[PATCH v2] rollback: add a config knob for entirely disabling the command

Johan Björk jbjoerk at gmail.com
Wed May 4 18:20:47 EDT 2016


Did the ui.progressive patch ever get accepted? If so this sounds like a
perfect candidate.

/Johan

On Wednesday, May 4, 2016, Sean Farley <sean at farley.io> wrote:

>
> Gregory Szorc <gregory.szorc at gmail.com <javascript:;>> writes:
>
> > On May 4, 2016, at 06:56, Kevin Bullock <
> kbullock+mercurial at ringworld.org <javascript:;>> wrote:
> >
> >>> On May 4, 2016, at 01:02, Gregory Szorc <gregory.szorc at gmail.com
> <javascript:;>> wrote:
> >>>
> >>> 'hg rollback' is already marked as deprecated. Can we rename it
> (debugrollback?), remove it entirely, require a config option to enable it,
> or move it to an extension?
> >>
> >> Remember that "deprecated" in Mercurial doesn't mean "slated for
> removal", it means "not recommended or supported but will continue to work
> as it did". I think having a knob to disable it is the best we can do.
> >
> > I know. It's just that this command is a massive footgun that when fired
> can lead to permanent data loss. That violates a core principle of VCS
> tools: don't lose my data. I feel that's a bigger concern than BC.
>
> Yeah, I agree with Greg here. We have the (BC) flag for things that are
> really bad like this. We've even done it before in the past with 'strip
> -b' (a flag that was confused for '--branch' but actually meant 'do not
> create a backup file).
> _______________________________________________
> Mercurial-devel mailing list
> Mercurial-devel at mercurial-scm.org <javascript:;>
> https://www.mercurial-scm.org/mailman/listinfo/mercurial-devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.mercurial-scm.org/pipermail/mercurial-devel/attachments/20160504/67ee01cc/attachment.html>


More information about the Mercurial-devel mailing list