Yes Mike... please disregard this patch for now because no point in
development without any requirement.
Thanks & Regards
Prabhu GS
On Mon, 2010-10-04 at 11:01 -0400, C. Michael Pilato wrote:
> Are you saying, then, that you'd like us to disregard your patch for now?
>
> On 09/29/2010 08:50 AM, prabhugnanasundar wrote:
> > Mike,
> > Sorry for the delayed response.
> > Your point is very much a valid one, Mike. But since we used
> > *re.match*, "perf" would not match "superfun". But your point really
> > holds good when "super" would match "superfun" repo. Actually I was not
> > aware of the -s option(thank you for that) while I coded this. I wanted
> > the regex part only to match the like-repo names in the config. But now,
> > this serves my purpose. I don't see any specific requirement from anyone
> > for this regex part. I would like to do the further enhancements if it
> > really is required... Thanks for the suggestions Mike :)
> >
> >
> > Prabhu GS
> >
> > On Fri, 2010-09-24 at 11:23 -0400, C. Michael Pilato wrote:
> >> On 09/22/2010 09:07 AM, prabhugnanasundar wrote:
> >>> Hi,
> >>>
> >>> The svnperms.py script reads the svnperms.conf file, which needs the
> >>> section rules for all the repos uniquely. This was quite tough when we
> >>> have n number if repos. I wished that svnperms.py reads regex matches
> >>> and applies the configuration rules and came up with this patch.
> >>> Mid-way of my coding work, I came to know that -s switch would let us
> >>> specify the section name explicitly, but not the regex thing.
> >>> I modified the code to search for a matching section via regex, which
> >>> might be really helpfull for certain cases.
> >>> I have attached the log file and the patch file with this mail for
> >>> review.
> >>
> >> I'm not familiar with this script, but won't your change break compatibility
> >> with existing users of the script? Imagine that someone is using the script
> >> today with 'svnperms.py -s perf', but their configuration file also has a
> >> section called "[superfun]" (something that "perf" would match, albeit
> >> incompletely. Will the new script, invoked the old way, read the wrong
> >> configury?
> >>
> >> Perhaps you should tie this change of interpretation to a new command-line
> >> flag, perhaps a capitalized "s" (-S).
> >>
> >
>
>
Received on 2010-10-05 11:19:53 CEST