Updates to 1. zsh autocomplete script and 2. slow link type warning message

[jorgeorpinel]

• Have you followed the guidelines in our
  Contributing document?

• N/A Does your PR affect documented changes or does it add new functionality
  that should be documented? If yes, have you created a PR for
  dvc.org documenting it or at
  least opened an issue for it? If so, please add a link to it.

---

[shcheklein]

@jorgeorpinel why does your commit history look strange? rebase?

[efiop]

@shcheklein that is how merges look :) That is why I usually advise people to

git remote add upstream https://github.com/iterative/dvc
git fetch upstream
git rebase upstream/master
git push -f

when they are submitting PRs 🙂

[shcheklein]

@shcheklein that is how merges look :) That is why I usually advise people to

it depends on your workflow with merge, right? :) If you always use a new feature-branch from master or if you don't squash commits it should not include commits that are already merged.

[efiop]

@shcheklein Well, in general merge can result in stuff like that. You are lucky if you get "fast-forward"(which is effectively a no-conflict rebase), but with "rebase" you always get a nice-looking PRs only with your commits 🙂

[ghost]

It is hard to review sort + changes 😅

[ghost]

@jorgeorpinel , besides sorting and adding the new commands, did you change anything?

[jorgeorpinel]

@efiop thanks for letting me know about hidden commands, I'll remove those comments in a bit...

@shcheklein its just merged from upstream/master. Should I be rebasing always? I'm not too used to rebasing but I'm happy to adopt

@MrOutis yes I should have sorted in one commit and changed in another, will keep in mind in the future. The actual updates are just to the descriptions of the commands, to match the current help output of each one.

[jorgeorpinel]

@MrOutis @shcheklein the new completion script is done now and I tried it manually on my system, please review.

[jorgeorpinel]

Also, after rebasing upsteam/master this history includes my changes to git diff (to output short commit hashes) again but those were already merged in #1906. I'm not sure why that happened.

[pared]

I like the change, thought I think we should consider some changes when handling message strings in general, @jorgeorpinel, had to push another change updating the test warning message. As @Suor pointed in #1942 this is not desired workflow. Maybe we should create a task for creating separate file with warning and error messages?

[efiop]

@pared Not sure it is worth a separate file. It is not like you are going to reuse those other than in the test. Defining it as a constant somewhere in the same file seems good enough.

[jorgeorpinel]

A constant in the appropriate class or file may be enough like Ruslan points out but it also sounds like a good practice to extract strings to its own file(s) if we're thinking about i18n in the future. M 2¢

UPDATE: Do we need an issue for this?

[efiop]

@jorgeorpinel @pared already added #2026 , but forgot to reference 🙂

[pared]

Sorry for not referencing it here

[jorgeorpinel]

All good. OK so what's missing for this PR? Can it be merged? 😃

[efiop]

@jorgeorpinel Let's wait for @MrOutis to review.

[ghost]

Please, reformat the docstrings accordingly to https://github.com/google/styleguide/blob/gh-pages/pyguide.md#38-comments-and-docstrings

[jorgeorpinel]

Done. That docstring Args wrapping was an accident 😅

[efiop]

Thanks! 🎉

stale

[jorgeorpinel]

Thank you! I noticed some of the - options in some commands don't come up in auto-complete though :p may need to revise again soon (probably same for bash).

p.s. see also #1924