On Thu, Jan 03, 2008 at 10:18:10AM +0000, Colin Watson wrote:
That's a well-documented damned-if-I-do-damned-if-I-don't which has nothing to do with how much we care about groff.http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=196762 Fortunately Brian expressed an intent to work on the remaining pieces recently.
Yes. In fact, I hope to have character classes implemented by the end of the day. Then I'll send a patch, and then I'll start work on applying attributes to those character classes.
Just so everything goes smoothly, I want to know if it's a problem if I use std::map (and std::string and std::vector) in implementing character classes. I ask because it seems like every attempt has been made to avoid usage of the standard library. My need is to be able to map strings into an array/vector of character classes, and I need each character class (pointer) to be able to belong to more than one string.
Also note that my implementation of character classes makes contiguous ranges very efficient, but everything else is stored as individual entries.
I have one further question: there are already flags in charinfo for BREAK_BEFORE and BREAK_AFTER. If I clear the BREAK_BEFORE flag on a glyph, will that prevent the line from being broken before that glyph (and similarly the BREAK_AFTER flag)? If so, kinsoku shori handling might be done by the end of the week.
-- brian m. carlson / brian with sandals: Houston, Texas, US +1 713 440 7475 | http://crustytoothpaste.ath.cx/~bmc | My opinion only a typesetting engine: http://crustytoothpaste.ath.cx/~bmc/code/thwack OpenPGP: RSA v4 4096b 88AC E9B2 9196 305B A994 7552 F1BA 225C 0223 B187
signature.asc
Description: Digital signature