, and remind them periodically their core font use is not good for their users or for Fedora. Is there any documentation on how to avoid this?

Re: Identifying remaining core font users

2009-11-11 Thread Richard W.M. Jones
On Wed, Nov 11, 2009 at 09:01:43AM -0600, Bruno Wolff III wrote: On Wed, Nov 11, 2009 at 13:11:03 +0100, Nicolas Mailhot nicolas.mail...@laposte.net wrote: Therefore, I'd like to identify remaining core font users, and remind them periodically their core font use is not good for their

Re: Identifying remaining core font users

2009-11-11 Thread Nicolas Mailhot
Le mercredi 11 novembre 2009 à 14:32 +, Richard W.M. Jones a écrit : I emailed him. It has to be said that maybe text in the OCaml Graphics module only works right now for people using fixed in a ISO-8859-1 locale or whatever [in reality, it works for a whole lot more than that], but

RE: Identifying remaining core font users

2009-11-11 Thread Avery, David [DENTK]
To: Development discussions related to Fedora Subject: Re: Identifying remaining core font users On Wed, Nov 11, 2009 at 04:41:32PM +0100, Nicolas Mailhot wrote: It seems the people maintaining text libs are not interested in backends that fail if you use codepoints outside a specific encoding, or when you

RE: Identifying remaining core font users

2009-11-11 Thread Nicolas Mailhot
Le mercredi 11 novembre 2009 à 10:00 -0700, Avery, David [DENTK] a écrit : I have a bigger problem with this, I use fedora boxes to talk to older devices (solaris 2.6 and M88KV4 unix machines) that can never be upgraded to newer X clients. Since the fedora ( or any xorg) servers are talking to

Re: Identifying remaining core font users

2009-11-11 Thread Adam Williamson
On Wed, 2009-11-11 at 09:01 -0600, Bruno Wolff III wrote: On Wed, Nov 11, 2009 at 13:11:03 +0100, Nicolas Mailhot nicolas.mail...@laposte.net wrote: Therefore, I'd like to identify remaining core font users, and remind them periodically their core font use is not good for their users or

Re: Identifying remaining core font users

2009-11-11 Thread Bruno Wolff III
On Wed, Nov 11, 2009 at 11:40:26 -0800, Adam Williamson awill...@redhat.com wrote: On Wed, 2009-11-11 at 09:01 -0600, Bruno Wolff III wrote: On Wed, Nov 11, 2009 at 13:11:03 +0100, Nicolas Mailhot nicolas.mail...@laposte.net wrote: Therefore, I'd like to identify remaining core

Re: Identifying remaining core font users

2009-11-11 Thread Gilboa Davara
On Wed, 2009-11-11 at 13:11 +0100, Nicolas Mailhot wrote: Hi, It has been plain since 2003¹ our new font access standard would be fontconfig. Since then most users of the old core fonts X11 backend have migrated, but there are still a few stragglers. Unfortunately these stragglers matter.

Re: Identifying remaining core font users

2009-11-11 Thread Gilboa Davara
On Thu, 2009-11-12 at 06:51 +0200, Gilboa Davara wrote: I own both icewm and idesk. As far as I know, both icewm and idesk are linked against xft and should not default to core fonts. (Unless I completely misunderstanding something...) - Gilboa OK. Did some reading. I more-or-less

Re: Identifying remaining core font users

2009-11-11 Thread Nicolas Mailhot
Le jeudi 12 novembre 2009 à 06:51 +0200, Gilboa Davara a écrit : I own both icewm and idesk. As far as I know, both icewm and idesk are linked against xft and should not default to core fonts. (Unless I completely misunderstanding something...) I've been asked to filter out xft matches next