Update fish completions

[andrewmeyer]

Update the fish completions profile to support newly added commands.
Some unseen work was also done: notably organizing the ungrouped commands.
Adds support for new features:

• limit to lunar cycles
• support csv output

Completions Added

• add
• aggregate (fixes Fish Shell completions: aggregate is not included #289 )

Completions Extended

• report
• log
• stop (Stop frame at given time #233)

Tests Ran:

• tests were ran against the current master position: 30313be
• Due to the nature of completion profiles, proof of testing can be tricky, I encourage another fish user to vet these additions as well.

add

chimera  ~/  watson add --from 2019-07-04-00:00 --to 2019-07-04-23:00 "take the day off"
Adding project take the day off, started 4 days ago and stopped 4 days ago. (id: c9874b9)

chimera  ~/  watson add --from 2019-07-04-00:00 --to 2019-07-04-23:00 "take the day off" +test
Adding project take the day off [test], started 4 days ago and stopped 4 days ago. (id: b163455)

chimera  ~/  watson add --from 2019-07-04-00:00 --to 2019-07-04-23:00 "take the day off" +test2 --confirm-new-tag 
Tag 'test2' does not exist yet. Create it? [y/N]: y
Adding project take the day off [test2], started 4 days ago and stopped 4 days ago. (id: 7411689)

chimera  ~/  watson remove 7411689
You are about to remove frame take the day off [test2] from 00:00 to 00:00, continue? [y/N]: y
Frame removed

chimera  ~/  watson remove b163455
You are about to remove frame take the day off [test] from 00:00 to 00:00, continue? [y/N]: y
Frame removed.

• we see here that I didn't understand how to properly enter my time, this may need to be clarified in the docs, not unlike watson stop --at. That is out of scope for this particular PR though.

aggregate

chimera  ~/  watson aggregate --project "take the day off" -G
Mon 01 July 2019 - 00s


Tue 02 July 2019 - 00s


Wed 03 July 2019 - 00s


Thu 04 July 2019 - 00s
  take the day off - 00s


Fri 05 July 2019 - 00s


Sat 06 July 2019 - 00s


Sun 07 July 2019 - 00s


Mon 08 July 2019 - 00s

csv support

chimera  ~/  watson aggregate --csv       Mon 08 Jul 2019 04:24:40 PM CDT
project,to,tag,from,time
take the day off,2019-07-04 23:59:59,,2019-07-04 00:00:00,0.0

lunar support

chimera  ~/  watson log --luna -G
Thursday 04 July 2019 (00s)
  c9874b9  00:00 to 00:00          00s  take the day off

[k4nar]

I approve the idea, but I didn't test it 😄 .

[andrewmeyer]

@nesmyslny might be able to help test this

[ybc37]

Sorry for the late reply. I'm short of time, but hope to be able to test it in the next days.
Thanks so much for this :)

[ybc37]

I tested the changes and IMO it looks good! Thank you for fixing and improving the completions 💯

Not related to this PR: Personally I find the logic with required conditions for some options a bit confusing (e.g. __fish_watson_has_from). Example:

$ watson add -
# tab -> only completion result: `--from`

$ watson add --from "..." -
# tab -> completion results: every other option but `--from`

First I expected that if there's a --from the completion will only show --to and when those required options are set, completion gives all other options. That would be consistent IMHO.

But honestly, I don't expect such a logic (not sure how other completions do this; never thought about it). I'm happy with "dumb" completions 😄

I totally see this is a feature and don't want to push against it as this is just my opinion. Just wanted to give some feedback 😉

[andrewmeyer]

that is odd and somewhat unintended. The intended functionality is to only show --to if --from is provided. I'll play with that when I can and see if I can fix it in a future PR. good find!

-edit:
this was a bug: it was introduced by using __fish_watson_has_from instead of __fish_watson_using_command
fixed in 63d0199

[jmaupetit]

Thanks @andrewmeyer 🙏