couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Benoit Chesneau <>
Subject Re: splitting rcouch in multiple repo
Date Tue, 11 Feb 2014 10:43:48 GMT
On Thu, Feb 6, 2014 at 6:44 PM, Paul Davis <> wrote:
> I've managed to split out the following rcouch applications:
> couch_index
> couch_mrview
> couch_replicator
> couch_collate
> couch_httpd
> couch
> And I've pushed them to their respective repositories on branches
> named import-rcouch.
> This is the procedure I used to do the extract:
> Hooray merges!
> On Thu, Feb 6, 2014 at 3:49 AM, Benoit Chesneau <> wrote:
>> On Wed, Feb 5, 2014 at 10:13 PM, Paul Davis <>wrote:
>>> So this turned into more of a thing than I thought it'd be though I
>>> have at least gotten couch_index split out onto a branch I think. I
>>> need to do some diff's to make sure that its identical and then have
>>> Benoit read the history to make sure its all sane. My notes on the
>>> requisite git gymnastics are here:
>>> I'll do more work on this tonight or tomorrow morning but I'm off line
>>> for the rest of the afternoon today.
>> Yup the move from apps/ to src/ then back to apps/ was unfortunate . i can
>> eventually rebase and provide a new branch at
>> 941a8236c33a248b77abc8fa546a082d08c18bd9
>> not sure it's interresting for all to have all the steps. (though the slip
>> of couch_httpd may be interresting...
>> - benoit

Checked this morning.

Apparently all repos are OK except the couch_mrview one . The tests
09- and 10- has disappeared from it. I am trying to see if i can
create a new branch with them without losing the history.

I didn't launched the tests yet. I have to check actually how you are
launching the tests in the bigcouch branch now. Maybe I could revisit
that in the branch to follow what you did. For myself I changed the
test paths so i could launch them simply by using the prove command.

About the tests I am wondering if we shouldn't move test/etap* and the
util to the couch app in a test folder ....

- beoit

View raw message