-
-
Couldn't load subscription status.
- Fork 17.1k
Revert "stdenv: pURL implementation (#421125)" #453322
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
|
I'm confused by the commit message:
The second commit seems very much unrelated. What am I missing? Please clarify the reason for the revert in the commit message, aka refer to the cases that were broken for Eval. |
it is the revert based on the latest master which i had locally (88ee2ac). This was the result of a standard |
Thanks, never seen that! |
|
Merging this to unbreak the various issues we have with meta depending on src. To clarify my comment on the other PR: I understand this will be necessary, but we will need better tooling to make such a change with confidence. For me that means:
Once we do this, we can much better judge the impact of such a change. We can then tell whether it was just these 1-2 broken packages - or whether there were a lot more elsewhere. I don't have an immediate solution for you, but the above two things have been on my list for a while and I have been thinking about them. |
I agree that would be helpful. The cplex thing was a huge hack anyway, tbh for that case I would be fine with some bespoke change (like explicitly setting a purl). |
|
Note that there are quite a few of those hacks in nixpkgs. |
|
sure, just saying it'd be helpful to get a feel for how many - I'd be fine with 5 or even 50, but perhaps not 500 :) |
|
My guess would be somewhere in the 10s but never really did a full scan will hack something together to get you a list (should not take that long to write) |
|
With a simple script, it got the following, which have non-evaluating src: List of non-evaluating src |
#421125 introduced a problem with checkMeta=true for nixpkgs-review. checkMeta defaults to false in general.
The scenario happened for thunderbird-bin in combination with an unsupported platform.
This PR prevents any evaluation which might be problematic
There were concerns, that a revert may be a better option. I keep this PR open and let the community decide: #453291
Things done
passthru.tests.nixpkgs-reviewon this PR. See nixpkgs-review usage../result/bin/.Add a 👍 reaction to pull requests you find important.