Observed data as References

[Yuri05]

The main goal is: additionally to the storage of observed data as part of PK-Sim/MoBi project (Status Quo): enable storage of only the Reference to the corresponding data source, e.g.:

• local, absolut: C:\Data\MySuperData.xlsx
• local, relative: ..\..\Data\MySuperData.xlsx
• SVN:

  https://mysvnserver/svn/mysvnrepo/!svn/bc/3/branches/2.0/ObservedData.xls 

  Highlighted part points to the specific revision; e.g. here: Revision 3
• GitHub:

  https://github.com/Open-Systems-Pharmacology/MyGitHubRepo/blob/451f35b2a523a6bda9c4c98177c5fd996ffc7e63/ObservedData.xls

  Highlighted part points to the specific commit. Alternatively: point to released versions of the data, e.g.

  https://github.com/Open-Systems-Pharmacology/MyGitHubRepo/releases/download/v2.0/ObservedData.xls

Suggested workflow:

• 2 Import modi should be available in PK-Sim/MoBi: „Import as data“ (=status quo) and „Import as reference“
  • Import mode can be changed under program options
  • PK-Sim/MoBi check at every program start if the Import mode was saved in the user options. If not: user is prompted to set the option explicitly (same program behavior as in case of chart watermarks)
• When new observed data is imported: Import mode used during the import is saved for every imported data set.
• In case of „Import as data“ – no changes to the status quo. Imported data is saved as part of the project and reloaded from project when it’s opened (no connection to the original data source is required anymore)
• In case of „Import as reference“:
  • When importing from new data source: checksum of the data source is calculated and saved for every imported data set:
  • When closing a project: all ObservedData with import mode „Reference“ will NOT be saved. Only the reference to the used data source and import settings are saved in the project.
  • When opening a project: all ObservedData with import mode „Reference“ are automatically reimported (based on the saved data source reference and import settings)
    • If data source is not available: observed data sets will be shown as “Missing” (+ Warning is shown to the user)
      • To be discussed: how to proceed with Simulations/Comparisons/PIs using Missing observed data sets
    • If data source is available but the data was changed (e.g. perform CheckSum-comparison): observed data sets will be shown as “Changed” (+ Warning is shown to the user)
      • To be discussed: how to proceed with Simulations/Comparisons/PIs using Missing observed data sets
  • When exporting: only ImportConfiguration+ReferenceToDataSource+Checksum is exported, but not the data itself. This applies to:
    • Export project as snapshot
    • Export PI to Matlab/R
    • Save ObservedData in Working Journal
    • Save ObservedData as pkml
    • Save ObservedData as template
    • When exporting observed data to file (excel, csv, …): the data is exported

Additionally:

• It should not be possible to edit observed data in projects anymore. In case of changes: “Update data set from a new data source” workflow should be used instead
• It should be possible to make a data set imported in “Reference” mode persistable in project (means: data will remain in project)