From koffice-devel Sun May 27 02:01:30 2001 From: "Dom Lachowicz" Date: Sun, 27 May 2001 02:01:30 +0000 To: koffice-devel Subject: Re: wvWare 2 X-MARC-Message: https://marc.info/?l=koffice-devel&m=99096248431722 > > - How are we going to represent Unicode strings within the library? > > STL strings? (related: is iconv portable? I heard of some problems on > > FreeBSD) > >On systems with no iconv or broken iconv, you can use "libiconv" - >this is what GLib does. Otherwise you have to maintain your own >codeset conversion tables, which I'm told is basically insanity. >(Though it might make sense if you only had one or two codesets to >worry about.) Hi guys, libiconv is a tremendous asset. I'd really like to use an iconv implementation here. It's quite reliable and very portable - both wvWare 1.0 and AbiWord use it. Abi is known to run on about 6 different Unices and win32, be, mac, qnx, etc... wvWare is known to run on a superset of Abi's supported platforms (os/2, s/390, etc...) As I've said, both use iconv, so it's quite portable. So what do we do to represent strings? Using stl strings isn't a bad idea. We can't use something like QString directly, but maybe we want to mimick its behavior. Abi has some UCS2, UTF8, and C string classes internally that are quite generic and not related to any particular platform or implementation - very vanilla and generic C++ code. I think that maybe if we "beef" those classes up a bit, they could be put to good use here. Dom _________________________________________________________________ Get your FREE download of MSN Explorer at http://explorer.msn.com _______________________________________________ Koffice-devel mailing list Koffice-devel@master.kde.org http://master.kde.org/mailman/listinfo/koffice-devel