From: Vadik M. <vad...@mt...> - 2004-09-01 16:28:26
|
=F3=D2=C5=C4=C1 01 =F3=C5=CE=D4=D1=C2=D2=D8 2004 19:01, Michael Jennings = =CE=C1=D0=C9=D3=C1=CC: > On Wednesday, 01 September 2004, at 12:14:40 (+0400), > > Vadik Mironov wrote: > > These two patches replace broken getword function in file.c with the > > not broken one (I hope). Also i made small warkaround about one bug, > > and i submit another patch later, but i need this function, so it > > will be really cool, if somebody merge these patches to the CVS. > > Well, if someone had really wanted to fix this, they probably would've > just stolen the code from libast: > > http://www.eterm.org/libast/html/strings_8c-source.html#l00340 > > No offense, but your code is...well, iffy. Your use of goto's and > labels is completely unnecessary. > > Michael Well, character handling not an easy task and it doesnt matter how <iffy> t= he=20 code looks like. It should be robust, understandable (try to imagine how yo= ur=20 function will be executed under various circumstances), simple and well=20 documented. I would like to have an simple function in library with clear=20 docs and visible path of execution.=20 =2D-=20 With best regards, Vadik Mironov =09 <vad...@mt...> PS. No defense :-) Could you please tell me what happened in this function with such a little= =20 program? int main(int argn,char *argc[]) { get_word(1, "\"Anne got cr"); return 0 ; } |