This is the mail archive of the
mailing list for the glibc project.
Re: [PATCH] manual: Clarify the documentation of strverscmp [BZ #20524]
- From: Michael Kerrisk <mtk dot manpages at gmail dot com>
- To: Florian Weimer <fweimer at redhat dot com>
- Cc: libc-alpha <libc-alpha at sourceware dot org>, Michael Kerrisk-manpages <mtk dot manpages at gmail dot com>
- Date: Sun, 11 Sep 2016 16:35:00 +0100
- Subject: Re: [PATCH] manual: Clarify the documentation of strverscmp [BZ #20524]
- Authentication-results: sourceware.org; auth=none
- References: <20160829202145.A2548403DC4A9@oldenburg.str.redhat.com> <CALxWeYpXBMo6+d3QTtbz1rM2Jv0Gn+G5OTNDV8Sze06Gp9hnrw@mail.gmail.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com>
On Thu, Sep 8, 2016 at 2:10 PM, Florian Weimer <firstname.lastname@example.org> wrote:
> On 08/31/2016 01:56 PM, Florian Weimer wrote:
>> Perhaps we should just say the comparison happens in an
>> implementation-defined manner and is only consistent within a single
>> process (just like strcoll). It would simplify matters and would us
>> allow to fix the algorithm. :)
> Any further comments?
Not really any from my part, except:
> I was only half-joking. Maybe we should indeed leave the exact sorting
> order unspecified.
I'm not sure you can get away with that. I mean, there might be
applications out there that depend on the current weird behavior(?).