D4809: narrow: move the ellipses server capability to core
martinvonz (Martin von Zweigbergk)
phabricator at mercurial-scm.org
Mon Oct 1 06:08:02 UTC 2018
martinvonz added inline comments.
INLINE COMMENTS
> martinvonz wrote in hg.py:582-583
> This patch seems to do more than what it says, and more than what the previous patch did. Can we separate out this functional change? IIUC, this makes it so we start writing the ellipses requirement to the clone if the source repo has the ellipses capability. I'm not sure that's the right thing to do -- just because the server supports ellipses doesn't necessarily mean that all users want that. I know they user/client has no say on that decision (it's controlled by the "narrowservebrokenellipses" config), but we might want to change that. OTOH, since that's how the server currently works, it's probably fine to set the requirement based on the server capability. I mostly just wanted to make sure keep this in mind. (But I would still prefer to see the refactoring separated from the functional change.)
Oh, I think you actually meant to leave this change for the next patch, right? Because ELLIPSES_REQUIREMENT that's used below doesn't seem to exist yet.
> pulkit wrote in test-narrow-clone.t:48
> I just ran `run-tests.py -i` and it popped up and I accepted it. I will check if the test pass without this change or not.
I forgot to say earlier that it's a common problem with `run-tests.py -i` that it doesn't know where to put lines around optional lines. So I'm pretty sure the test will pass even if you revert this change.
REPOSITORY
rHG Mercurial
REVISION DETAIL
https://phab.mercurial-scm.org/D4809
To: pulkit, durin42, #hg-reviewers
Cc: martinvonz, mercurial-devel
More information about the Mercurial-devel
mailing list