#319 Opening Multiple Datasets

Enhancement
closed-fixed
David Benn
8
2013-05-16
2012-04-18
Sara Beck
No

It would be nice to be able to open multiple text files (such as data reports) so that analysis can be performed on them as a whole. One way this could be done is with an "Append to existing dataset" checkbox added to the file dialog.

A further enhancement would be to let the user load data from the AID, then add data from text files to that data set. I don't think it would be necessary to allow it to work the other way around (i.e. loading data from the AID after text files have already been loaded) if that would create a simpler interface.

Discussion

  • David Benn
    David Benn
    2012-04-19

    Appending datasets only from files could be implemented first since that seems to be the primary use case. Adding the ability to append from an AID loaded dataset is in principle no harder than from a file loaded dataset, so possible if we want to do that as well.

     
  • Sara Beck
    Sara Beck
    2012-08-06

    In light of the fact that the AAVSO is strongly urging observers to check their observations before uploading them to the AID, I have come to the realization that there is currently no good way to do this. If VStar would let people add their own data (most conveniently through the AAVSO Upload format plugin) to a dataset for the same star downloaded from the AID, VStar could meet this need very nicely.

    I don't think it matters which file must be loaded first (if that is a constraint) but it is important that the two datasets be distinguishable from one another possibly with a different plot symbol. Remember that both datasets could contain multiple band colors.

    Once the two datasets are together, the user should be able to create phase plots or perform other analysis on the combined dataset.

     
  • Sara Beck
    Sara Beck
    2012-08-06

    • priority: 3 --> 8
     
  • David Benn
    David Benn
    2013-01-07

    Question: Should each loaded dataset (e.g. from AID and data file) have an observation list tab created for it? The alternative is to merge all data into a single table. Some datasets (observation sources) have a different number of columns (data fields); that's true for AID vs AAVSO upload format datasets. I was initially thinking of merging them into a single table.

     
  • Sara Beck
    Sara Beck
    2013-01-07

    I don't think it really matters whether you merge the files into one table or keep them separate. Which do you think would be easiest for you, David?

    Does it make any difference in terms of how they are treated otherwise?

     
  • David Benn
    David Benn
    2013-01-08

    Thanks Sara. I'm trying to figure out the most usable approach. I think a single table tab is least confusing.

    Note to self: Need to think in terms of additive load, and whether the loaded datasets have been or need to be converted to HJD. Also, if an additive load is requested, arbitrary column source should be used for the single obs list. An additional field per obs indicating source may be useful in disambiguation. Info... dialog should also indicate additive load.

     
  • David Benn
    David Benn
    2013-01-10

    If column source is the same as one from existing, use same, else arbitrary column source.

     
  • David Benn
    David Benn
    2013-01-20

    Implemented in http://vstar.svn.sourceforge.net/viewvc/vstar?view=revision&revision=946

    Currently, file based datasets can be loaded additively. AID datasets and those from http-based plugins (BSM, APASS) must be loaded first. Internally there is nothing stopping these being added. UI components (checkboxes) need to be added for this.

     
  • David Benn
    David Benn
    2013-05-16

    • status: open --> closed-fixed