introduce dvc update for ext repos#2218
Conversation
dvc update for ext reposdvc update for ext repos
bc45a8e to
d03ed49
Compare
dvc update for ext reposdvc update for ext repos
8834cee to
e16aafc
Compare
dvc update for ext reposdvc update for ext repos
556ef47 to
e21320c
Compare
dvc update for ext reposdvc update for ext repos
bfa2f99 to
70f5177
Compare
|
@Suor Updated my patch to get rid of enter/exit. Please take a look 🙂 |
There was a problem hiding this comment.
can we reuse force or something, to avoid this locked flag manipulation ... does not look really nice
There was a problem hiding this comment.
force doesn't affect locked stages. But I'll take a look.
|
please, update zsh/bash autocompletes if it's not done yet |
There was a problem hiding this comment.
are we running status for the locked stage?
There was a problem hiding this comment.
We do, but not for the dependencies. So update available will only be visible if you dvc unlock import.dvc
There was a problem hiding this comment.
is there a way to check the status before updating it then?
There was a problem hiding this comment.
Yes, if you dvc unlock import.dvc first and then run dvc status
There was a problem hiding this comment.
Hmm ... not an easy interface ... don't see an easy way to make it better ... may be make dvc status target.dvc ignore the lock?
There was a problem hiding this comment.
@shcheklein Maybe dvc update something.dvc --dry then? Changing dvc status itself doesn't feel right.
There was a problem hiding this comment.
@shcheklein In any case, that additional command can be introduced on top, as it doesn't affect the rest of the logic.
There was a problem hiding this comment.
--dry sounds like a good start
There was a problem hiding this comment.
Why status ignores locked stages? This doesn't make sense from consistency POV only from optimization one. Is locked about optimization?
There was a problem hiding this comment.
Yes, locked is also about not actively going and checking dependencies of the stage(e.g. not re-downloading external repo, not going to s3 to check for external s3 dependency etc). Also, locked is kinda making a data source from any stage, so showing status for its dependencies doesn't seem like a desired behaviour.
dvc update for ext reposdvc update for ext repos
Fixes treeverse#1774 Fixes treeverse#2139 Fixes treeverse#2201 Signed-off-by: Ruslan Kuprieiev <ruslan@iterative.ai>
dvc update for ext reposdvc update for ext repos
Signed-off-by: Ruslan Kuprieiev <ruslan@iterative.ai>
|
add a few tickets to update lock flag description, lock/unlock command references, import-url command reference with the new semantics (in case we change it) |
Signed-off-by: Ruslan Kuprieiev <ruslan@iterative.ai>
|
@shcheklein Added #2238 for import-url. I don't think |
|
@shcheklein Ah, you probably mean that those now should mention that it not only |
|
@efiop I meant docs, btw :) I think it worth mentioning on their command references that there is an update command. We'll see if it makes sense. |
|
@shcheklein Got it. Change for |
|
|
||
| out = repo.find_out_by_relpath(self.def_path) | ||
| repo.fetch(out.stage.path) | ||
| to.info = copy.copy(out.info) |
There was a problem hiding this comment.
Is it safe to change passed argument here?
There was a problem hiding this comment.
@Suor yes, we are effectively doing to.info.save(), but it will run later anyway to verify that everything is in check.
Have you followed the guidelines in our
Contributing document?
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.
treeverse/dvc.org#474