Update/Improve geoCSV generator

[DocOtak]

Background

Right now the geoCSV generator follows the inherited code where it hard codes the qa flag to 0. There are other issues I've noted such as all the field_* headers do not have the same number of columns/labels in the actual csv block (6 vs 7). The single digit qa flag also does not seem to support multiple issues with casts. However, I don't know if this is actually invalid for a "geoCSV". Since this matches the old QA software, addressing this feels like a non blocking issue.

Things that probably need to happen:

• update field_unit, field_type, field_standard_name, and field_missing to have the same number of fields as the actual data

• convert the qa field to be a bit field so they can be logical ORed together and logically ANDed apart.:

  • 0001 - invalid time
  • 0010 - invalid nav
  • 0100 - time outside range
  • 1000 - nav outside of range

  examples:

  • invalid time and invalid nav -> 3 (0011)
  • invalid time and nav outside of range -> 9 (1001)
  • everything ok -> 0 (0000)
  • outside nav and time range 12 -> (1100)

Decisions

Since I do not really know how this geoCSV is used and don't want to break anything, some input from R2R is needed. Specifically, I think either:

• Specific instructions to update the geoCSV output (including keep it as is)

OR

• Permission to unilaterally fix the above issues without specific R2R input

Additional would be nice:

• Someone or software that can validate the geoCSV headers/output

---

Edits:

• Added that I didn't know if this actually is a malformed geoCSV
• Added that I didn't think leaving this unresolved was blocking (for QA in production).
• Added what decisions I think need to be made