cassandra-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From aaron morton <>
Subject Re: Reoganizing drivers
Date Mon, 20 Jun 2011 04:26:08 GMT
I can see the drivers have moved to

Just wondering where that path is available on git://

These are the remote branches I can find 

$ git ls-remote  | grep drivers
From git://
20635cec24389d83b146af51fa902fcf2d21491b	refs/remotes/tags/drivers
dd06878fa6b143dbff1e1e338087041b1b230d48	refs/tags/drivers
20635cec24389d83b146af51fa902fcf2d21491b	refs/tags/drivers^{}


Aaron Morton
Freelance Cassandra Developer

On 8 Jun 2011, at 05:01, Eric Evans wrote:

> On Tue, 2011-06-07 at 18:40 +0200, Sylvain Lebresne wrote:
>> On Tue, Jun 7, 2011 at 3:18 PM, Jonathan Ellis <>
>> wrote:
>>> Sounds fine as far as it goes, but don't we want some concept of
>>> branches/tags for driver releases too?
>> Our idea so far (Eric can correct me if I'm wrong :)) was to consider
>> the drivers directory as the 'trunk' for drivers, and create branches
>> and tags for them alongside the cassandra ones.
> Yup.  In fact, I already tagged the Python and Java drivers as
> tags/drivers/<lang>/<version> during the last release (neither of those
> driver artifacts corresponded to the same SVN rev, nor did they
> correspond to the rev for 0.8.0).
>> Truth is, I even think that consider the drivers as a whole is not
>> granular enough. It's unlikely the different drivers will move at the
>> same pace.
> As far as I know, there is no reason that a tag (say
> tags/drivers/py/1.1.1) can't point to a subdirectory of drivers/ (i.e.
> drivers/py).  In fact, that's how the tags mentioned above were done
> (except those pointed to branches/cassandra-0.8.0/drivers/<lang>).  I
> think it just boils down to a matter convention.
>> *But*, we believe that moving the drivers up one level is at least a
>> first step towards something better than the status quo.
> Yeah, even if we decide to do something different later on, this is an
> improvement over what we have now. 
> -- 
> Eric Evans

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message