Mailing List ArchiveSupport open source code!
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]Re: tlug: Introduction; Looking for Help on i18n
- To: k.wetzel@example.com
- Subject: Re: tlug: Introduction; Looking for Help on i18n
- From: "Stephen J. Turnbull" <turnbull@example.com>
- Date: Fri, 7 Nov 1997 08:54:12 +0900 (JST)
- Cc: tlug@example.com
- Content-Transfer-Encoding: 7bit
- Content-Type: text/plain; charset=us-ascii
- In-Reply-To: <3462100E.3405C9@example.com>
- References: <199711050600.GAA01129@example.com><3460A91B.B609A094@example.com><m0xTPLz-00000LC@example.com><3462100E.3405C9@example.com>
- Reply-To: tlug@example.com
- Sender: owner-tlug@example.com
(By the way, it's not clear to me whether you are subscribed to the list; if you're getting two copies and it's annoying you, say so and I'll stop sending to both.) >>>>> "Kai" == Kai Wetzel <k.wetzel@example.com> writes: Kai> The LIP has to determine what we can provide in the library Kai> and what would be out of scope or is beeing worked on by Kai> other people. I can't say much about this myself, since I Kai> only know that my Linux/X11 has vertially no support for Kai> anything outside of the "C" locale :( This I can help with. X11 intends to use the system's native locale support, and by default locale support is provided by libc. In order to use the support provided by X11, you should do something like the following in every module that requires locale support from Xlib: #define X_LOCALE #include <X11/Xlocale.h> This substitutes X11's functions for libc's. The database is in /usr/X11R6/lib/X11/{i18n,locale,nls} some or all of which are symbolic links to each other. This database is not compatible with the POSIX format. If you don't have that database, you're in trouble, of course. This can also be done at runtime by use of the "Netscrap_X11_liblocale.so" hack: you make a wrapper function to interface what libc functions will be calling to the _Xsetlocale function. Then run existing locale-ized programs such as Motif programs with `LD_PRELOAD=/path/to/Netscrap_X11_liblocale.so proggy'. These tricks both work with non-X programs, but you do have link against Xlib, of course. And if you're not running under X you may have to provide facilities like fonts and input methods yourself. I'd provide the code, but (a) I forget where I left it and (b) you should figure this out for yourself anyway because the code used to let Netscrap work with Japanese intentionally omits some locale features, and you need to decide if that's appropriate. >> C/C++? Support? NOT. There is some support for POSIX locales >> in Linux glibc, but no documentation I can find (I haven't >> looked that hard, and have not joined the ML, though) Kai> Well, there is e.g. the class "locale" and there seems to be Kai> some other facilities. I don't know how complete they are Kai> and I don't know how much of them is usable in common Kai> implementations of the standard C++ library. Right, but without the database they're nothing. Some of the locale data are shit^H^H^H^Halpha (I believe this is true for Korean); without docs, you can't tell which are good. And some locales have simply gone missing and nothing is said about attempts to put them together (eg, Japanese). Kai> Well, I think both is needed, extending existing classes as Kai> well as potentially introducing classes which are mainly Kai> there to provide i18n features. Examples from the Java API Kai> would be Locale, Collation (was: "TCollator"), Date/Time Kai> formats (was: "formatter"), etc. Of course some things may Kai> effect the core classes fundamentally, such as fonts. I may be wrong, but I suspect that if you look closely you will see that some or all of these classes are actually integrated into very low levels of the API. Of course you _need_ a locale class to manage the various aspects of locale (for example, a Frenchman reading mail from the US might very well want the currency symbol to be `$' but the decimal separator to be `,', so custom locales will be necessary for some cranky users). But the fact that it's user-visible doesn't mean that it doesn't go all the way to the root of the IO classes. Again, I think you have it backwards about fonts; the fonts actually come last in the display pipeline: markup -> format dates, etc -> line-breaking by font metrics -> display using fonts and so can be grafted on at the last second. Ciao. Off to teach. Steve Next TLUG meeting is Saturday Dec. 13, 1997 (possibly Nov. 13?) --------------------------------------------------------------- a word from the sponsor: TWICS - Japan's First Public-Access Internet System www.twics.com info@example.com Tel:03-3351-5977 Fax:03-3353-6096
- Follow-Ups:
- Re: tlug: Introduction; Looking for Help on i18n
- From: Kai Wetzel <k.wetzel@example.com>
- References:
- tlug: Simple Ghostview Question
- From: bickel@example.com
- tlug: Introduction; Looking for Help on i18n
- From: Kai Wetzel <k.wetzel@example.com>
- tlug: Introduction; Looking for Help on i18n
- From: "Stephen J. Turnbull" <turnbull@example.com>
- Re: tlug: Introduction; Looking for Help on i18n
- From: Kai Wetzel <k.wetzel@example.com>
Home | Main Index | Thread Index
- Prev by Date: Re: tlug: Introduction; Looking for Help on i18n
- Next by Date: Re: tlug: Introduction; Looking for Help on i18n
- Prev by thread: Re: tlug: Introduction; Looking for Help on i18n
- Next by thread: Re: tlug: Introduction; Looking for Help on i18n
- Index(es):
Home Page Mailing List Linux and Japan TLUG Members Links