[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: svn commit: r1241050 - /subversion/trunk/subversion/mod_dav_svn/mod_dav_svn.c

From: Julian Foad <julianfoad_at_btopenworld.com>
Date: Tue, 6 Mar 2012 12:39:27 +0000 (GMT)

Daniel Shahaf wrote:

> Daniel Shahaf wrote on Mon, Feb 06, 2012 at 18:20:28 +0200:
>> Stefan Sperling wrote on Mon, Feb 06, 2012 at 17:12:32 +0100:
>> > On Mon, Feb 06, 2012 at 05:59:04PM +0200, Daniel Shahaf wrote:
>> > > This still strips whitespace around ='s in the value:
>> > >    SVNHooksEnv "name = x = y"
>> > > will result in
>> > >    setenv("name", "x=y", 1)
>> > > whereas I believe it should result in
>> > >    setenv("name", "x = y", 1)
>> > > (and, to be honest, I'd be happy with
>> > >    setenv("name ", " x = y", 1)
>> > > as well).
>> > >
>> > > WDYT?  How should it behave?
>> >
>> > I agree.
>> > would telling svn_cstring_split() to no strip whitespace suffice?
>>
>> I assume that should result in the third setenv() case above, so +1.
>
> Ping?  trunk_at_HEAD still strips whitespace around equal signs in the value.

My tuppence-worth?  I agree that the current behaviour as stated above is wrong.  Unless there is precedent to the contrary, I think it should do no stripping at all.  If you can find precedent for some stripping in such a setting, then follow the precedent.  Note that it's not only possible to strip spaces before and/or after the first '=' character, but also before "name" and/or after "x = y".

- Julian
Received on 2012-03-06 13:40:04 CET

This is an archived mail posted to the Subversion Dev mailing list.