Add docs for CSV_EXPORT2

[berland]

Closes #193

[oyvindeide]

This looks good, thanks for adding documentation! Could you add a test checking that the description and examples are valid rst? They are, but for future use. Could be something like this: https://github.com/equinor/semeio/blob/master/tests/jobs/correlated_observations_scaling/unit/test_examples.py

Also, can you link it to the corresponding issue?

[berland]

Also, can you link it to the corresponding issue?

Not enough privileges to link to issues..

[oyvindeide]

Ahh, closes #214

[rnyb]

MAKE_DIRECTORY <CONFIG_PATH>/ert/output/csv
CSV_EXPORT2 <RUNPATH_FILE> <CONFIG_PATH>/ert/output/csv/.csv monthly F* W* TCPU TIMESTEP

I would suggest to write output to the scratch disk (instead of the project disk),
in .../<case_dir>/share/someoutputfolder/...

[rnyb]

Would it be useful to add to the examples or description some tips on how to avoid too large csv files?

[sondreso]

Ahh, closes #214

This needs to be in the description of the PR ☺️ (And probably point to the issue and not the PR?)

[oyvindeide]

Ahh, closes #214

This needs to be in the description of the PR ☺️ (And probably point to the issue and not the PR?)

Such high requirements! :p

[berland]

Would it be useful to add to the examples or description some tips on how to avoid too large csv files?

Thinking of vector selection or time frequency? Example tip?

[berland]

MAKE_DIRECTORY <CONFIG_PATH>/ert/output/csv
CSV_EXPORT2 <RUNPATH_FILE> <CONFIG_PATH>/ert/output/csv/.csv monthly F* W* TCPU TIMESTEP

I would suggest to write output to the scratch disk (instead of the project disk),
in .../<case_dir>/share/someoutputfolder/...

Done.

[rnyb]

Would it be useful to add to the examples or description some tips on how to avoid too large csv files?

Thinking of vector selection or time frequency? Example tip?

Both, i see time interval examples is covered now. Though convenient the * notation may easily dump much more vectors than the user needs(?) But maybe this should be left to the user to judge...?

As an example, using WOPT*, WWPT*, WGPT* instead of W* can make a huge difference

[oyvindeide]

Everything looks good from my perspective, but @rnyb is probably better placed to review the contents of the docs. So when he is happy this can be merged.

[rnyb]

Everything looks good from my perspective, but @rnyb is probably better placed to review the contents of the docs. So when he is happy this can be merged.

I think it looks good now

[berland]

NB: "weekly" will not work before komodo 2020.10 is stable.

[oyvindeide]

NB: "weekly" will not work before komodo 2020.10 is stable.

Is it a lot of work to split that into a separate PR that we can merge when that is closer in time? Normally that should be so close that it would not be a problem, but might be delayed.

Other than that, rebase and it is good to go.

[oyvindeide]

Great, can you also create an issue or a PR straight away so we dont forget?

[berland]

Rebased.

[oyvindeide]

LGTM! Thanks for adding docs!