How about renaming svn:ignore to svn:status-ignore or svn:status_ignore,
or svn:ignore-for-status since that's what it is.
Colin Watson wrote:
> On Thu, Mar 06, 2003 at 10:23:51AM -0500, Tim Moloney wrote:
>
>>Branko ??ibej wrote:
>>
>>>The reason for "svn:ignore" is to keep certain files out of the "?"
>>>output of svn status, and only marginally to help svn import and
>>>recursive svn add. And that's it.
>>
>>As far as I've seen, svn:ignore is only documented for the status
>>subcommand. It doesn't appear to be documented for the either the
>>add or import subcommands.
>>
>>Its use is certainly inconsistent.
>
>
> Why is it inconsistent?
>
> [cjwatson@riva ~]$ svn pg svn:ignore | fgrep .bash_history
> .bash_history
> [cjwatson@riva ~]$ svn st | fgrep .bash_history
> [cjwatson@riva ~]$ svn st .bash_history
> ? .bash_history
>
> This seems pretty much identical to the behaviour of add to me. If you
> don't mention something explicitly and it's in svn:ignore, it's ignored.
> If you do mention it explicitly, then it's processed even though it's in
> svn:ignore. To me this feels like a simple and quite natural rule.
>
>
>>It is non-intuitive that svn:ignore isn't always honored and more
>>non-intuitive that it would be honored by a subcommand but only if a
>>certain switch is used.
>
>
> Actually I find the current behaviour very intuitive. Intuitiveness,
> remember, is a highly relative concept.
>
>
>>I would prefer that we strive for consistent and intuitive software.
>>Since I've read this sentiment stated by others on this list, it really
>>surprises me that several people state strongly that they want to keep
>>svn:ignore's usage this way.
>
>
> That would probably be because intuitiveness is not absolute.
>
> Cheers,
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Mar 6 18:42:57 2003