Marcus> I am sure that the font I am trying to assign to this Marcus> face, is available as a bold-italic version. You've checked this with xfontsel or something similar that actually displays the font? Is it possible that you have an unscaled font that your X server/fontserver is trying to scale to that size? The implementation of font searching in many common servers is pretty broken, and inconsistent across vendors. We do the best we can, AFAIK, but we haven't had a real X grok look at that stuff in a long time. Make sure other tools find the font you expect. Marcus> It would be nice if someone could tell me whats going on Marcus> there. It could be a Custom bug. Custom is a nice idea that didn't really get finished before being made the default. But all the people who know much about it have recently become employed. However, I do get the bold-oblique expected in this situation (Debian Linux, pretty close to vanilla sid). So maybe you have an old X resource for that face or something like that. If not, try (define-specifier-tag 'a-unique-tag) (set-face-font 'bold-italic "-adobe-courier-bold-o-normal-*-*-120-*-*-m-*-iso8859-1" 'global '(a-unique-tag) 'prepend) in your init file. -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."

Marcus> I think the problem is, that xemacs does some wildcard Marcus> match on the font name, when requested through customize Marcus> and then it gets my ugly bitstream-courier-version font. I Marcus> wish I could specify the font-vendor with customize :-) I don't recall if Customize creates a 14-hyphen XLFD. If it wildcards where possible (which is consistent with the general quality of Customize code), you may be able to fool it by setting the family to "adobe-courier" in the Customize buffer for the font. This of course won't work in the menus, which strip off the foundry information before even displaying the menu to you. Marcus> I tried, seems to work. What is that specifier tag ? An attempt to get custom to leave the specification alone. The main purpose for built-in specifier tags is to specify that some face settings should be used only in certain circumstances. For example, the 'x tag says "only use this specification on X displays", the 'color tag says "only use this specificatoin on color displays". That allows you to use color only on displays that can handle it, but use a different font or size for emphasis on mono/grayscale displays. However, it also affects what specifications are changed by a call to set-face-font and other specifier manipulations. See the Lispref info manual, topic Specifiers, for the dense but very complete full story. Marcus> I come to the conclusion that it is currently the best, to Marcus> avoid italic fonts completely. I also searched this list, Marcus> and it seems that there are more people discovering this Marcus> behaviour. Italic fonts are hard. The terminology is confusing, because most people think it means "slanted", but that's not correct (although most familiar italic fonts _are_ slanted). -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."

Marcus> Is there a way to manage italic and bold-italic faces Marcus> independently from outside xemacs, say with xresources ? X resources can be used (as long as you're not going to use a Windows or GTK port, and they may work on GTK). See etc/Emacs.ad (search for "^! Fonts.$") and the Info documentation (XEmacs | Customization | X Windows IIRC). -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."

Marcus> I played a little with custom and encountered a completely Marcus> different behavior when I define my own face. faces.el works quite hard to ensure that the 'default, 'italic, 'bold, and 'bold-italic fonts contrast, at least with 'default. That may have something to do with it. Marcus> Where does the the face bold-italic come from (where is it Marcus> declared) ? How can I find such things out (apropos does Marcus> not help)? It's faces.el. You might look at getting make-face to put a property on the face name (which is a symbol, so it has a plist) to document where it comes from. I don't know how this is done for variables and functions, though, but I doubt it's all that arcane. -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."

Jeff Mincy <jeff(a)delphioutpost.com&gt; writes: Yeah, this is what I am really missing for a long time. Often I write some text and want to lookup an english word (because I am from german), that I don't know. Currently I have to use a web-browser for this task. W3 is a nop for this because my local Dict. makes use of a lot of html-tables, which w3 does not render fast enough (a joke;) it takes minutes). Some time ago I started such a mode but I dropped it because I found it that hard to parse html with elisp (my local dict only supports html output). Yours Marcus -- Yes, God had a deadline. So he wrote all in LISP. marcus(a)antbear.org http://www.antbear.org

Jeff> What needs to happen to make a package? 1. Get a volunteer. 2. Volunteer decides whether file fits into existing package (eg, generalize edict.el -- not a good idea though, edict requires Mule). If so, check with maintainer, if none, volunteer to be maintainer, if too lazy :), ask Steve Y. Add file(s) to package in CVS, update Makefile. 3. If need new package, pick non-conflicting name, mkdir, put files in there, steal Makefile from similar package, file off identifying marks of other package, insert relevant details for own, get Steve Y's approval (a formality if the volunteer intends to be maintainer), check in. Of course, all steps require sanity checking to make sure package builds, still works, etc, but the basic operations are that simple. Refinements include improving customize support, adding autoload cookies so that auto-autolads eliminates the need to `require' in init files, etc. But that can come later. -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."

From: Steve Youngs <youngs(a)xemacs.org&gt; Date: Thu, 06 Sep 2001 08:16:52 +1000 |--==> "SJT" == Stephen J Turnbull <turnbull(a)sk.tsukuba.ac.jp&gt; writes: Jeff> What needs to happen to make a package? SJT> 3. If need new package, pick non-conflicting name, mkdir, put files in SJT> there, steal Makefile from similar package, file off identifying SJT> marks of other package, insert relevant details for own, get Steve SJT> Y's approval (a formality if the volunteer intends to be SJT> maintainer), check in. The package in question, "dictionary", has my approval for addition into the XEmacs packages. It already has an active maintainer so I'll ask them if they'd like to maintain the XEmacs package version. If they don't want to I'll call for a volunteer here to do it. Sound good? I found the author (Torsten.Hilbrich(a)gmx.net) to be responsive in handling the one issue that came up for xemacs (a coding system issue in v21.1). The dictionary package already seems like it is pretty much packaged up ready to go, and I assume that the author would want to do the xemacs version. But, if not and you need a volunteer, then I guess[1] I volunteer. Footnotes: [1] assuming that I understand what is involved from your previous description of what is necessary. -jeff

On Mon, 03 Sep 2001 15:14:12 EDT, Jeff Mincy said: Typographers distinguish between 'italic' and 'oblique' however. The distinction is (basically) that an oblique font *is* a slanted version of the standard one, whereas an italic font has serifs designed for the slanted version. Incidentally, this also means that there's no such thing as an italic san-serif font (at least not for typographers). And just because Microsoft thinks there's such as thing as an Italic Ariel doesn't make it so..... -- Valdis Kletnieks Operating Systems Analyst Virginia Tech

Marcus> So where could I start looking at customize to understand Marcus> how it works? Maybe I can fiddle with it a little... Is Marcus> cus-face.el a starting point? cus-edit.el, cus-face.el. wid-edit.el (if you want to see how the user interface works). Widget has a man page, not terribly complete. fonts.el for the font model the rest of XEmacs uses (which the Customize programmers evidently don't like). Face manipulations require familiarity with specifiers, see Lispref on topic Specifiers. What I think should be tried is to migrate all custom functionality to specifiers, using specifier tags to identify Custom-specific specifications (such as the user's saved spec in custom.el, the most recent custom-set specification, etc). If this were done right, Custom specifications and set-face-*-style specifications could coexist. I think specifier tags are also a natural way to identify "themes" of related faces. This would allow you to specify themes for particular specifier-locales, such as particular windows or buffers, in a natural way. (However the UI still needs to be written.) But the current code goes about this in a very different way. The aproach chosen in the half-done theme implementation we have does have the advantage of (probably) being portable to GNU Emacs with little effort. -- University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091 _________________ _________________ _________________ _________________ What are those straight lines for? "XEmacs rules."

On Mon, 03 Sep 2001 14:53:07 +0200, Marcus Geiger said: Was the breakage in the original debian package, or was it locally-inflicted breakage that happened to be cleaned up by re-installing? If it's self-inflicted, that's life. ;) If on the other hand it was broken when Debian shipped it, somebody should make a note of that in the PROBLEMS file... /Valdis