DVCS comparison for our organisation: decided

Daniel Pittman daniel at rimspace.net
Tue Oct 16 11:35:46 BST 2007


Ben Finney <bignose+hates-spam at benfinney.id.au> writes:
> Ian Clatworthy <ian.clatworthy at internode.on.net> writes:
>> Ben Finney wrote:
>> > Our organisation has chosen Mercurial, based largely on
>> > more-mature and better-documented support for repository
>> > commit-time actions.
>> 
>> We did add some improvements to the hooks doc in 0.91 but they
>> weren't rolled out well enough in two ways:
>> 
>> * it looks like 'latest' is still pointing to the 0.90 docs
>> * the hooks.txt ReST source never got converted to html on the web site.
>> 
>> I assume you found the information in the User Guide OK
>
> I was only able to find this past the deadline, which was enough of a
> black mark against Bazaar's documentation that it wasn't considered.

For what it is worth I had previously looked at that specific
documentation; it didn't really touch on getting the "smart server"
integrate with the hooks well.

[...]

>> 2. Are there pre-packaged scripts we ought to provide for common
>>    hook tasks rather than leaving it up to users to provide the code?
>>    That sounds easy to do, either directly in the code base or on
>>    a wiki page.
>
> The ones that come to mind are:
>
>   * run an arbitrary shell command pre-commit, and reject the commit
>     if that command has a non-zero exit code
>
>     This would be to allow 'make test' or whatever shell command is
>     needed to run a test suite.

As noted elsewhere this is potentially painful for longer running
commits but useful when, for example, we need to enforce a "no spaces in
filenames" rule.

[...]

>> We need to get the core Bazaar product right regardless. Any
>> additional feedback you can provide us to make sure Bazaar Just
>> Works would be much appreciated.
>
> I believe there is work currently underway to improve Bazaar's startup
> time? We didn't measure it, but the perceived difference in startup
> time between Mercurial and Bazaar was quite marked (in Mercurial's
> favour) for a simple cold (i.e. nothing cached) startup of the tool.

Actually, it lost in the lowest end hot cache case as well; cold cache
on my machine (1.7GHz Pentium-M, 2 x 5200RPM HDD RAID1) was 0.845s vs
4.307s respectively.  Cache-hot was 0.285s vs 0.637s.

Those numbers are stable to around .001s on my system and Bazaar was run
second so probably got some benefit from Python being loaded by
Mercurial in the cold-cache situation.

[...]

> I hope this is of some help.

I also expect that we will be putting the results of our assessments
online in some public format -- and dropping a note about it to the
relevant lists -- since it does help everyone.


I also want to note that I am very surprised that Mercurial actually won
our assessment; I had expected Bazaar, which I found generally pleasant
(if a fraction sluggish on startup) to use, or Git to be the winning
choice.

Regards,
        Daniel
-- 
Daniel Pittman <daniel at cybersource.com.au>           Phone: 03 9621 2377
Level 4, 10 Queen St, Melbourne             Web: http://www.cyber.com.au
Cybersource: Australia's Leading Linux and Open Source Solutions Company




More information about the bazaar mailing list