-
Notifications
You must be signed in to change notification settings - Fork 15
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
Bump SSSP to 1.3 #578
Bump SSSP to 1.3 #578
Conversation
As discussed in AiiDA meeting, I move on without the protocol change in aiida-quantumespresso. But it means we need to explicitly override the pseudo_family when creating a builder. Anyhow, I don't want to block the this SSSP bump which Nicola asks for the QeApp explicitly. |
1528a7b
to
6c738fa
Compare
There still couple of tests failed and I have no idea how to fix those, I guess it requires to haveing a proper |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #578 +/- ##
==========================================
+ Coverage 80.74% 80.86% +0.12%
==========================================
Files 49 49
Lines 3418 3429 +11
==========================================
+ Hits 2760 2773 +13
+ Misses 658 656 -2
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. |
Hi @unkcpz which one? I saw all the tests passed. |
|
Ha, nice. Some failed in my local environment, I guess I messed up with the environment somehow. Can you give this branch a test? I also suggest testing opening a finished workchain after updating the app. In this PR, I also fix an issue that the post_install of pseudo library will fail in the AppStore. |
Thanks for the very detailed comment, I’ll convert it to an issue for future improvements. For the PR itself, can you give it a look? I want to make another release and build the quantum mobile again on Monday.
Sent from Outlook for iOS<https://aka.ms/o0ukef>
…________________________________
From: Xing Wang ***@***.***>
Sent: Thursday, December 14, 2023 11:20:47 PM
To: aiidalab/aiidalab-qe ***@***.***>
Cc: Jusong Yu ***@***.***>; Mention ***@***.***>
Subject: Re: [aiidalab/aiidalab-qe] Bump SSSP to 1.3 (PR #578)
1. the unit tests and fixtures must be refactored, there are too many duplicates and quite a mess. 2) In aiida-quantumespresso, the ProtocolMixin is used and there are protocols defined by yaml files for every workchains. We probably use the same way rather than override it in the code, however, I see this against the plugin where the get_builder is used and the builder is created out of air.
1. Indeed, we need to update the tests and fixtures. Initially, when I wrote the fixture for the workchain and plugin-related tests, the concept of the plugin was still in its early stages. Now, with the plugin architecture having stabilized, we could update the test with a better mind.
2. Using ProtocolMixin, one defines all the parameters in the YAML file. When creating the builder, it fetches these parameters based on the protocol . However, QEapp has a different way. In QEapp, one can still define the protocol and related parameters in the YAML file. But when the user selects the protocol from the GUI, the GUI will fetch the corresponding parameter immediately and show the values to the user. At this stage, the user can modify the GUI to override the parameters. In the final submission step, we read all the parameters from the GUI (the protocol value and the user override values) and use these values to create the builder, so we don't need to read the YAML file in the get_builder function again.
Unless we want to define some other protocol-related parameters, that don't have their corresponding value in the GUI, we can read the YAML in the get_builder function. By the way, the get_builder function is quite flexible. The plugin developer can define the YAML and load it into the get_builder. If this is a common use case, one can improve the get_builder` function to standardize it.
@mbercx<https://github.com/mbercx> knows better ProtocolMixin, feel free to comment.
—
Reply to this email directly, view it on GitHub<#578 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ACDMFRMM4BMCSS4G2ONXRULYJN3T7AVCNFSM6AAAAABARYVQXKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNJWG44DSNRZGQ>.
You are receiving this because you were mentioned.Message ID: ***@***.***>
|
I merged this anyway so as not to block the release of Quantum Mobile, in principle I should not merge it without approve. |
Following tests are all good:
|
The test fail because the default value will read fromaiida-quantumespresso
where the pseudo version in the protocol not updated yet.