From: David G. <go...@py...> - 2004-11-11 03:56:01
|
[David Goodger] >> The "trim" attributes would have to match the context for the >> substitution to be applied. [Felix Wiemann] > Why? Given the definition above, I can insert an en-dash with > spaces around |--| like this |--| and I can insert an en-dash > without spaces-|--|-by surrounding it with hyphens (which is > sometimes needed, too). I misunderstood. I was thinking about my previous proposal for dashes, like: .. |M| unicode:: U+02014 .. EM DASH :trim: - And similarly for spaces: .. |emsp| unicode:: U+02003 .. EM SPACE :trim: I had originally thought of this for spaces: .. |M| unicode:: U+02003 .. EM SPACE :trim: So "word-|M|-word" would result in an em-dash, and "word |M| word" would result in an em-space. The substitutions would be context-sensitive. Perhaps not that great of an idea. But now that I do understand what you meant, I don't like it so much. Needing to write extra hyphens in order not to get spaces around an em-dash is ugly and a kludge. Even target cases, like "50-|--|-100", are ugly. I'm thinking that ":trim: -" might not be such a good idea after all. -- David Goodger <http://python.net/~goodger> |