incubator-ooo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <gst...@gmail.com>
Subject Re: DEV300 vs OOO340 (was: fetch-all-cws.sh)
Date Thu, 07 Jul 2011 18:33:11 GMT
On Thu, Jul 7, 2011 at 13:57, Mathias Bauer <Mathias_Bauer@gmx.net> wrote:
> Hi Greg,
>
> On 07.07.2011 04:09, Greg Stein wrote:
>
>> On Tue, Jul 5, 2011 at 05:20, Mathias Bauer <Mathias_Bauer@gmx.net> wrote:
>>>...
>>>> (guess the fact that yours is smaller is caused by me having used DEV300
>>>> and you OOO340, so some CWSes that are empty for you have content for me)
>>>
>>> Indeed I got more empty cws because I checked against OOO340. As this is
>>> the version we want to use, I updated cws-list.txt accordingly.
>>
>> I wanted to highlight this particular fact.
>>
>> Is there consensus that we will build the single Hg repository based
>> on OOO340? (with separate bookmarks for all CWSs, pulled against that
>> tag)
>>
>> Previously, I had thought the consensus was DEV300_m106. I believe the
>> choice here implies what "trunk" will end up as: DEV300 or OOO340. I
>> thought that I'd heard there was work completed on DEV300 that is
>> *not* part of OOO340. If that is true, then what should we do that
>> work?
>>
>> I have no opinion, but the fetch-all-cws.sh script is
>> written/documented to work from a DEV300 repository. That will need to
>> be updated.
>>
>>
>> Also, to clarify: OOO340 is the hg.services.openoffice.org/OOO340
>> repository? And that I can get OOO340 by using the DEV300.hg bundle,
>> then pulling from that repository?
>
> does that mean that you are working on that and - even more important -
> also on the generation of the svn dump?
>
> Just to avoid work duplication!

I've been trying to focus us on the scripts so that all of us can experiment.

I'll get them updated for OOO340 as a base, build a single Hg, and
publish that (as a bundle) on people.apache.org for others to review.
Of course, you can also review the output of the scripts on your own,
since that is precisely how I'll construct the single repository.

After that, I'll start trying the svn conversion. I think we're going
to need to fix up the conversion tool to properly handle the
tags/bookmarks and turn them into branches. Thankfully, I have Python
Fu for that, but I don't know that codebase; it will take a while. I
have a friend that works on Hg, so we can probably get those changes
moved upstream pretty easily.

Again, my goal is to get all of this scripted. That will allow
everybody to do this in parallel until we deem the process "correct".
At that point, then sure: I'll do the "final conversion" and pass that
to Infrastructure.

How does that sound?

Cheers,
-g

Mime
View raw message