After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 137544 - Task priority spinner not effective.
Task priority spinner not effective.
Status: RESOLVED FIXED
Product: planner
Classification: Other
Component: General
unspecified
Other Linux
: Normal trivial
: 0.12
Assigned To: Lincoln Phipps
planner-maint
Depends on:
Blocks:
 
 
Reported: 2004-03-17 23:53 UTC by Lincoln Phipps
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: 2.3/2.4


Attachments
Standalone patch to fix this one bugzilla issue. (16.33 KB, patch)
2004-04-06 00:31 UTC, Lincoln Phipps
none Details | Review

Description Lincoln Phipps 2004-03-17 23:53:44 UTC
The task priority spinner is not effective at all. Looking at the code I
can see why !. It hasn't got anything ;) nor is there anything in the schema.

Easy for me to add this as I've already changed schema for the short-name
field so know what to do to make this work for XML and SQL.

We should make this work because people using Planner currently have to use
 something else to do workload balancing/levelling. The priority field
would be expected to be used by these non-Planner tools to allow you to
define which task loses out when there is contention for resources.

Potentially Planner's own future (hypothetically speaking)
leveller/scheduler would also need such a  priority field. Until Planner
gets this, we need a way of signalling how to manage contention to other
projects: the priority field is the thing to use at this time.

The range should be Default = 1, range = 0, 1...integermax. Planner will
not apply any meaning to any value at this time i.e. 0,1 or 9999 have no
meaning to Planner.

Rgds,
Lincoln.
Comment 1 Richard Hult 2004-03-18 00:09:57 UTC
Sounds like a good plan :)
Comment 2 Lincoln Phipps 2004-04-06 00:31:19 UTC
Created attachment 26383 [details] [review]
Standalone patch to fix this one bugzilla issue.

OK - I've given in and will chop my patches into smaller
things. I've just got a lot more familiar with using
patch to reverse patches. It works very well.

Attached is the Task priority patch. This fixes,

http://bugzilla.gnome.org/show_bug.cgi?id=137544
(Task priority spinner not effective.)

The priority spinner is now effective and gets stored.
It was always displayed but never used (I guess never
worked in MrProject or Planner). Range is quite
arbitarily set to 0..9999. In the future I'd expect
0 to imply no priority defined and 1..9999 to mean a
priority. FYI: MS Project 2000 has priority values
from 1 to 1000 so we're able to map to/from that as well
as potentially have a finer grain when within our own
environment.

As to wether 1 is high or low is up to the user at this
time; planner doesn't use this in any scheduling.

WARNING: this changes schema. An old Planner will
not read any new files which have "priority" in
them (This isn't a bug but how Planner works in its
backward compatibility stuff).

Why is this important ? - well Planner doesn't have
an internal load leveller but there are a number of
external ones and people have expressed some interest
in external scheduling. You could use the priority
field to help tweak how these 3rd party rank tasks
for scheduling.

Note as mentioned: Planner doesn't yet use this field.
Its for discussion - but planner could use this field
in the future e.g. if you have TASK 1,2,3 and TASK2 and
3 are both linked to TASK 1 as FS and both use the same
set of resources which are now at MAX (100%) then the
"higher" priority task out of 2 or 3 should take precedence
over the "lower" priority task. But as we know this
isn't as easy as it sounds but having a priority
field is certainly a very helpful hint to any scheduler
as its defined by the Project Manager (human) and they
are the best arbitrator of precedence in the end.

Rgds,
Lincoln.
Comment 3 Richard Hult 2004-04-06 20:38:34 UTC
Thanks a lot, committed!