Re: [REVISE coding-cookie] `coding:' cookie as lisp hook

Ville> Hmm, this is a Mule package, right? We'd need a Ville> mule-packages/fsf-compat-mule or something. Not necessarily. A Mule package is one that contains constructs that harm a non-mule XEmacs just by loading them. Non-ISO-8859-1 characters, for example. However, with no eval-when-compile constructs and with the call to -enable deleted, this should be OK; there's no way to get any of those functions called unless the user does something like M-: (coding-cookie-whip-me-beat-me-stomp-me-with-your-high-heels-ooh :-). Ville> ...and there's already some support for them in core and Ville> latin-unity. I feel uncomfortable about scattering Ville> supposedly "broken" functionality around. Well, as Ilya points out in the .texi, this is a difference of opinion. I'm not going to introduce the functionality he wants in 21.4, and not in 21.5 any time soon. We really need a "cookie" facility that recognizes usages that are enforced by standards, where the relevant packages can push a recognizer (eg, a function or regexp) on a list. Then python-mode can push "-\*-.*coding: \(\s_*\)", and psgml "<meta http-equiv=\"content-type\" content=\".*charset=\(.*\)\"" etc on that list, and be confident they'll work because they are enforced by the language standard. What I object to is random hand-maintained cookies hiding in comments. But there really is a lot of demand for this among a certain class of users, especially Japanese and Russians. Those users don't have anything reliable: EUC-JP, Shift JIS, and KOI8 (and Big 5, but we don't seem to have so many Taiwanese users) are for better or worse quasi-national standards, and they're not reliably distinguishable from international standard encodings also in common use by any of our usual tricks. There's some hope for 21.5 with Ben's new detection mechanism, but for now a lot of users prefer to use cookies. As a maintainer, I'm not particularly happy, but we need to support our users. Just tag it like RMail. "New and obsolete; if you don't know, you don't need it; if you aren't using it, don't start." :-) -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN My nostalgia for Icon makes me forget about any of the bad things. I don't have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py