August 2004 - openSUSE M17n - openSUSE Mailing Lists

No compose characters possible with asian language setting
by akimbo337＠gmx.de 16 Nov '09

16 Nov '09

Hello! I'm currently using SuSE 9.0 pro for both german and japanese input. This works fine so far, by setting the variables LC_CTYPE to ja_JP.UTF-8 and LANG to de_DE.UTF-8, but nevertheless I experienced a problem: With the LC_CTYPE set to japanese and an input method running (currently kinput2), I cannot type european language compose characters any more - at least the ones I used to produce with two sequent keystrokes ( ´ and e for é for example). The rest of my (german) keyboard layout works all fine, and with LC_CTYPE set to german, é's, à's, ô's, etc pose no problem, but then, of course, I cannot input japanese... This problem seems also to be existent for other language settings that use an input method, I tried it out with a friend who uses chinese input (also on SuSE 9.0 pro). We (my friend and I) are currently writing an article about asian language input with StarOffice/OOo, so we would really appreciate any help or hint concerning this problem. Thanks. Marcel Boeing

5 6

No skim panel in KDE 3.3
by Andrej Keiper 05 Sep '04

05 Sep '04

Hi, everybody, in my KDE 3.3 the skim panel doesn't show up in the system tray. If I start "scim-panel-kde" the two processes scim-panel-kde and skim are running but I see no panel. If I change the scim config file to the skim panel it's the same. The GTK panel works fine. Does it work for somebody? I use the packages from ftp.suse.com/pub/projects/m17n/9.1/. Andrej

2 2

Re: [m17n] GCC and Unicode
by Stephen Holmes 25 Aug '04

25 Aug '04

Fantastic - this anwers my quesiton - thanks! S t e p h e n H o l m e s Localisation Manager, Novell Tel: +353-1-605-8087 Cell: +353-833-5027 Fax: +353-1-605-8200 Novell Inc., the leading provider of information solutions "This e-mail was sent from a 100% Linux desktop. We're back, we're loud, and we're serious" >>>Mike FABIAN <mfabian(a)suse.de> 08/23/04 3:43 pm >>> Stephen Holmes <sholmes(a)novell.com> *******: >This isn't directly related to SUSE - I just happen to be using it, >but I had a question about GCC's handling of an input file encoded >as UTF-8. If I have something like > > char * lstr = Schöne Grüße > >Will the compiler compile this to UCS-4 internally, or just leave it >as a byte sequence? Here is an answer from our compiler expert Michael Matz <matz(a)suse.de>: matz> For string literals basically the byte values are transfered literally matz> from the input file to the object file. (C99 is more complicated than matz> this, but with UTF-8 files this is what it boils down to). matz> matz> With newer GCC versions (3.4 onward) normal string literals have their own matz> character set, which can be chosen by the user. By default this also is matz> UTF-8, so the behaviour is the same, i.e. UTF-8 strings are copied matz> literally to the object file. matz> matz> This is different with _wide_ character literals, i.e. strings of the form matz> LBlabla . With GCC 3.3 the input bytes are simply widened, i.e. such matz> wide character literal contains the same bytes as the UTF-8 encoding, matz> except that each byte is extended to 32 bit. With GCC 3.5 this is matz> different. There the input characters (in UTF-8) are converted into the matz> wide character set, which is UTF-32 (i.e. UCS-4) by default on machines matz> where wchar is 32 bit wide (i.e. all linux platforms). matz> matz> Hence, with gcc 3.3 there is no recoding involved. UTF-8 strings will be matz> placed into the object file. matz> matz> With newer gcc there is recoding for _wide_ string literals, from UTF-8 to matz> UCS-4 (i.e. 32 bit per character). Without user intervention there is no matz> recoding for normal string literals, i.e. UTF-8 strings remain such. matz> matz> > If I have something like matz> > matz> > char * lstr = Schöne Grüße matz> matz> So, given this (assuming the input is UTF-8): matz> char *str = Schöne Grüße; matz> wchar *wstr = LSchöne Grüße; matz> matz> with our compiler 'str' will be in UTF-8, 'wstr' will be in something matz> strange (UTF-8 but each byte extended to 32 bit). matz> matz> with newer compilers and no special options while compiling 'str' will be matz> in UTF-8, and 'wstr' will be in UCS-4. -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian ***** -- To unsubscribe, e-mail: m17n-unsubscribe(a)suse.com For additional commands, e-mail: m17n-help(a)suse.com

1 0

GCC and Unicode
by Stephen Holmes 23 Aug '04

23 Aug '04

Hi Folks, This isn't directly related to SUSE - I just happen to be using it, but I had a question about GCC's handling of an input file encoded as UTF-8. If I have something like char * lstr = Schöne Grüße Will the compiler compile this to UCS-4 internally, or just leave it as a byte sequence? Thanks in advance. Steve. S t e p h e n H o l m e s Localisation Manager, Novell Tel: +353-1-605-8087 Cell: +353-833-5027 Fax: +353-1-605-8200 Novell Inc., the leading provider of information solutions "This e-mail was sent from a 100% Linux desktop. We're back, we're loud, and we're serious"

2 1

RE: [m17n] GCC and Unicode
by 我不知道 23 Aug '04

23 Aug '04

Hi, Cannot read you question code in IE. Where did you use Unicode string? Identifier or literal? If it is in identifier, GCC does not support it, beacause GCC thought C99 standard has problem. If it is in literal, your usage has problem. As I remembered, accroding GCC document, source code will first be transfered to UTF-8 encoding, then be parsed. YANG RongWei 2004-8-23 Shanghai, China >From: "Stephen Holmes" <sholmes(a)novell.com> >To: <m17n(a)suse.com> >Subject: [m17n] GCC and Unicode >Date: Fri, 20 Aug 2004 15:56:01 +0200 > >Hi Folks, > >This isn't directly related to SUSE - I just happen to be using it, but I had a question about GCC's handling of an input file encoded as UTF-8. If I have something like > > char * lstr = Sch�ne Gr��e > >Will the compiler compile this to UCS-4 internally, or just leave it as a byte sequence? > >Thanks in advance. >Steve. > >S t e p h e n H o l m e s >Localisation Manager, Novell > >Tel: +353-1-605-8087 >Cell: +353-833-5027 >Fax: +353-1-605-8200 > >Novell Inc., the leading provider of information solutions >"This e-mail was sent from a 100% Linux desktop. We're back, we're loud, and we're serious" _________________________________________________________________ ��ĵ��ʼ�ϵͳ�� MSN Hotmail�� http://www.hotmail.com

1 0

kinput2 and dead keys
by Rowdy Rednose 21 Aug '04

21 Aug '04

If I enable kinput2, I am no longer able to use dead keys. Does anyone know, how I can use both?

3 3

Re:_Re:_[m17n]_Re:_[m17n]_SCIM_not_OK!
by Gustaf Kugelberg J�nsson 17 Aug '04

17 Aug '04

Hi again, >>You mean >> >>export XMODIFIERS=@im=SCIM >>emacs & > >>and then "Control+Space" doesn't work? Works >fine for me and many >>others. it didn't work, nothing at all happened, except that emacs started. (actually, once or twice the first few letters I typed turned red, but that was all) I tried launching scim and scim -d after and before the above, but it returned the same error: failure to find/load "socket server module" in both cases. I noticed that my .xim is long, complex (if clauses etc) and possibly outdated. Can that be the reason? Gustaf

2 1

Re:_[m17n]_SCIM_not_OK!
by Gustaf Kugelberg J�nsson 13 Aug '04

13 Aug '04

Sorry, I will use the mailing list! >You mean > >export XMODIFIERS=@im=SCIM >emacs & >and then "Control+Space" doesn't work? Works >fine for me and many >others. On SuSE 9.1, SLES9 and on the current >development version of >SuSE. Thanks, I think that I have tried exactly that, but I am not sure, I will try it when I get home (no internet at home). I seem to remember that I should change my .xim though, which I am not quite competent to do. It's not clear though how I choose language and and then variety (traditional). >> PS Is it my imagination or does suse get worse >> for each release, in terms >> of CJK handling? > It's your imagination. A lot has been improved > for CJK in SuSE 9.1. > Really. Oh, I didn't mean that it was your fault, and I agree that utf was badly needed. Also, it certainly is more integrated now, I was just missing those packages, and the "j" directory on ftp that made it easy to find them afterwards. > We have a much more packages than will fit > on the CDs and the people creating the CDs > always have to find a good > compromise. That was what I suspected. So is there no way to get those other packages, excepting the CJK from your site? I will download the missing packages, and the new scim. thank you. > If you still miss something, just tell me, I'll > upload it. One thing I have missed is CEDICT and CCDICT, the Chinese equivalents of Edict, though they've never been in SUSE, as far as I know. They have been part of Debian and can also be downloaded, but I suppose not everyone is aware of there existence. What I really miss most is kpackage, I think its name was, I much better tool for installing downloaded packages than YAST2, that simply doesn't work on my system, it refuses to find files on the HD, it complains that they are not on the DVD eventhough I chose an RPM from the HD. Thank you, Gustaf

2 1

Re: Autohinter and Bangla/Bengali fonts
by Mike FABIAN 10 Aug '04

10 Aug '04

Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: > MuktiNarrow looks OK - maybe disabling the autohinter for Likhan would > be a good idea. OK, I did that, the fontconfig rpms in ftp://ftp.suse.com/pub/projects/m17n/9.1 contain this change now. -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

1 0

newbies: japanese + polish
by Marek Wawrykiewicz 10 Aug '04

10 Aug '04

Hi, I'm new to this group. And I've big (big for me) problem. I use SuSE 9.1 Professional with polish language (coding iso-8859-2) and this work very well. But lastly I need use some japanese texts and I have problem is possible use iso-8859-2 with japanese coding? What about console? Sorry for stupid questions, but I'm ... lamer. Best regards, Marek

2 4

Re: [m17n] Re: Bangla/Bengali fonts
by Mike FABIAN 10 Aug '04

10 Aug '04

Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: > On Mon, 2004-08-09 at 20:14 +0200, Mike FABIAN wrote: >> Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: >> >> > Well, Mitra and Ani are not being actively developed anymore (basically >> > they were designed as "proof of concept fonts", and released even before >> > Microsoft released the Open Type specs for Bengali) - and currently >> > Akaash is going a rewrite. So I don't think it would be a good idea to >> > include those fonts at this stage. >> >> Mitra wasn't included anyway because it is apparently not >> available anymore from > > Ohh - Mitra is included within the Ani tarball. I see, you are right. Somehow I missed that. -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

1 0

Re: Autohinter and Bangla/Bengali fonts
by Mike FABIAN 09 Aug '04

09 Aug '04

Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: > pointing out the patch. However, before going ahead, please verify the > rendering in some actual text editor/web browser. For sample text, you > can go to http:/www.bengalinux.org/bn . The main issue we have been > facing with the FreeType2 autohinter is that glyphs in a rendered string > fail to align properly (horizontal alignment, that is). An example of > proper rendering is at > http://sayamindu.randomink.org/ramblings/images/akaashabnormal.jpg > Note how the top "matra" lines up more or less perfectly - this is > something which we have not got with the Freetype autohinter > (unpatched). I tried with http://www.mandrakelinux.com/l10n/bn.php3 now. Result with "Mukti Narrow" with autohinter is here: ftp://ftp.suse.com/pub/projects/m17n/konqueror-mukti-narrow-hinting-true.png The top "matra" lines apparently line up well. When using Likhan instead of "Mukti Narrow", it looks less good though: ftp://ftp.suse.com/pub/projects/m17n/misc/konqueror-likhan-hinting-true.png Is that acceptable? -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

1 0

Re: [m17n] Re: Bangla/Bengali fonts
by Mike FABIAN 09 Aug '04

09 Aug '04

Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: > Well, Mitra and Ani are not being actively developed anymore (basically > they were designed as "proof of concept fonts", and released even before > Microsoft released the Open Type specs for Bengali) - and currently > Akaash is going a rewrite. So I don't think it would be a good idea to > include those fonts at this stage. Mitra wasn't included anyway because it is apparently not available anymore from http://savannah.nongnu.org/download/freebangfont/ I have now removed Ani and Akaash, new rpms are at the usual place: ftp://ftp.suse.com/pub/projects/m17n/9.1/noarch/fonts-bangla-0.20040805-0.1… ftp://ftp.suse.com/pub/projects/m17n/9.1/src/fonts-bangla-0.20040805-0.1.sr… -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

1 0

Autohinter and Bangla/Bengali fonts
by Mike FABIAN 09 Aug '04

09 Aug '04

Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: > Also, if you using the autohinter of FreeType2 in SuSE, you'll need > to disable it for the two fonts, since autohinter does not work very > well with Indic fonts. To do so, add the following lines to > /etc/fonts/fonts.conf: > > <match target="font"> > <test name="family"> > <string>Mukti Narrow</string> > </test> > <edit name="hinting"> > <bool>false</bool> > </edit> > </match> > <match target="font"> > <test name="family"> > <string>Likhan</string> > </test> > <edit name="hinting"> > <bool>false</bool> > </edit> > </match> If the autohinter were really bad for Bengali fonts, I think it would be better to disable it for all Bengali fonts and for each font separately. Like this: <match target="font"> <test name="lang" compare="contains"> <string>bn</string> </test> <edit name="hintstyle" mode="assign"> <const>hintnone</const> </edit> </match> But actually I have the impression that these fonts look better when the autohinter is used. Can you please have a look at these two screenshots: ftp://ftp.suse.com/pub/projects/m17n/misc/mukti-narrow-hinting-false.png ftp://ftp.suse.com/pub/projects/m17n/misc/mukti-narrow-hinting-true.png The one without the hinting has more smeared out gray, therefore it looks a bit bolder but considerably more blurred. Maybe the reason why the hinted version looks better is that the SuSE freetype2 package has the patch from http://www.kde.gr.jp/~akito/patch/freetype2/2.1.7 applied which improves the hinting results for CJK fonts a lot. Quite possibly this helps for Indic fonts as well. When you wrote that the autohinter doesn't work well for Indic fonts, you were talking about an unpatched freetype2, didn't you? -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

1 0

Re: [m17n] Re: Bangla/Bengali fonts
by Mike FABIAN 09 Aug '04

09 Aug '04

Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: > I would suggest that you include MuktiNarrow and Likhan for the time > being - we are still not very happy with the quality of the other fonts. Is it necessary to remove the other fonts? They are not very big anyway, maybe it is better to leave them in the package for testing? -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

1 0

Re: [m17n] Re: Bangla/Bengali fonts
by Mike FABIAN 09 Aug '04

09 Aug '04

Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: > MuktiNarrow is a sans-serif, while Likhan will be a serif font. OK, I added these to the fontconfig rules to be the fontconfig setup to be prefered in Bengali locales. Packages for SuSE Linux 9.1 are here: ftp://ftp.suse.com/pub/projects/m17n/9.1/i586/fontconfig-2.2.96.20040728-2.… ftp://ftp.suse.com/pub/projects/m17n/9.1/i586/fontconfig-devel-2.2.96.20040… ftp://ftp.suse.com/pub/projects/m17n/9.1/src/fontconfig-2.2.96.20040728-2.1… ftp://ftp.suse.com/pub/projects/m17n/9.1/x86_64/fontconfig-2.2.96.20040728-… ftp://ftp.suse.com/pub/projects/m17n/9.1/x86_64/fontconfig-devel-2.2.96.200… By the way, if Likhan is a serif font, shouldn't it also use serif glyphs for the Latin alphabet? Currently Likhan has sans-serif glyphs for the Latin alphabet. -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

1 0

[m17n]_SCIM_not_OK!
by Gustaf Kugelberg J�nsson 07 Aug '04

07 Aug '04

Hi, I have also had troubles with SCIM, mine won't start at all. Is there a reasonable manual somewhere? I haven't found one and I know too little about backends and such to get it started without help. The built in manual presumes that I am an expert, which I am not. Could anyone give me some pointers? Thanks, Gustaf

2 1

SCIM not OK? 9.1 Pro New installion
by Isen 06 Aug '04

06 Aug '04

Hi, I just install my brand new SuSE 9.1 Pro on my computer, and SCIM not working. I already tried to put KDE locale to Chinese, but SCIM also not working. My question: 1. Can I have english locale with Working chinese (simplified) input. 2. Is there any configuration file to tweak to make SCIM working? thanks =========================================================================================== "Gabung INSTANIA, dapatkan XENIA. Daftar di www.telkomnetinstan.com, langsung dapat akses Internet Gratis.. Dan ..ikuti "Instan Smile" berhadiah Xenia,Tour S'pore, Komputer,dll, info hub : TELKOM Jatim 0-800-1-467826 " ===========================================================================================

2 5

Re: [m17n] Re: Bangla/Bengali fonts
by Mike FABIAN 06 Aug '04

06 Aug '04

Sayamindu Dasgupta <sayamindu(a)randomink.org> さんは書きました: > On Thu, 2004-08-05 at 18:01 +0200, Mike FABIAN wrote: > The glyphs for MuktiRegular and MuktiBold were withdrawn by the company > which donated them due to possible copyright infringements > (http://www.gnu.org.in/software/software.html#akruti) - so we had to > withdraw those fonts too :-( Thank you. I made a SuSE rpm containing the fonts from http://www.nongnu.org/freebangfont/ and will submit this for inclusion into SuSE Linux >= 9.2. An rpm for SuSE 9.1 for testing is here: ftp://ftp.suse.com/pub/projects/m17n/9.1/noarch/fonts-bangla-0.20040805-0.1… ftp://ftp.suse.com/pub/projects/m17n/9.1/src/fonts-bangla-0.20040805-0.1.sr… Can you please tell me which of these fonts should be the preferred font in /etc/fonts/fonts.conf for sans-serif serif monospace respectively? -- Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian 睡眠不足はいい仕事の敵だ。

1 0

SUSE :: l10n in Bangla/Bengali
by Jamil Ahmed 05 Aug '04

05 Aug '04

Hello, We [www.ankurbangla.org] is working towards Bangla support in GNU/Linux. Currently we are working on Fedora Linux, Mandrake Linux, Gnome, KDE. You can see the status from the url's below. Fedora Linux http://elvis.redhat.com/cgi-bin/i18n-status?page=status&locale=bn Mandrake Linux http://www.mandrakelinux.com/l10n/bn.php3 Gnome 2.6.x http://l10n-status.gnome.org/gnome-2.6/bn/index.html KDE http://i18n.kde.org/stats/gui/HEAD/bn/index.php We are interested to work on SUSE Linux too! Let us know the general procedure - how to start on this issue. Waiting for your valued reply. :-) Thanks, `Jamil

2 5

Converting Japanese filenames to UTF-8
by David Nettles 03 Aug '04

03 Aug '04

I did the right tthing and backed up my SuSE 8.2 Pro system to an external 160GB USB2.0 drive formatted with reiserfs. My SuSE 8.2 Pro system was installed in Japanese. I took my copy of SuSE 9.1 Pro and installed it. I plugged in the USB2.0 drive, it mounted read-only, and proceeded to try to recover my home directory only to find all of the Japanese filenames were 100% unreadable-scramble (文字化け!!!). I realize that SuSE 9.1 uses UTF-8 -- and I am really happy about that -- but I need to recover the files from this backup. I now have filenames that used to be "第８課のノート.txt" that now look like "□□機■□.txt" or "?????????.txt" or any other unreadable malarky. I did look into the utility *convmv* and tried it using a variety of source encodings -- jis, jis7, sjis, shift_jis, iso-2022-jp. None worked. :( Any ideas? I am sure UTF-8 will be much better, but I really need to be able to restore from this archive first. Perhaps part of my problem is that I have no idea what the source encoding would have been on the SuSE 8.2 Pro system. I did do the entire installation in Japanese, so it kept the default for Japanese, I imagine. I really want this new SuSE 9.1 environment to work. -- David Nettles web: http://www.miteyo.org email: tetsuoni3000(a)yahoo.co.jp __________________________________________________ GANBARE! NIPPON! Yahoo! JAPAN JOC OFFICIAL INTERNET PORTAL SITE http://mail.ganbare-nippon.yahoo.co.jp/

2 4

UTF-8 enooding can't print GB2312
by Isen 02 Aug '04

02 Aug '04

Hi, as adviced by Mike I set my encoding to en_US.UTF-8. I can happily type chinese character, and print it. (Open Office can't be used for SCIM?) but the problem is, I CAN open document with chinese character from my old SuSE installation, but CAN NOT print it. I believe it's because in the past I used something like zh_cn.gb2312. is there anyway I can read & print my old file. or, is there any web site I can read about all the chinese problem & how to fix it in SuSE 9.1. thanks. =========================================================================================== "Gabung INSTANIA, dapatkan XENIA. Daftar di www.telkomnetinstan.com, langsung dapat akses Internet Gratis.. Dan ..ikuti "Instan Smile" berhadiah Xenia,Tour S'pore, Komputer,dll, info hub : TELKOM Jatim 0-800-1-467826 " ===========================================================================================

2 3

Re: [m17n] Chinese fonts anti-aliasing problem
by pangaoyong 02 Aug '04

02 Aug '04

> "pangaoyong" <pangaoyong(a)126.com> ��ϕ��ޤ��: > > > I select the Chinese TrueType fonts AR PL SungtiL GB but I still can't get it > > working fine, this is my new Konqueror setting. > > [HTML Settings] > > AutoDelayedActions=true > > AutoLoadImages=true > > AutomaticDetectionLanguage=0 > > ChangeCursor=true > > DefaultEncoding= > > Fonts=AR PL SungtiL GB,AR PL SungtiL GB,AR PL SungtiL GB,AR PL SungtiL GB,AR > > PL SungtiL GB,AR PL SungtiL GB,0 > > FormCompletion=true > > HoverLinks=true > > MaxFormCompletionItems=10 > > MediumFontSize=10 > > MinimumFontSize=7 > > ShowAnimations=Enabled > > UnderlineLinks=false > > The Arphic PL fonts don't have embedded bitmaps. You said you liked > the way English looked when you used the bitmap font "Adobe Helvetica" > for English. But you cannot achieve that for Chinese with the "AR PL > SungtiL GB" of course as this font doesn't have bitmaps, only > outlines. > > You can get the outlines rendered with anti-aliasing but apparently > you don't like this. Or, you can switch the anti-aliasing off by > adding the rule > > <match target="font"> > <edit name="antialias"> > <bool>false</bool> > </edit> > </match> > > to your ~/.fonts.conf. Then the outlines will be rendered in black and > white only without the gray. But they will still be not as perfect as > hand tuned bitmaps. > > -- > Mike FABIAN <mfabian(a)suse.de> http://www.suse.de/~mfabian > ˯�߲��Ϥ��¤Δ�� > > -- > To unsubscribe, e-mail: m17n-unsubscribe(a)suse.com > For additional commands, e-mail: m17n-help(a)suse.com > > > > ----------------------------------------------------------- Yes, you are right, when I switch off the anti-aliasing by adding the lines, the fonts rendered in black and white without the gray, it looks better then ever. Thanks ----------------------------------------------------------- ��ϵ�http://www.126.com��260Mȫ��! �ṩ��ʼ��ֻ��ѹ��, ��ʱ��ʼ��Ϣ!

1 0