Change Request: fix a fas2 vs fas1 regression

[Toshio Kuratomi]

Ricky Zhou wrote:
> On 2008-05-01 06:24:32 PM, Toshio Kuratomi wrote:
>> I've just committed this change to the fas git repo that fixes a change in 
>> the dump format from fas1 to fas2.  In fas1, the role_type was accurate 
>> when dumped from the account system.  In fas2, role_type was always "user" 
>> even if the person was a sponsor or administrator.
>>
>> I'd like to have this applied to the production FAS so that the Package 
>> Status Scripts can be run again.
> +1 for the role_type fix
> (If the last field, the number of people sponsored, isn't vital to those
> scripts, then we can fix that last omissions after the freeze.)
> 
Shoot.  I didn't know that field was broken as well.  Am I correct that 
that's the number of people that the given person has sponsored?

If so, here's a new version of the patch that fixes that as well 
(committed to fas's git).  It makes two db queries inside of the method 
body[1]_ per request.  I'm pretty sure that it could be made into a 
single query per request but I don't think that's a necessary change at 
the moment.

This has been tested on pt3 for plain text and json calls, with a 
groupname and without a groupname specified.  It imports a few methods 
from sqlalchemy but otherwise it only touches the dump method and its 
template just like the last patch.

.. _[1]: There are additional queries that we don't have much control 
over for authentication and such.

-Toshio
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fas2-dump-role-sponsor.patch
Type: text/x-patch
Size: 3608 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/fedora-infrastructure-list/attachments/20080502/3f5af30a/attachment.bin>