Revive link checker

[wirednkod]

Also checks the PR before merging (when PR is opened).

e.g. https://github.com/Polkadot-Blockchain-Academy/pba-content/actions/runs/5463571326/jobs/9944430712?pr=682#step:4:88

We should also add a rule on settings that the PR cannot be merged unless there is at LEAST 1 approval from an admin or smthng

[JoshOrndorff]

I think this is a good change. But I think we get @nukemandan's opinion before merging.

FWIW this is a middle ground between what I installed in #368 (checks every branch) and what Nuke installed in #648 (only checks main). This one only checks branches that have PRs to main open.

Honestly, this is probably the best idea of all.

[wirednkod]

IMO - there is no reason to have a link checker if everything gets merged without being checked.
The earlier the issue appear, the easier it is to be fixed. This way the instructor will make sure his dirs and links are fine, and the issue will not be delegated to a maintainer or someone else (as I did) to fix the issues;

[nuke-web3]

duplicate #690 ?

See #686 too for maybe a yet untried minimal CI running to add value option 🙏

[wirednkod]

#690 is a DO NOT MERGE TEST PR :) So we keep this as the "track"

[JoshOrndorff]

Why is this closed? We need serious CI on this repo. @nukemandan As a PBA session goes on you get more and more pissed at contributors for missing things like broken links or closing / characters in img tags. Why are you so opposed to just having a CI that checks it?

[wirednkod]

I merged with main and the checker already captured some link issues. I think it is important to keep this one - but its your call 100%

[nuke-web3]

This repo is moving to an archive, and thus we will no longer need to try and fix things, a new monorepo book is imminent that we can get better CI on. Here is the WIP that will be destroyed when approved and the HEAD of that repo used to start the new public facing content repo with slides embedded: https://github.com/Polkadot-Blockchain-Academy/testing-only

[JoshOrndorff]

@wirednkod I think it is time to get this one in :D

I can't reopen it, but maybe you can since it is your PR.

[nuke-web3]

The book is the repo to use only now. This repo can and should have issues and PRs migrated.
you can migrate this over there, except you don't need to. https://github.com/Polkadot-Blockchain-Academy/pba-book/blob/main/.github/workflows/checks.yml

Please review
https://polkadot-blockchain-academy.github.io/pba-book/contribute/index.html

[wirednkod]

Thank you @nukemandan for the feedback. This, though, was adviced to revive and run here.

[nuke-web3]

OK well I am sadly not involved at this time to help out, although that may change (@Asamartino will let me know) I would advocate not trying to maintain this repo in addition or instead of the book, that this CI work implies is what you are steering for.

If you do decide to abandon the book or maintain changes between these repos, I suggest strongly pulling from the book first for the slides before restarting work here, as I did clean some MD and associated img/other files and fixed a lot of broken and stale links there (SDK migration as well).

One key factor the book provides that this repo does not: a replacement for notion as a home for the content in sequence. So yet more work would be needed that I tried hard to avoid in making the book will be required to share and update things easily.

Happy to advise on things and help clarify things as I do know I was not able to communicate with faculty past UCB to motivate anything I was doing toward the book.

Best of luck in HK if I am not part of things at all team ❤️

[wirednkod]

@JoshOrndorff This still needs some links to be fixed (I'll take care of it)
@nukemandan sorry to hear about your non-involvenment. I am not aware of the book's content but I'll ask around or cherry-pick whatever changes need to be passed here, if any. Thank you for your advices though - highly appreciated 🙏🏽

[JoshOrndorff]

Vitalik's entire website is down right now. I suspect it won't be permanent, and we may be able to restore this one.

[wirednkod]

Alright then. This is not a hurry anyways so we could wait for it