mj@myrias.UUCP (Michal Jaegermann) (06/19/89)
A header file <string.h> for a distributed version of dLibs contains,
among other things, a following macro:
#define bcopy memcpy
This should be replaced with
#define bcopy(source,dest,length) memcpy((dest),(source),(length))
Note an order of arguments. This is still not exactly right, since
memcpy returns 'dest' and bcopy returns some unknown integer, but it
should do in most situations when you will need this macro.
The above costed me some head scratching during a port of some code.
Love y'a all
===============================
Michal Jaegermann
Myrias Research Corporation
Edmonton, Alberta, CANADA
...{ncc,alberta}!myrias!mj
==============================mj@myrias.UUCP (Michal Jaegermann) (06/20/89)
> > #define bcopy(source,dest,length) memcpy((dest),(source),(length)) > A correction to my posting from yesterday. Use memmove instead of memcpy. bcopy is supposed to handle correctly overlapping moves. Sorry... Michal Jaegermann Myrias Research Corporation Edmonton, Alberta, CANADA ...{ncc,alberta}!myrias!mj ==============================
to_stdnet@stag.UUCP (06/25/89)
From: dal@syntel.UUCP (Dale Schumacher) [mj@myrias.UUCP (Michal Jaegermann) writes...] > A header file <string.h> for a distributed version of dLibs contains, > among other things, a following macro: > > #define bcopy memcpy > > This should be replaced with > > #define bcopy(source,dest,length) memcpy((dest),(source),(length)) It should really use memmove(), but that's basically correct. I fixed this in my header a while ago, but it got left out when I posted the last set of fixes. I got bitten pretty badly by it as well :-) This and several other bug reports (and FIXES, which is even better) have recently been posted, many by Michal. While it is somewhat of an embarassment to have my mistakes ponited out in gory detail, I can put my ego aside and simply be glad of two things. The problems are being fixed, so others won't have the same trouble, and people must actually be using dLibs if they are finding bugs. John Stanley has also privately sent me a few bug reports, the latest on a problem with fullpath() in qualifying '.'. I'm not planning to post fixes myself, although others can feel free to do so if they wish. I'll be collecting these suggestion up and reviewing them for inclusion in the 1.3 release to be posted with the next release of Sozobon C (no, I don't know when that will be). Until then, enjoy what does work, fix what doesn't and keep reporting bugs! PS. I'll be gone until the 30th of June at the X'hibition '89 conference. -- Dale Schumacher 399 Beacon Ave. (alias: Dalnefre') St. Paul, MN 55104-3527 ...bungia!midgard.mn.org!syntel!dal United States of America "I may be competitive, but I'm never ruthless"
apratt@atari.UUCP (Allan Pratt) (06/27/89)
> From: dal@syntel.UUCP (Dale Schumacher) > ... While it is somewhat of an > embarassment to have my mistakes ponited out in gory detail, I can put > my ego aside and simply be glad ... Try being an Atari Systems Programmer for a while. :-) ============================================ Opinions expressed above do not necessarily -- Allan Pratt, Atari Corp. reflect those of Atari Corp. or anyone else. ...ames!atari!apratt