hey, reviewboard is pretty cool. i'm wonder about conventions to using
it. the problem is how to correlate the review with the patch. since a
JIRA number may contain multiple patches, we need a way to identify a
particular patch. perhaps we should name our patches
ZOOKEEPER-<JIRA#>_<revision#>, then we can use the patch name when we
create a review board request. what do you all think?

thanx
ben

Hey Ben, if we have multiple patches in the same jira, perhaps we  
should use subtasks and name the review requests accordingly? Does it  
work for the cases you're referring to?

I was wondering, though, what other projects do to maintain the jira  
consistent with respect to discussion on the review board, if  
anything. We should at least post the link to the review board request  
so that we have it documented.

-Flavio

On Jan 9, 2011, at 7:47 AM, Benjamin Reed wrote:

> hey, reviewboard is pretty cool. i'm wonder about conventions to using
> it. the problem is how to correlate the review with the patch. since a
> JIRA number may contain multiple patches, we need a way to identify a
> particular patch. perhaps we should name our patches
> ZOOKEEPER-<JIRA#>_<revision#>, then we can use the patch name when we
> create a review board request. what do you all think?
>
> thanx
> ben

flavio
junqueira

research scientist

[EMAIL PROTECTED]
direct +34 93-183-8828

avinguda diagonal 177, 8th floor, barcelona, 08018, es
phone (408) 349 3300    fax (408) 349 3301

i'm talking about iterations of the same patch. if you review a patch
and the contributor updates the patch and you want to review it again,
how do you correlate the reviews with the different iterations of the
same patch?

ben

On 01/11/2011 01:51 PM, Flavio Junqueira wrote:
> Hey Ben, if we have multiple patches in the same jira, perhaps we
> should use subtasks and name the review requests accordingly? Does it
> work for the cases you're referring to?
>
> I was wondering, though, what other projects do to maintain the jira
> consistent with respect to discussion on the review board, if
> anything. We should at least post the link to the review board request
> so that we have it documented.
>
> -Flavio
>
> On Jan 9, 2011, at 7:47 AM, Benjamin Reed wrote:
>
>> hey, reviewboard is pretty cool. i'm wonder about conventions to using
>> it. the problem is how to correlate the review with the patch. since a
>> JIRA number may contain multiple patches, we need a way to identify a
>> particular patch. perhaps we should name our patches
>> ZOOKEEPER-<JIRA#>_<revision#>, then we can use the patch name when we
>> create a review board request. what do you all think?
>>
>> thanx
>> ben
>
> *flavio*
> *junqueira*
>
> research scientist
>
> [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>
> direct +34 93-183-8828
>
> avinguda diagonal 177, 8th floor, barcelona, 08018, es
> phone (408) 349 3300fax (408) 349 3301
>
>

Here's an RB example with multiple (6) patch revisions:
https://review.cloudera.org/r/402/

I suspect we'd have to do something like you (ben) suggest - use rev#
in the patch name (jira) and
reference that name in RB when adding the description for the updated
diff. It's ugly given the
lack of integration btw rb and jira.

I wonder how the other teams are handling this (hbase? ccing Todd).

Patrick

On Tue, Jan 11, 2011 at 2:30 PM, Benjamin Reed <[EMAIL PROTECTED]> wrote:
> i'm talking about iterations of the same patch. if you review a patch and
> the contributor updates the patch and you want to review it again, how do
> you correlate the reviews with the different iterations of the same patch?
>
> ben
>
> On 01/11/2011 01:51 PM, Flavio Junqueira wrote:
>>
>> Hey Ben, if we have multiple patches in the same jira, perhaps we should
>> use subtasks and name the review requests accordingly? Does it work for the
>> cases you're referring to?
>>
>> I was wondering, though, what other projects do to maintain the jira
>> consistent with respect to discussion on the review board, if anything. We
>> should at least post the link to the review board request so that we have it
>> documented.
>>
>> -Flavio
>>
>> On Jan 9, 2011, at 7:47 AM, Benjamin Reed wrote:
>>
>>> hey, reviewboard is pretty cool. i'm wonder about conventions to using
>>> it. the problem is how to correlate the review with the patch. since a
>>> JIRA number may contain multiple patches, we need a way to identify a
>>> particular patch. perhaps we should name our patches
>>> ZOOKEEPER-<JIRA#>_<revision#>, then we can use the patch name when we
>>> create a review board request. what do you all think?
>>>
>>> thanx
>>> ben
>>
>> *flavio*
>> *junqueira*
>>
>> research scientist
>>
>> [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>
>> direct +34 93-183-8828
>>
>> avinguda diagonal 177, 8th floor, barcelona, 08018, es
>> phone (408) 349 3300fax (408) 349 3301
>>
>>
>
>

On reviewboard you can see the diff between patches. Is this the kind of
feature you were looking for?

On Tue, Jan 11, 2011 at 5:30 PM, Benjamin Reed <[EMAIL PROTECTED]> wrote:

> i'm talking about iterations of the same patch. if you review a patch and
> the contributor updates the patch and you want to review it again, how do
> you correlate the reviews with the different iterations of the same patch?
>
> ben
>
>
> On 01/11/2011 01:51 PM, Flavio Junqueira wrote:
>
>> Hey Ben, if we have multiple patches in the same jira, perhaps we should
>> use subtasks and name the review requests accordingly? Does it work for the
>> cases you're referring to?
>>
>> I was wondering, though, what other projects do to maintain the jira
>> consistent with respect to discussion on the review board, if anything. We
>> should at least post the link to the review board request so that we have it
>> documented.
>>
>> -Flavio
>>
>> On Jan 9, 2011, at 7:47 AM, Benjamin Reed wrote:
>>
>>  hey, reviewboard is pretty cool. i'm wonder about conventions to using
>>> it. the problem is how to correlate the review with the patch. since a
>>> JIRA number may contain multiple patches, we need a way to identify a
>>> particular patch. perhaps we should name our patches
>>> ZOOKEEPER-<JIRA#>_<revision#>, then we can use the patch name when we
>>> create a review board request. what do you all think?
>>>
>>> thanx
>>> ben
>>>
>>
>> *flavio*
>> *junqueira*
>>
>> research scientist
>>
>> [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>
>>
>> direct +34 93-183-8828
>>
>> avinguda diagonal 177, 8th floor, barcelona, 08018, es
>> phone (408) 349 3300fax (408) 349 3301
>>
>>
>>
>