This is the mail archive of the
binutils@sourceware.cygnus.com
mailing list for the binutils project.
Re: Absolute paths in BFD
- To: binutils at sourceware dot cygnus dot com
- Subject: Re: Absolute paths in BFD
- From: Chris Faylor <cgf at cygnus dot com>
- Date: Sat, 22 Apr 2000 16:34:49 -0400
- References: <3902086F.48AC66FE@cygnus.com>
On Sat, Apr 22, 2000 at 04:15:44PM -0400, Timothy Wall wrote:
>I've noticed quite a few instances in BFD where absolute paths are
>tested by expecting the first character to be '/' (grep for "== '/'" to
>see how often this is done). This test breaks under cygwin when drive
>letters are used as part of a pathname.
Just to pick nits, it's actually a problem under Windows, not Cygwin
specifically. Cygwin tries to mask the x:\ madness.
>There should probably be a bfd function for absolute path-ness, and that
>called instead of explicitly doing the leading slash test.
>
>I was about to make a small patch when I realized that this check is
>going on in many more places than the bug I was tracking.
I seem to be fighting a losing battle to convince everyone to just use
Cygwin POSIX-style pathnames, so this seems like a great idea for
the MS-DOS pathname purists. It would be nice to make bfd able to
understand at least the x:/foo/bar (i.e., without the backslashes)
version of MS-DOS syntax.
I don't think we want to go as far as allowing '\' as a path separator,
though.
cgf