home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.c
- Path: sparky!uunet!portal!dfuller
- From: dfuller@portal.hq.videocart.com (Dave Fuller)
- Subject: Re: Memory functions - memcpy & memmove
- Message-ID: <BxFKAw.1A5@portal.hq.videocart.com>
- Organization: VideOcart Inc.
- References: <1992Nov8.225440.6545@sq.sq.com>
- Date: Mon, 9 Nov 1992 03:56:55 GMT
- Lines: 25
-
- I've said it before. I could really not give a shit if the code isn't
- perfect. I don't get paid to do this, and i try to answer a lot of the
- questions. If someone wants more than a signed int of chars moved, then
- change the code. I only want to see replys when the code is just flat
- out WRONG.
-
- second, why the hell would it even need to match the ansi code for
- memmove. this isnt even called memmove. it was designed to take the
- place of it where it didn't exist yet. Therefore, there isnt a damn
- standard to follow. so someone needs to cast (char *) to call this.
- I couldn't care less. don't like it, change it. but it works, so
- please quit giving me some B.S. style guide lessons.
-
- ALSO, note that the so called correction missed the point of the code.
- the code WILL NOT, let me repeat that, WILL NOT fail if size == 0 initially.
- because the first thing in the for loop is size --. this will make it
- -1, and will drop out. Of course, if the size were forced to be unsigned
- by some other programmer, then we have a problem. BUT, guess what.
- I wrote this, not another programmer. and it works. and if someone
- else breaks it by changing a definition without seeing the effects
- it has, then GOOD.
-
- Dave Fuller
- dfuller@portal.hq.videocart.com
-
-