March 2005 - openSUSE M17n - openSUSE Mailing Lists

japanese fonts in firefox revisited
by Mike Shegedin 10 Mar '05

10 Mar '05

OK. I had it working before. It was sweet. But upon doing a new SuSE install, I can't seem to get Japanese printing under FF to work again. However, a few things are different. This time, I'm guessing that font embedding is enabled -- that is, FF is embedding the fonts into the ps file when printing because GS doesn't complain about missing fonts. However the resulting mozilla.ps file is around 320k which I think is about the same size as the file generated when I didn't have font-embedding working so I'm not sure about that. I'm curious to know what turns font embedding on and off when printing from FF. It would not seem to be the qtconfig setting right off because even when I turn off font embedding there, the resulting ps filesize seems to be the same as when printer font embedding is enabled as well as the fonts refered to within the resulting printed ps file. While I'm at it, do qtconfig settings have any bearing on any of this? The ps file currently being generated has lots of Numbus fonts which clearly are not Japanese fonts. So how do I convince FF to output using a Japanese font? FF displays Japanese just fine using SuSE's Japanese fonts or other Japanese TT fonts I have installed. Just can't get to them when printing. BTW I have wrappers generated for all of my installed Japanese fonts, so I think that as long as GS had a file with Japanese text and refered to a Japanese font then I would be in business. Mike S ghostscript tells me that the fonts embedded in my ps file are __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1

Re: [m17n] Japanese printing in Firefox
by Mike FABIAN 06 Mar '05

06 Mar '05

Mike Shegedin <sandynomike(a)yahoo.com> さんは書きました: >> Do you have the ghghostscriptjcjkackage installed? [...] > Yes, I had recently installed ghghostscriptjcjkHere > are the GS packages I have now: > ghghostscriptonts-other-7.07.1rcrc207.1 > ghghostscriptonts-std-7.07.1rcrc207.1 > ghghostscriptibrary-7.07.1rcrc207.1 > ghghostscriptjcjk0021119-211 > ghghostscript11-7.07.1rcrc207.1 > > After installing ghghostscriptjcjkI also ran the > following > SuSuSEconfig-module ghghostscriptjcjkper your > directions on > hthttp/wwwwwususeedemfmfabianususejcjkhghostscriptthtmlhghostscriptbut OK. > I wasn't able to do much after that. Let's check whether your firefox/Mozilla prints with fontembedding or not (Both is possible, depending on the settings in firefox/Mozilla). Print to a file instead of directly to the printer and then run Ghostscript on that file: gs mozilla.ps You will see messages about fonts which are loaded on standard output. Do you see any messages about missing fonts? If Mozilla prints without font embedding, it writes a request for the standard Japanese font name "Ryumin-Light" into the Postscript output because this font is available in most Japanese PostScript printers and also supported by Ghostscript with the ghostscript-cjk extras. I.e., if you Mozilla prints without fontembedding, you should see a message about loading Ryumin-Light in the standard output of Ghostscript. On the other hand, if you Mozilla prints with font embedding, you will not see Ghostscript loading Ryumin-Light because the fonts are already contained in the Postscript document. You should then see which fonts were embedded by grepping in the Postscript file, for example: mike@kibou:~$ grep %%BeginResource mozilla.ps %%BeginResource: CMap IPAPGothic.Regular.0.0_cmap %%BeginResource: CIDFont IPAPGothic.Regular.0.0 mike@kibou:~$ -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

2 6

Re: [m17n] should empty keybindings be allowed in scim?
by James Su 05 Mar '05

05 Mar '05

Hi, Yes. And it's easy to fix. I can provide a patch against 1.2.0, so that you can commit it into SL 9.3. Regards James Su Mike FABIAN wrote: >Tobias Burnus <burnus(a)net-b.de> $B$5$s$O=q$-$^$7$?(B: > > > >>>Maybe it should be allowed to disable keybindings by setting them to >>>empty? Zhe Su, what do you think? >>> >>> >>> >>Hmm, better would be some methode to disable it which leaves the keys >>untouched. >> >> > >For the trigger hotkey, maybe. > >But I think it might be useful if empty values for all configurable >keybindings were allowed. Maybe one does not need a certain function >at all and doesn't want to have it on any key in order not to hit it >accidentally. Then it might be nice to make the key binding empty in >the scim configuration dialog. > >Currently it is not possible to make such key bindings empty, if >nothing is found in the configuration files, fallback key bindings >which are hard coded in the source code are used. > >I think this would be easy to change to accept empty values to be able >to disable certain key bindings, but would that make sense? > >Zhe, what do you think? > > >

2 1

Re: [m17n] skim and option reloading
by James Su 05 Mar '05

05 Mar '05

Hi, Thank you very much for your report. I'll try my best to make SCIM better and better. Your suggestions and bug reports are really very helpful for our development. Regards James Su Mike FABIAN wrote: >Tobias Burnus <burnus(a)net-b.de> さんは書きました: > > > >>- used input methods: I can call reload from the menu plus call --reload. >> >> > >After that your new hotkey settings should be active. At least that >works fine for me. Do you experience problems already here? > > > >>If I choose exit, it restartes but the hotkey does no longer work. >> >> > >Better don't use that exit button in the scim-panel, as it kills scim. >GTK2 and QT applications (which use scim via the input module plugin >support) will restart scim, but old-style X applications which use >scim via XIM will not restart scim. And such XIM clients like xterm, >mlterm, java, (X)Emacs, ... will often hang if their XIM server is >suddenly killed. > >This "exit" button is only there in a debugging build of scim. >It is not very useful though. > >Currently I build the scim packages with --enable-debug because it is >helpful to track down problems. Maybe I switch that of for the SuSE >9.3 release. > > > >>If I kill everything scim related, >> >> > >That is also problematic, for the same reasons as using this "exit" >button. > > > >>then the applet restarts (automatically) and I finally get the >>updated input method. => Reported to scim bugtracker. >> >> > >Restarting everything scim related cleanly is indeed a bit difficult. >I often have the problem that the scim-panel-gtk behaves >strangely after a > > killall scim-launcher; killall scim-panel-gtk; > scim -d > >often some buttons are missing or the fonts are too small. Killing >scim-panel-gtk once again often helps. > >Currently the only really clean way to restart scim seems to be to >exit all applications using scim first (usually that means to exit >your X session as well) and then restart. > > > >>- somehow the trigger commands (control-s) have problems after such a >>forced restart. >> >> > >I cannot remember that I ever had that problem, but as I wrote >above, scim often behaves strange after a forced restart. > >I have no problems with reloading the options though. At least I can >change the key bindings and then reload the options either from the >panel or from the command line without problems. > > > >>I think I will stay with scim and try very hard _not_ to set options >>which require the restart of scim since this seems to cause problems. >> >> > >There are options which do indeed require a restart and then it may >become a bit difficult if you want to avoid to restart your X session >(usually works for me though. But sometimes I need to kill and restart >several times). > > > >>Hopefully, in SUSE 9.3 it will work better. >> >> > >Of course we are trying to make it work as good as possible and >reports from users like you certainly help a lot. > >Zhe Su, the scim author does a wonderful job with scim, although there >are still some minor bugs, I think it is great already and a huge >improvement over the input methods we used by default for SuSE Linux ><= 9.1. > > >

1 0

scim/skim question
by Tobias Burnus 05 Mar '05

05 Mar '05

Hello, I'll just started to use skim/scim (first for my first steps in Chinese and as support for the Chinese members in our group). (I've the new version from projects/m17n/ on a SUSE Linux 9.2 system with de_DE.UTF-8.) In the skim configuration dialog I can enter keyboard shortcuts to enable/disable scim and to change to the next input, but they don't seem to be effective. Is there a simple method to disable/enable scim completely? I mean in a way I can quickly re-enable it. I tend to hit Shift+space rather frequently and get then some nice characters, e.g., instead of leaving vi (":q"). At "Scim server|input methods" of Skim config, I can enable/disable input methods, but it does not seem to have any effect on both the menu shown by the kicker applet or on the methods shown when cycling through. What is the difference between Keyboard and German/Keyboard>German/European by the way? Question of the complete novice in Chinese: Next to the input parameter field I see with "intelligent pinyin" a button labeled "中" and three other symbols still unknown to me. What is ment by those? By the way, I'll completely miss a description of how the input methods work. I figured out that, e.g., unicode works by entering the entering the hex code of the character, but what means "unicode", "utf-8", "de" next to it? Tobias

2 9

Re: [m17n] scim/skim question
by James Su 04 Mar '05

04 Mar '05

Hi, Why do you want to disable the trigger keys? You can use an unusual key binding for trigger key to prevent enabling scim by accident. Regards James Su Mike FABIAN wrote: >Tobias Burnus <burnus(a)net-b.de> $B$5$s$O=q$-$^$7$?(B: > > > >>I'll just started to use skim/scim (first for my first steps in Chinese >>and as support for the Chinese members in our group). (I've the new >>version from projects/m17n/ on a SUSE Linux 9.2 system with de_DE.UTF-8.) >> >>In the skim configuration dialog I can enter keyboard shortcuts to >>enable/disable scim and to change to the next input, but they don't seem >>to be effective. >> >> > >Did you click on "Reload config"? > > > >>Is there a simple method to disable/enable scim completely? I mean in a >>way I can quickly re-enable it. I tend to hit Shift+space rather >>frequently and get then some nice characters, e.g., instead of leaving >>vi (":q"). >> >> > >If you have the latest packages from projects/m17n, you can change >the trigger key from the command line like this: > > scim-config-agent -s /Hotkeys/FrontEnd/Trigger=Control+Alt+Shift+space > scim-config-agent --reload > >You could put such commands to set the trigger key to some weird, >inconvenient key combination which you cannot hit by accident in to a >script and execute such a script with a button on your KDE desktop for >example. Then you would have a button to disable and another button to >enable scim. > >Currently it is *not* possible to disable scim by making the value of >/Hotkeys/FrontEnd/Trigger empty, in that case scim uses Control+Space >which is hardcoded in the source code. > >Maybe it should be allowed to disable keybindings by setting them to >empty? Zhe Su, what do you think? > > > >>At "Scim server|input methods" of Skim config, I can enable/disable >>input methods, but it does not seem to have any effect on both the menu >>shown by the kicker applet or on the methods shown when cycling through. >> >> > >Looks like you need to reload the config. > > > >>What is the difference between Keyboard and >>German/Keyboard>German/European by the way? >> >> > >I think none. > > > >>Question of the complete novice in Chinese: Next to the input parameter >>field I see with "intelligent pinyin" a button labeled "$BCf(B" and three >>other symbols still unknown to me. What is ment by those? >> >> > >The symbol which looks like a crescent moon toggles between full width >and half width. You can use that for example to write full with $B#L#A#T(B >$B#I#N(B. The next button to the right which shows a period and a comma >toggles between the full width and the half width forms for >punctuation. > > > >>By the way, I'll completely miss a description of how the input methods >>work. >> >> > >Yes, some more documentation is needed. Some nice descriptions made by >users are already available on the internet and I plan to add some >description of the various scim input methods to my home page as well >after SuSE 9.3 is released. > > > >>I figured out that, e.g., unicode works by entering the entering >>the hex code of the character, but what means "unicode", "utf-8", >> >> > >No, the 4 digit Unicode code point, i.e. UCS2 (=UTF-16 in the BMP). > > > >>"de" next to it? >> >> > >"de" next to what? > > >

2 1

Japanese printing in Firefox
by Mike Shegedin 04 Mar '05

04 Mar '05

Hi. I'm using SuSE 9.2 w/Japanese support. I can input/display/print Japanese in most apps including konqueror and Openoffice. However I cannot print Japanese under Firefox. All I get are little boxes in place of Japanese Characters when printing. I know it's Japanese printing under Firefox is possible because I had Firefox printing Japanese about 6 months ago under Mandrake thanks to many hours of guesswork which finally lead to the magical combination of qtconfig font substitutions and Firefox settings. Unfortunately I lost it at some point and never managed to get it going again. Since then I'm a confirmed SuSE convert but I really need to get this working. Help! Mike __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

3 2