On Tue, 2006-11-14 at 14:58 +0100, Bruno Haible wrote: > Yoann Vandoorselaere wrote: > > > I don't think Chinese users will find it nice if you exclude them from > > > correct functioning of your program because of "performance" or "library > > > size". > > > > I don't think you are qualified to decide in place of the application > > developer whether the application should handle localized input or not. > > Hehe, it's my role as gettext maintainer to encourage internationalization :-) > > > I'm not advocating to not use them: I'm advocating to let the developer > > choose. Some of the library/program using GnuLib are used in embedded > > system where size matter, and where you won't see anything else than > > standard ASCII as input. > > OK, embedded systems. What I can offer, as a compromise, is to introduce > flags like > NO_CHINESE_USERS > NO_JAPANESE_USERS > NO_KOREAN_USERS > NO_TURKISH_USERS > UTF_8_ALL_THE_WAY > so that > - when the first three are defined or the last one is defined, strstr uses > the byte-for-byte implementation, > - when additionally NO_TURKISH_USERS is defined, strcasestr uses the > byte-for-byte implementation, > - when UTF_8_ALL_THE_WAY is defined, iconv becomes a trivial nop. > > With names chosen like this, the user of gnulib or of your software will > know explicitly which compromises he's making. > > Would you be satisfied with that?
The ability to disable localized input will certainly be useful to certain specific project. However, I find the proposed flags name harsh. Using "HANDLING" or "INPUT" in place of "USERS" sound more appropriate to me, don't you think? -- Yoann Vandoorselaere | Responsable R&D / CTO | PreludeIDS Technologies Tel: +33 (0)8 70 70 21 58 Fax: +33(0)4 78 42 21 58 http://www.prelude-ids.com
