"submit" command naming - just "bundle" preferred?

[Aaron Bentley]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ian Clatworthy wrote:
> Aaron Bentley wrote:
> 
>>> How about
>>>  send -o OUTPUTFILE
>> Given a send command that actually sends mail, this is pretty good.  -o
>> will be rarely used, for sure.
> 
> FWIW, I use "bzr bundle | less" quite a lot. It gives me the total set
> of changes committed in a feature/fix branch. I gather I could get the
> same from some combination of switches to diff?

Yes.  diff -r submit:..-1

> "bzr bundle | less" is
> faster to type though and fast enough as an operation that I'm happy
> right now using it.

I use this alias:
sdiff=cdiff --check-style -r submit:

Compared with bundle, that gives pretty colors, style checks and it
shows uncommitted changes (note the lack of "..-1"), and no base64 junk.
 And it's faster.

> After it gets through review, I switch to my integration branch, merge
> it there, push to the public mirror of my integration branch and
> pqm-submit.

This is because you only have a single public branch?

> After pqm succeeds, I delete the original feature/fix
> branch. My patches directory therefore ends up being my "personal
> archive" of changes submitted.

Deleting branches and keeping bundles seems odd to me.

> To clarify the outcomes so far, "submit" will be renamed "send", bundle
> will be an alias and we'll continue discussing how best to configure
> send to do the actual email/submission bit? Or is submit staying and
> send is only the latter?

I'm loathe to call it "send" before we make it possible to send.

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGoDE20F+nu1YWqI0RAoVKAJ9O831RYguzWJzbXRdc593g4gMT1QCggE54
1Is3ObuPyXItVdipBnkVZO8=
=3XmM
-----END PGP SIGNATURE-----