Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!usenet.ins.cwru.edu!news
From: bammi@dsrgsun.ces.cwru.edu (Jwahar R. Bammi)
Newsgroups: comp.os.minix
Subject: Re: POSIX headers, which of the below are correct
Keywords: POSIX
Message-ID: <1989Dec1.233711.24354@usenet.ins.cwru.edu>
Date: 1 Dec 89 23:37:11 GMT
References: <682@augean.OZ: <22014@princeton.Princeton.EDU>
Sender: news@usenet.ins.cwru.edu
Reply-To: bammi@dsrgsun.ces.cwru.edu (Jwahar R. Bammi)
Organization: Case Western Reserve University
Lines: 33
In-reply-to: nfs@notecnirp.Princeton.EDU (Norbert Schlenker)

In article <22014@princeton.Princeton.EDU>, nfs@notecnirp (Norbert Schlenker) writes:
>The only legal way to do this correctly, I believe, is something like:
>
>FILE *fopen(const char *path, const char *mode)
>{
>...
>  char localpath[PATH_MAX+1];
>...
>  strcpy(localpath, path);
>...
>  creat(localpath, 0666);
>...
>}
>
>So there is extra overhead (both time and space) for doing this right.
>I understand why Earl did what he did; I did almost the same thing for
>my package, but had the audacity to ignore gcc's complaints.
>
with gcc atleast (i dont have access to any other ansi C compiler) you can
cast a (const char *) to a (char *) without having to make a local copy.
(or vise versa). i dont know for sure if these casts are legal ansi C, but
gcc -pedantic does'nt complain. (assuming of course you really want to do
that). another thing that struck me as strange (unless it has changed recently)
are the protos for strchr, strrchr and the like: 
the proto says `char *strchr(const char *s, int c)' and what you are
really doing is returning either a pointer into s or NULL. should'nt
the return type be `const char *' ? or are you really supposed to return a 
pointer into a writable copy of s? i used casts in all such cases.
--

bang:   {any internet host}!dsrgsun.ces.CWRU.edu!bammi	jwahar r. bammi
domain: bammi@dsrgsun.ces.CWRU.edu
GEnie:	J.Bammi