Re: Re: [Monotone-devel] net.venge.monotone.workspace-merge.migration ready for review

[Daniel THOMPSON]

Nathaniel Smith wrote:
> I know it's handy having _MTN/revision be usable for shell scripts...
> maybe we should keep _MTN/revision as it is for now, but as a totally
> cosmetic file (i.e., monotone writes it, but never reads it), and

I don't know about anyone else but I find the facility to alter the
revision very helpful.

One obvious example is to 'rescue' a built workspace if I perform a
kill_rev_locally without first updating the workspace to a different
revision.

A less common example is after database rebuild an existing workspace
can be made to work again.

I also find myself playing with it a lot when hunting for regressions.
This is, in fact, more to workaround the fact that meld can't compare
against historic revisions. Clearly rather than hold back monotone it
would be better improve meld.

Basically the facility to change revision id without changing the
workspace does find use with some of us (anyone else use it for anything
useful?). However such functionality could easily become a flag for 'mtn
update' providing it didn't whine about 'abuse' if the original revision
id doesn't exist.

-- 
Daniel Thompson (STMicroelectronics) <address@hidden>
1000 Aztec West, Almondsbury, Bristol, BS32 4SQ. 01454 462659

If a car is a horseless carriage then is a motorcycle a horseless horse?