On Wed, Jul 29, 2009 at 03:42:24PM +0200, Guido Günther wrote:
> dch has:
>
> if ($mtime == $newmtime && ! $opt_create &&
> (!$opt_r || ($opt_r && $opt_force_save_on_release))) {
>
> warn "$progname: $changelog_path unmodified; exiting.\n";
> exit 0;
> }
>
> So it will abort if we do a release (opt_r) and opt_force_on_release is
> true. Isn't that logic reverted? Shouldn't we save the changelog if
> opt_r is true and opt_force_save_on_release is also true:
$opt_force_save_on_release is an unfortunate name, but this is correct
behavior. When the variable is true, this means that we are requiring
the user to save the changelog (thus updating the mtime) so we know they
have confirmed their changes.
The description in the man page isn't quite clear. How does this sound:
--force-save-on-release
When --release is used and an editor opened to allow inspection of the
changelog, require the user to save the changelog their editor opened.
Otherwise, the original changelog will not be modified. (default)
--
James
GPG Key: 1024D/61326D40 2003-09-02 James Vega <james...@debian.org>
signature.asc
Description: Digital signature
