This mail is an automated notification from the task tracker
of the project: Savane.
/**************************************************************************/
[task #777] Latest Modifications:
Changes by:
Mathieu Roy <[EMAIL PROTECTED]>
'Date:
mer 22.09.2004 à 11:42 (Europe/Paris)
------------------ Additional Follow-up Comments ----------------------------
[(Error - Not Found)]
Note: to remove annoying values, use
savane/update/1.0.4/delete-field-status-leftover.sql
/**************************************************************************/
[task #777] Full Item Snapshot:
URL: <http://gna.org/task/?func=detailitem&item_id=777>
Project: Savane
Submitted by: Mathieu Roy
On: mer 22.09.2004 à 10:37
Should Start On: mer 22.09.2004 à 00:00
Should be Finished on: jeu 23.09.2004 à 00:00
Category: Packaging
Priority: 5 - Immediate
Resolution: Cancelled
Privacy: Public
Assigned to: yeupou
Percent Complete: 0%
Status: Closed
Effort: 0.00
Planned Release:
Summary: update script forbidden unplanned use of Status field
Original Submission: Last year, it was decided that the Status field would
only be a boolean for Open/Closed.
Now, we realized that old installation that migrates from pre-Savane havent got
the Status field values different from Open/Closed removed. This is true for
LCG Savannah and GNU Savannah.
Usage of field values different from Open/Closed does not make the trackers
unusable by break the logic of many things.
A script must be written to make this change smoothly for users.
What remained to be determined is:
- whether we map usage of Status to Resolution
- whether we do not map resolution and just update
the field status in the way it makes sense
- whether we copy Status into a custom field
In all case, we'll have to 'update the field status in the way it makes sense'.
We need some user input from CERN to make a decision. We'd like to hear also
some info about the case at Savannah -- are they many groups using the field
Status in this way? What's the way to go in your opinion?
Anyway, a decision will have to made today -- but in the worse case, it will
still possible to modify the script we will provide.
This script will be included in update/1.0.4 but wont have any effect on
installation made since Savane is release -- more precisely, since the 2003
CERN branch. For instance, Gna! is unaffected by this problem.
Regards,
Commentaires
------------------
-------------------------------------------------------
Date: mer 22.09.2004 à 11:42 By: Mathieu Roy <yeupou>
Note: to remove annoying values, use
savane/update/1.0.4/delete-field-status-leftover.sql
-------------------------------------------------------
Date: mer 22.09.2004 à 11:41 By: Mathieu Roy <yeupou>
The decision made is to warn the few groups in this case and to remove the
annoying values that are default for the trackers (group_id='100').
Groups using this field will be warned that soon this values will no longer
work at all (currently, there's more than 5 serious issues it could creates)
and are asked to tell how the migration of this old data could be done for them.
I suggest you do the same at GNU Savannah.
-------------------------------------------------------
Date: mer 22.09.2004 à 11:10 By: Mathieu Roy <yeupou>
Hello,
Can you tell us how many groups are really using this status wrong values at
GNU Savannah? At CERN, only one groups is in this case, so we won't write a
script just for one group, it is faster to do the job at hand.
CC List
-------
CC Address | Comment
------------------------------------+-----------------------------
ype |
beuc |
beu |
For detailed info, follow this link:
<http://gna.org/task/?func=detailitem&item_id=777>
_______________________________________________
Message posté via/par Gna!
http://gna.org/
