On Wed, Mar 24, 2004 at 06:12:05PM -0500, Volker Quetschke wrote: >>>I also had a look at the code and reached pretty much the same conclusion >>>as Volker. Replacing all 'isslash (*src)' and 'isslash (src[x])' in >>>normalize_posix_path by "isdirsep ()" would yield more consistent results. >>>I know this code is delicate but the possible drawback isn't obvious. >> >>speed disadvantage for nebulous gain. >I will not comment on the nebulous gain, but it is not necessary >at all to use sdirsep () or both '\' and '/' in normalize_posix_path. > >Just let the function do what its comment says, namely >"\'s are converted to /'s in the process."
Easy enough to change the comment. I've done that. I've sent a patch to Corinna for comment which should have a minimal performance hit on parsing posix paths but will a have slightly larger effect if it backslashes are hit. >I.e. put: > >--- snip --- >for (char *p = dst; (p = strchr (p, '\\')); p++) > *p = '/'; >--- snap --- So, check every character in the string to see if it's a backslash then use another loop to iterate over the string to check for other stuff. No thanks. As you note in your later message, this wouldn't work anyway. dst is "const" for a reason. -- Christopher Faylor spammer -> [EMAIL PROTECTED] Cygwin Project Leader Red Hat, Inc. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/