[Monotone-devel] Re: Monotone-devel Digest, Vol 24, Issue 8

[jc007b6093]

>On Tue, Apr 05, 2005 at 06:05:11PM -0500, Matthew A. Nicholson wrote:
>> Nathaniel Smith wrote:
>> >As for "status" and "commit" taking that long, I assume it's because
>> >we currently read the entire working copy off of disk scanning for
>> >changes, and you're waiting on IO.  It's entirely possible, even
>> >straightforward, to fix this, and let people use something more like
>> >CVS's timestamps to detect differences; just someone needs to write
>> >it...
>>
>> But please don't make any timestamp stuff the default (at least not like
>> make).
>
>Of course not!
>
>How about:
>
>  function working_dir_change_detection_mode()
>    return "reckless"
>  end
>
>?  ;-)
>
>-- Nathaniel

How about providing an option to 'prune' the search against parts of the
working tree which are unlikely to have changed? E.g. monotree has the
Batick and Gxl libraries in the tree along with all their javadoc - this
will never change until a new version is imported. It would also be good
for directories which only ever contain derived objects.

Another alternative is to combine the two and be able to set the detection
algorithm for each sub-tree.

Joel