[PATCH 1 of 2] mq: add qfreeze command to protect patches from careless modification

[FUJIWARA Katsunori]

At Sun, 7 Jun 2009 18:10:42 +0300,
timeless wrote:
> 
> Note: I'm not commenting on whether this is the right thing to do,
> merely the English. (Which sadly also has typos which a spell checker
> could have caught.)

Hi, timeless. Thank you for your detailed comments.

> Personally, I find freeze --release to be unintuitive.

ok, I will mention it later in this mail.

> > +    Frozen patch prevents qrefresh, qdelete, qfold and qfinish from
> > +    careless modifiction. qrename is allowed, because it does not
> > +    modify content.
> 
> Use qfreeze to prevent qrefresh, qdelete, qfold and qfinish commands
> from accidentally acting upon a patch.
> Patches that are frozen can be renamed with qrename because it does
> not change the patch's content.

Thank you for your correction.

> > +    If none of '--series', '--applied', '--unapplied' nor patch name is
> > +    specified, qfreeze treats 'qtop' as a target patch.
> 
> 'nor' isn't appropriate here.

is "and" appropriate ? sorry for my poor English...

> > +    With --release option, this command releases frozen status of target
> > +    patches.
> 
> omit 'option'

ok.

> > +    With --list option, this command shows all frozen patches.
> 
> omit 'option'

ok.

> > +    # v(erbosity from implicit target specification)
> > +    v = ((opts['series'] and 1 or 0) +
> > +         (opts['applied'] and 1 or 0) +
> > +         (opts['unapplied'] and 1 or 0))
> > +    # c(ombination of target specification)
> > +    c = (v + (patches and 1 or 0))
> 
> this seems awfully clever....

thanks :-)

> > +            raise util.Abort(_("-l cannot accept patch specification"))
> 
> cannot is strange.

is "does not" appropriate ? or is "invalid combition with -l" better ?

> > +            raise util.Abort(_("invalid combination of target specification"))
> 
> odd

Please let me confirm your suggestion. Which would you means by "odd":

  1. this message is redundant = should be shorten
  2. this checking is redundant = such combination should be acceptable
  3. other

> personally, i'd probably use:
> 
> hg qfrozen --freeze patch
> hg qfrozen --thaw patch
> hg qfrozen --list

Oh, adoption of 'frozen' as command name is out of my English ability !
Thank you for your suggestion.

You assume that the message such as:

    one of --freeze/--thaw/--list should be specified

is shown, when none of them is specified, do not you ?

----------------------------------------------------------------------
[FUJIWARA Katsunori]      fujiwara at ascade.co.jp(foozy at lares.dti.ne.jp)