Datasets window with shorter dataset names and possibly structured in subbranches

[marcomass]

Make possible to structure the content of the Datasets window in subbranches, within the current only Private/Public ones.

[marcomass]

@acanakoglu @akaitoua Hi Abdulrahman. You moved this issue to Milestode 2, but it is really very unconfortable to deal with current very long name. Is it so long/difficult to implement a solution for this?

[acanakoglu]

Postponed until new repository management

[akaitoua]

@andreagulino can you take care of this, since you updated the dataset meta data.

[marcomass]

After discussion with akaitoua, andreagulino and acanakoglu it was defined that this is a visualization issue in the web interface, and that it can be solved as follow:

• keep the current dataset file name, which is an identifier of the dataset composed of:
  • job_
  • query name
  • user name_
  • time stamp of submitted job
  • dataset name
• adding in the dataset metadata (which andreagulino is building) the following separate attributes and values
  • full_name, equal to the current dataset name stored on disk
  • group, equal to dataset query name
  • time_stamp, equal to time stamp of submitted job
  • dataset_name, equal to dataset name as expressed in the query that generated the dataset
• the web interface will get these attributes and values from the dataset metadata and will use them to structure the visualization of the private dataset tree according to the following levels:
  • group
  • dataset easy name, equal to time_stamp + dataset_name
  • dataset sample Visualization of the public dataset tree will be structured accordingly, by filling (at dataset loading time) the same metadata attributes as follow:
  • full_name, equal to the current dataset name stored on disk
  • group, specified at dataset loading time
  • time_stamp, empty (by now)
  • dataset_name, equal to current dataset name