On Mon, Nov 10, 2003 at 11:48:20PM +0100, Alain Bench wrote: > Let's take the case from the other side. You can define a locale > with unmunged name by providing a full path as name to localedef(8), > this disables the downcasing and removing of dashes and underscores in > charset names. So say: > > | # localedef -i es_ES -f UTF-8 /usr/lib/locale/es_ES.UTF-8 > > ...to create a real es_ES.UTF-8 shown in --all-locales and which > will be selected by your LANG in priority over the es_ES.utf8 one. Would you believe it? I can't find my glibc build directory :-( I'm in the middle of recompiling glibc. > > installed all locales from the [glibc] build process > > It uses something as "localedef -i es_ES -f UTF-8 es_ES.UTF-8" > without path, which creates the munged version es_ES.utf8 in the default > directory /usr/lib/locale/. oh > >>>> Was the invalid byte raw in sent folder, on sent folder display, or > >>>> the copy/paste generated it? > > Thanks for the files. Both messages are identical¹ to my copies from > the list: Sébastien's is clean, yours is invalid. I'm sorry still don't > know where the problem was, but it seems to have vanished... Now your e > acutes quoted, copy/pasted to askcc and to body are perfect (??). interesting. . . > [$send_charset] > >>> I purposely changed it to UTF, in order to encourage its use. > >> UTF aware mailers (vast majority) will not notice anything. People > >> with old systems or old mailers unable to see UTF will just be > >> annoyed > > I should swap in a pro-UTF signature line when I send UTF email? > > Yes. And perhaps demonstrate its superior capabilities when they are > needed: Mix Spanish and Hebrew in same mail, use nice math symbols, IPA > phonetic pronunciation alphabet, and so on. LOL :-) > No need, no use. Because it > creates unnecessary problems, which is rather dis- than encouraging. I'd rather encourage UTF's use, since it simplifies everything. You only need one encoding no matter what language you want to write in. You _can_ mix languages if you want. You have many auxhiliary characters at your disposal. You can get the Euro symbol without resorting to 8859-15 or something. You can use the Euro symbol in Hebrew text (since 8859-8 doesn't include it - Israelies deal with Euros a lot more than with Dollars, you know. . .). In short, it removes all the complexity of dealing with different "types" of characters. It brings us back to the days of ASCII being all that was necessary for most normal people. > ¹ Not really identical copies: Your evil iPlanet MTA changes a little > bit everything for you. But not touching our todays problem. That > needed to be verified. Yeah, my reply to your private reply vented a bit about OOL's SMTP server. Another annoyance is its virus scanner, which deletes virus attachments and replaces them with text/plain attachments explaining that OOL has robbed me of my email. > Mutt 1.2.5 users: Do us all a favour, set in_reply_to="%i" in muttrc, so > threading is not messed up by silly mail servers. Mutt 1.2.5 users: Do yourselves a favor and switch to the 1.5.x series :-) > Everybody: Let's burn silly iPlanet mail servers. Dump ashes to > trashcan. And void trashcan. I'm all for it ... you'll first have to convince CableVision to fork their OOL SMTP servers over, though. . . - Dave -- Uncle Cosmo, why do they call this a word processor? It's simple, Skyler. You've seen what food processors do to food, right? Please visit this link: http://rotter.net/israel
Attachment:
pgpV102lEfnY2.pgp
Description: PGP signature