Re: [Quilt-dev] [PATCH] mail-single-patch.diff

[John Vandenberg]

On 9/18/05, Gary V. Vaughan <address@hidden> wrote:
> Hi John,
> 
> John Vandenberg wrote:
> > On 9/18/05, Gary V. Vaughan <address@hidden> wrote:
> >
> >>This patch allows calling quilt mail with a list of patch names
> >>after the usual arguments, and mails only those patches rather
> >>than the entire series.  If only a single patch is named then it
> >>is put into the bady of a single mail (that's what I'm using to
> >>send this patch)
> >
> >
> > Hi Gary,
> >
> > This patch makes `mail' the first command to operate on multiple
> > patches outside of the series.  While I would enjoy being able to send
> > a single patch, or a sub-series, I don't think quilt should generate
> > emails that can't be applied on the receiving end.  In other words, in
> > order to send an adhoc set of patches, `mail' should first verify that
> > the dependency graph of the selected patches is sane.
> 
> I disagree: tools shouldn't try to outsmart their users.  If I have
> 100 patches in my series that I posted yesterday, and I've just revised
> patch 100 as a result of feedback and want to repost the revision, I
> don't want quilt to post all 100 patches again -- I'm exaggerating but
> you see my point?
> 
> Maybe a warning, or requiring a --force argument is reasonable?

Warnings and --force sounds good.  In order to do introduce
functionality that uses adhoc sets of patches, we need tools to
qualify what an adhoc set of patches is: i.e. is it a series, or a
guarded series, or multiple individual patches that have no
dependencies.  Then we can warn the user appropriately, and they can
use --force in an informed manner.

> Otherwise, why require users to go around quilt and paste the patches
> they need to repost into their regular MUA?

I am not suggesting that quilt mail shouldn't tackle this task.  It
should.  But please, when it does, it should maintain the sanity of
the series it sends.  IMO, it should not revert to being just another
script that can email files, in the hope that eventually someone will
add the smarts later.

> As it stands mail is
> mostly useless... all I can do is post my whole series once.  And
> repost the entire series again when I've added a new patch to it that
> I want to open for peer review...

I have no reservations about quilt mail sending a single patch.  For
multiple patches, why not author a new/temporary series file, verify
it works, and then use quilt mail.  This could be combined together
into a single operation.
--
John