Skip to content
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

fix: update to swipl v9.3.15 #787

Merged
merged 8 commits into from
Nov 24, 2024
Merged

fix: update to swipl v9.3.15 #787

merged 8 commits into from
Nov 24, 2024

Conversation

jeswr
Copy link
Collaborator

@jeswr jeswr commented Nov 19, 2024

fix: update to swipl v9.3.15

@jeswr jeswr enabled auto-merge (squash) November 19, 2024 00:10
@jeswr
Copy link
Collaborator Author

jeswr commented Nov 20, 2024

@JanWielemaker @josd I am assuming that we would not expect the recently added rwlocks tests to succeed in a browser build. If this is the case; could you please advise on how to skip them in the ctest here as I am not particularly familiar with ctest myself.

Ideally, could you push the changes skipping the ctest for rwlocks as a commit to this PR.

@JanWielemaker
Copy link
Member

Ideally, could you push the changes skipping the ctest for rwlocks as a commit to this PR.

Not sure what you mean by that. I have pushed a fix as SWI-Prolog/swipl-devel@bb29bab. This moves the test to the thread directory. I hope you can work it out from there.

@jeswr
Copy link
Collaborator Author

jeswr commented Nov 21, 2024

Ideally, could you push the changes skipping the ctest for rwlocks as a commit to this PR.

Not sure what you mean by that. I have pushed a fix as SWI-Prolog/swipl-devel@bb29bab. This moves the test to the thread directory. I hope you can work it out from there.

Thanks @JanWielemaker - moving the test to the thread directory solved the ctest issue.

We now have a new issue when running against SWI-Prolog/swipl-devel@21f46ad in particular, the yield value after calling swipl.prolog.call appears to have been changed from true to false. The test is here and the CI failure can be found here.

@JanWielemaker
Copy link
Member

If I read this correctly, there is nothing wrong with the yield value, but the .done field is not what you expect. It needs to have this value to satisfy the Query as iterator. I need to think. Possibly we can get around by setting some flag on the query that tells us it is an iterator or not? Otherwise we need to redesign the API a little.

@JanWielemaker
Copy link
Member

Pushed SWI-Prolog/swipl-devel@d5192b2 to address this and SWI-Prolog/swipl-devel@2910776 to get rid of the deprecated messages for $allocate and friends. Builds using emcc 3.1.72 and passes the Prolog (ctest) tests.

@jeswr
Copy link
Collaborator Author

jeswr commented Nov 21, 2024

If I read this correctly, there is nothing wrong with the yield value, but the .done field is not what you expect.

Correct, my bad!

Pushed SWI-Prolog/swipl-devel@d5192b2 to address this and SWI-Prolog/swipl-devel@2910776 to get rid of the deprecated messages for $allocate and friends. Builds using emcc 3.1.72 and passes the Prolog (ctest) tests.

Thanks for being so fast! CI now running with SWI-Prolog/swipl-devel@2910776 - lets see how it goes

@jeswr
Copy link
Collaborator Author

jeswr commented Nov 21, 2024

Ok, that's fixed those tests (yay!). There is now a new error later on file `'prolog.p'' does not exist; it looks like the file we provide is prolog.pl.

Was support for .pl extensions dropped? Is there a bug meaning that the last letter in the command is dropped?

@JanWielemaker
Copy link
Member

Nothing changed there. Could it be that I made a mistake with SWI-Prolog/swipl-devel@2910776? It seems that the node invocations of SWI-Prolog during the build work fine, but I'm not 100% sure they use this. Do you see anything wrong with this change?

@jeswr
Copy link
Collaborator Author

jeswr commented Nov 24, 2024

Looking now - I also think that the -q flag isn't being picked up see here

@JanWielemaker
Copy link
Member

Yip. The last character of each argument was lost. Fixed with SWI-Prolog/swipl-devel@848bc2d

@jeswr
Copy link
Collaborator Author

jeswr commented Nov 24, 2024

Nothing changed there. Could it be that I made a mistake with SWI-Prolog/swipl-devel@2910776? It seems that the node invocations of SWI-Prolog during the build work fine, but I'm not 100% sure they use this. Do you see anything wrong with this change?

@JanWielemaker yes - I ran the tests on the previous commit SWI-Prolog/swipl-devel@d5192b2 and they all pass and then fail on SWI-Prolog/swipl-devel@2910776

@jeswr
Copy link
Collaborator Author

jeswr commented Nov 24, 2024

Yip. The last character of each argument was lost. Fixed with SWI-Prolog/swipl-devel@848bc2d

Awesome - updated PR, lets see how the CI goes

@jeswr
Copy link
Collaborator Author

jeswr commented Nov 24, 2024

Seems to be fixed locally - thanks @JanWielemaker !

@jeswr jeswr merged commit f9c8dbe into master Nov 24, 2024
21 checks passed
@jeswr jeswr deleted the fix/update-swipl-v9.3.15 branch November 24, 2024 15:24
@jeswr
Copy link
Collaborator Author

jeswr commented Nov 24, 2024

🎉 This PR is included in version 4.0.15 🎉

The release is available on:

Your semantic-release bot 📦🚀

@jeswr jeswr added the released label Nov 24, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants