Getting a window saying the data (I'm viewing?) has been clustered before when it is not clustered

[Robert Leach]

USE CASE: WHAT DO YOU WANT TO DO?

Cluster data that is currently unclustered

STEPS TO REPRODUCE AN ISSUE (OR TRIGGER A NEW FEATURE)

Scenario 1:

1. Open an unclustered file (that has been clustered before, but was saved as a separate file... e.g. small_133x133.txt, which was saved as something like small_133x133/small_133x133_average_1.cdt and associated files)
2. Cluster it

Scenario 2:

1. In your file system, delete the directory containing the type of clustered files that you are going to re-cluster (e.g. the "complete" directory such as "small_133x133/complete"). Note: I still had an "average" directory.
2. Open the original unclustered data file
3. Cluster it

CURRENT BEHAVIOR

In each case, you get a warning that the data has been clustered before and asks whether they're sure they want to cluster:

EXPECTED BEHAVIOR

I see 1 of 3 possible behaviors I might expect:

1. Only provide the warning if the data they're looking at hasn't been clustered (with the options they've selected)
2. Never provide a warning and simply perform the clustering without confirmation.
3. If we're going to provide a warning, and they're looking at unclustered data, tell the user in the warning message where the clustered files are located so they can open them instead of recluster - better yet, give them the option in the warning window to select one of those files to open (and allow them to recluster despite the available files).

DEVELOPERS ONLY SECTION

SUGGESTED CHANGE (Pseudocode optional)

none

FILES AFFECTED (where the changes will be implemented) - developers only

unknown

LEVEL OF EFFORT - developers only

medium

COMMENTS

[Christopher Keil]

Does this happen on master or #213? I am assuming all new issues here are master.

[Robert Leach]

Both. And yes - if it's only in your 213 branch, I'll put it in the PR.

[Robert Leach]

Removing milestone: Clustering (automated comment)