aMule Bug Tracker - aMule
View Issue Details
IDProjectCategoryView StatusDate SubmittedLast Update
0000758aMuleMiscpublic2006-01-06 12:452006-01-07 04:00
Reporterlao 
Assigned ToXaignar 
PrioritynormalSeveritytrivialReproducibilityalways
StatusassignedResolutionopen 
PlatformOSOS Version
Product Version 
Target VersionFixed in Version 
Fixed in Revision
Operating System
Summary0000758: Sorting by multi-valued columns
DescriptionExamples of multi-valued columns would be the Sources column in the Downloads view, the Requests and Transferred Data column in the Shared Files view, etc.

They usually keep track of a current and a total count. Sorting should treat this value as a tie-breaker, akin to the multi-column sorting already implemented.

Presumably, this should yield to sorting by a different column, like this:

1a. Click on Column1 (single value : C1a)
2a. Click on Column2 (single value : C2a)

Now C2a is used for the primary sort, and C1a to break any ties.

1b. Click on Column1 (single value : C1a)
2b. Click on Column2 (two values : C2a, C2b)

Now C2a should be used for the primary sort, any ties should be resolved by C1a, and after that any remaining ties should be resolved by C2b.

That is Column1=Priority / Column2=Requests would produce the following:

File3 (Normal) (1/15)
File1 (High) (1/10)
File2 (High) (1/21)
File6 (High) (1/22)
File4 (Normal) (2/22)
File5 (High) (2/3)

Changing Column2=Requests to descending should either ignore the total count (keep it ascending) or, simpler, sort both descending, like this:

File4 (Normal) (2/22)
File5 (High) (2/3)
File3 (Normal) (1/15)
File6 (High) (1/22)
File2 (High) (1/21)
File1 (High) (1/10)


And similiarily for two multi-valued columns.
Steps To Reproduce
Additional Information
TagsNo tags attached.
Relationships
Attached Files
Issue History
Date ModifiedUsernameFieldChange
2006-01-06 12:45laoNew Issue
2006-01-06 16:27laoNote Added: 0001728
2006-01-07 04:00KryStatusnew => assigned
2006-01-07 04:00KryAssigned To => Xaignar

Notes
(0001728)
lao   
2006-01-06 16:27   
Alternatively, it could of course also sort first by C2a, then by C2b and then by C1a.

Then we'd get:

File1 (High) (1/10)
File3 (Normal) (1/15)
File2 (High) (1/21)
File6 (High) (1/22)
File5 (High) (2/3)
File4 (Normal) (2/22)

Possibly that's slightly more useful, though I'm not sure.

It's really just a cosmetic issue.