On Tuesday, December 22, 2009, 17:22:59, Thomas Fernandez wrote: >> No. If the font doesn't have an Unicode table, it's assumed to be >> CP1252 > Why? I'm sure you can point me to something.
Historical reasons (read: fonts from the Windows 3.x era). JS>> (I previously wrote ASCII, which was wrong). > Changing your mind? No, I just slipped. ASCII is a subset of CP1252 (it doesn't define characters with indexes 128-255, which CP1252 does). > I agree with the font, just not with the charset assumption. Back in the Windows 3.x days, the fonts didn't define any charset, so the displayed character was defined by font name and it's index. That's why for example, the EE (East European) version of Windows shipped fonts with CE in their name (Arial CE, etc.), which replaced characters above index 128 with characters that were more useful for languages used in this region. If you didn't have the correct font, you eg. saw é instead of č. However, the same substitution worked with other characters, which is how the Wingdings (and later Webdings) font worked. However, starting with Windows NT and 95, the fonts included a Unicode character table, which is incompatible with such substitutions. For compatibility reasons, these fonts are now marked as symbol fonts, and still included with Windows, but they shouldn't be used anywhere where they're expected to be rendered on the client side (there was a thread about this exact problem with the same message on thedailywtf recently - mail clients other than Outlook/Outlook Express would display the message with a P instead of the graphic, even if it was sent the way author intended). -- < Jernej Simončič ><><><><>< http://eternallybored.org/ > [ The Bat! 4.2.13.7 ALPHA on Windows 7 6.1.7600. ] Everybody is somebody else's weirdo. -- Dykstra's Law ________________________________________________________ Current beta is 4.2.13.8 | 'Using TBBETA' information: http://www.silverstones.com/thebat/TBUDLInfo.html
