Re: [Gnu-arch-users] "tla commit" generates a patch-set even if there are no changes

[Aaron Bentley]

Julian T. J. Midgley wrote:
> > Perhaps you could explain why the defaults are unwise?  What's the harm 
> > in an accidental empty commit, aside from a little embarassment?
>
>
> Conversely, perhaps you could explain the benefits?  So far, I see
> none (provided --force is available to allow the commit when it is
> needed).

The harm comes from changing the default behavior.  I have scripts that 
I run from cron that would be broken by that change.  tla isn't very 
user-friendly.  Scripts can make it smoother.  Let's not make tla 
script-unfriendly too.

> Why not go the whole hog and permit people to commit with empty log files?

tla does permit people to commit with empty log files.  Try "tla 
make-log; tla commit".

The case where it doesn't permit empty log files is the case where you 
commit with no log file at all.  There are two obvious cases why you 
might commit with no log:

1. You want commit to run $($EDITOR $(tla make-log)) for you.
2. You invoked commit in the wrong directory.

In the second case, the user can abort the process when they see the 
empty log by quitting.  Othewise, they're stuck.

Aaron

--
Aaron Bentley
Director of Technology
Panometrics, Inc.