You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@subversion.apache.org by cm...@collab.net on 2003/06/09 14:47:37 UTC

Re: svn commit: rev 6173 - in trunk: . doc/book doc/book/book doc/translations/french doc/user notes subversion/bindings/java/javahl/native subversion/bindings/java/javahl/src/org/tigris/subversion/javahl subversion/bindings/ruby subversion/clients/cmdline subversion/clients/cmdline/man subversion/include subversion/libsvn_client subversion/libsvn_wc subversion/tests/clients/cmdline subversion/tests/clients/cmdline/getopt_tests_data tools/client-side tools/client-side/psvn tools/dev www

kfogel@tigris.org writes:

> Author: kfogel
> Date: Sun Jun  8 21:13:41 2003
> New Revision: 6173

Dude.  This really doesn't seem like a change based on a real
community consensus.  Can I ask you to revert -- I mean, undo :-) --
this change until something more solid is formed?  

Perhaps we can explore other options, like making revert prompt you
first:

  $ svn revert -R .
  The operation you are about to perform will throw away any changes
  you've made to the target path(s).  Are you sure you want to do
  this? [y/N]

And then add a configuration variable 'revert-no-prompt = no' for a
sort of "expert mode".  Would the list-folk be okay with a solution
like this?

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn commit: rev 6173 - in trunk: . doc/book doc/book/book doc/translations/french doc/user notes subversion/bindings/java/javahl/native subversion/bindings/java/javahl/src/org/tigris/subversion/javahl subversion/bindings/ruby subversion/clients/cmdline subversion/clients/cmdline/man subversion/include subversion/libsvn_client subversion/libsvn_wc subversion/tests/clients/cmdline subversion/tests/clients/cmdline/getopt_tests_data tools/client-side tools/client-side/psvn tools/dev www

Posted by Garrett Rooney <ro...@electricjellyfish.net>.
cmpilato@collab.net wrote:

>kfogel@tigris.org writes:
>
>  
>
>>Author: kfogel
>>Date: Sun Jun  8 21:13:41 2003
>>New Revision: 6173
>>    
>>
>
>Dude.  This really doesn't seem like a change based on a real
>community consensus.  Can I ask you to revert -- I mean, undo :-) --
>this change until something more solid is formed?  
>

I agree, this change should be backed out.

>Perhaps we can explore other options, like making revert prompt you
>first:
>
>  $ svn revert -R .
>  The operation you are about to perform will throw away any changes
>  you've made to the target path(s).  Are you sure you want to do
>  this? [y/N]
>
>And then add a configuration variable 'revert-no-prompt = no' for a
>sort of "expert mode".  Would the list-folk be okay with a solution
>like this?
>

Ick.  I really dislike the idea of starting to add "yes I know what I'm 
doing" prompts.  We've managed to avoid them up to now, and I'd hate to 
see us cave in for this.

-garrett


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn commit: rev 6173 - in trunk: . doc/book doc/book/book doc/translations/french doc/user notes subversion/bindings/java/javahl/native subversion/bindings/java/javahl/src/org/tigris/subversion/javahl subversion/bindings/ruby subversion/clients/cmdlin

Posted by cm...@collab.net.
kfogel@collab.net writes:

> "Sander Striker" <st...@apache.org> writes:
> > > And then add a configuration variable 'revert-no-prompt = no' for a
> > > sort of "expert mode".  Would the list-folk be okay with a solution
> > > like this?
> > 
> > +1!  I've thought about that, but I didn't post because I didn't
> > want to be the poor bugger to suggest an option... ;) :)
> 
> Backed out the change with pleasure, but waaaay -1 on this alternate
> solution.  Yick :-).  What's so bad about just changing the name of
> the "resolve" command instead?  Solves the problem, minimal code
> change, no annoying options...

Nothing so bad about that solution.  Was just offering other another
one for people to toy with so we don't have to revisit this again
later.  Renaming 'resolve' is perfectly fine with me.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

Re: svn commit: rev 6173 - in trunk: . doc/book doc/book/book doc/translations/french doc/user notes subversion/bindings/java/javahl/native subversion/bindings/java/javahl/src/org/tigris/subversion/javahl subversion/bindings/ruby subversion/clients/cmdlin

Posted by kf...@collab.net.
"Sander Striker" <st...@apache.org> writes:
> > And then add a configuration variable 'revert-no-prompt = no' for a
> > sort of "expert mode".  Would the list-folk be okay with a solution
> > like this?
> 
> +1!  I've thought about that, but I didn't post because I didn't
> want to be the poor bugger to suggest an option... ;) :)

Backed out the change with pleasure, but waaaay -1 on this alternate
solution.  Yick :-).  What's so bad about just changing the name of
the "resolve" command instead?  Solves the problem, minimal code
change, no annoying options...


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

RE: svn commit: rev 6173 - in trunk: . doc/book doc/book/book doc/translations/french doc/user notes subversion/bindings/java/javahl/native subversion/bindings/java/javahl/src/org/tigris/subversion/javahl subversion/bindings/ruby subversion/clients/cmdlin

Posted by Sander Striker <st...@apache.org>.
> From: cmpilato@collab.net [mailto:cmpilato@collab.net]
> Sent: Monday, June 09, 2003 4:48 PM

[...]
> Perhaps we can explore other options, like making revert prompt you
> first:
> 
>   $ svn revert -R .
>   The operation you are about to perform will throw away any changes
>   you've made to the target path(s).  Are you sure you want to do
>   this? [y/N]
> 
> And then add a configuration variable 'revert-no-prompt = no' for a
> sort of "expert mode".  Would the list-folk be okay with a solution
> like this?

+1!  I've thought about that, but I didn't post because I didn't
want to be the poor bugger to suggest an option... ;) :)


Sander

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org

RE: svn commit: rev 6173 - in trunk: . doc/book doc/book/book doc/translations/french doc/user notes subversion/bindings/java/javahl/native subversion/bindings/java/javahl/src/org/tigris/subversion/javahl subversion/bindings/ruby subversion/clients/cmdlin

Posted by Steven Brown <sw...@ucsd.edu>.

> -----Original Message-----
> From: cmpilato@collab.net [mailto:cmpilato@collab.net]
> Sent: Monday, June 09, 2003 7:48 AM
> To: Subversion Developers
> Subject: Re: svn commit: rev 6173 - in trunk: . doc/book doc/book/book
> doc/translations/french doc/user notes
> subversion/bindings/java/javahl/native
> subversion/bindings/java/javahl/src/org/tigris/subversion/javahl
> subversion/bindings/ruby subversion/clients/cmdlin
>
>
> kfogel@tigris.org writes:
>
> > Author: kfogel
> > Date: Sun Jun  8 21:13:41 2003
> > New Revision: 6173
>
> Dude.  This really doesn't seem like a change based on a real
> community consensus.  Can I ask you to revert -- I mean, undo :-) --
> this change until something more solid is formed?
>
> Perhaps we can explore other options, like making revert prompt you
> first:
>
>   $ svn revert -R .
>   The operation you are about to perform will throw away any changes
>   you've made to the target path(s).  Are you sure you want to do
>   this? [y/N]
>
> And then add a configuration variable 'revert-no-prompt = no' for a
> sort of "expert mode".  Would the list-folk be okay with a solution
> like this?

I like the idea of revert prompting, like rm -i, and have an override, like
rm -f.  A user-level operation in a source control system that can
permanently destroy data really should confirm by default, as unlike any
other operation, it's not recoverable.

People are going to occasionally mistype/misthink 'revert' and 'resolve' no
matter what you call them (I've seen people type rm instead of cp by
accident) so changing the names doesn't seem like the right solution (and
the names are perfectly defined as-is).  Other than prompting by default,
the other possibility would be to have revert backup files by moving them
out of the way (like foo~, or  #foo# or whatever), but that would probably
work better with CVS than subversion due to revert also dealing with
directories.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org