4/12/02 [EMAIL PROTECTED] : >Using "non-ASCII" characters in email can produce strange results.
True, but that's no excuse. The machine should adapt, not the user. At least the Mac user. I must say that I experience almost no text-related problems with French in Emailer, apart from: - messages with encoded subjects from other mailers (I didn't bother correcting those yet) - messages I would SEND using Emailer 2.0v3, which could get truncated if not cleaned first using the "Avoid QP truncation" script - messages from hotmail (rarely other mailers, depends on their settings), easily handled by dirty scripts of mine that send the text to a browser and get the correct result from it (I had planned a much cleaner approach, but got lazy when my first test script proved good enough, I only intended to use iCab to test encodings, but well, in France "temporary" means "forever"); Yet I noticed too that Mail.app's default encoding confuses Emailer. I'll look for a fix when the problem reaches annoying factor 2. If someone has some time to throw at it, it won't be hard to script. My suggestion is a script that either: - detects the encoding based on known offending headers (declared encoding, X-mailer, other details) and then converts the text (I think there was an osax for that, maybe we can access Text Encodings directly nowadays) or - simply lets the user test different encodings until the message looks right (this would be much easier and almost as good, especially assigned to a keyboard shortcut via OSAMenu) 4/12/02 0:09 Helder Correia : >By my count, your signature is 25 characters long when it travels via >Emailer->Emailer, but it becomes 75 characters long when it travels via >Mail->Emailer. Apparently, Mail is converting each individual >high-ascii character in the original signature to three high-ascii >characters. > >What's even stranger is that none of the three high-ascii characters >generated by Mail (which I received as 0x92 0xE2 0xE0) match the >high-ascii character you meant to send (which I received as 0x89). Nothing strange, probably Unicode. Unicode is definitely the way to go (single encoding, all the characters from any language or quite), but unfortunately it spreads slowly because it's not vital for roman languages that accommodate various 256-characters encodings. And it has some flaws too, like the fact that accented chars may come in 2 flavors (the accented char itself, which is better from a typographer's point of view, or composited from the base char and the accent, which is sexier from a geek's perspective). ---- VRic ___________________________________________________________________________ To unsubscribe send a mail message with a SUBJECT line of "unsubscribe" to <[EMAIL PROTECTED]> or <[EMAIL PROTECTED]>
