sakai icon indicating copy to clipboard operation
sakai copied to clipboard

SAK-48907 Lessons restore lessons subpage navigation

Open hornersa opened this issue 1 year ago • 4 comments

Besides restoring the Lessons subpage nav code removed from SAK-48804, this proposed PR repairs UX issues that may not have yet been addressed when adapting the Lessons subpage nav to the new portal, Trinity.

For a demo of the proposed fix, refer to the screencast, SAK-48907-ProposedFix.mp4, attached to the corresponding jira.

hornersa avatar Jun 14 '23 16:06 hornersa

I've narrowed down a few dozen Codacy errors down to five, about which I don't know how to proceed-- one of which (the medium one) looks like a false error, and it's based on code that existed prior to SAK-48804.

hornersa avatar Jun 14 '23 23:06 hornersa

Responses regarding the wider context of this PR are in the conversation in the corresponding jira.

The defaults of my emacs editor (and my copying-pasting) were responsible for the odd indentation issues raised by the first commit of this PR. My revision to this PR will restore the prior version of this file, plus one fix to make the brace positions consistent within the class.

Other issues I’ll respond to where they’re raised.

hornersa avatar Jun 15 '23 21:06 hornersa

What's the state of this PR? Thanks

bgarciaentornos avatar May 17 '24 07:05 bgarciaentornos

I've been maintaining my own local branch of this feature, using the same paradigm as was supported in the past: the Lessons tool server side provides json to the portal, and the portal's client side code parses and renders the subpage navigation.

hornersa avatar May 17 '24 12:05 hornersa

@hornersa do you have any more updates to this PR?

ern avatar Aug 01 '24 20:08 ern

@ern - My code for this has evolved over time. Per comments in the corresponding jira back in May or so (https://sakaiproject.atlassian.net/browse/SAK-48907), I'm still dealing with several other priorities. It will take me a while to get back to trying to contribute similar code. I defer to you on whether we should close this PR. I surmise it would be easier for me down the road to open a different one, though it would be good to preserve the feedback in this one.

hornersa avatar Aug 01 '24 21:08 hornersa

is this your latest and best code addressing this issue?

ottenhoff avatar Aug 01 '24 22:08 ottenhoff

@ottenhoff - No, this code PR was submitted 13 months ago. As it was not incorporated, I made changes to it over time.

hornersa avatar Aug 01 '24 22:08 hornersa

can you point us to your best branch addressing this issue? thanks.

ottenhoff avatar Aug 01 '24 23:08 ottenhoff

@ottenhoff - Bear in mind that I've solely evolved the code for 23.x. On a temporary, provisional basis, I have pushed the current, in-development copy of Pacific Lutheran University's Sakai 23 code to github here:

https://github.com/hornersa/sakai/tree/23.x-PLU-SHARED-20240802

Github is indexing the code at the time I'm writing this, so searches through the git log in github might not work for a while. That said, after you copy this branch, I recommend searching for the following strings in the commit messages, listed chronologically from latest to earliest:

  • SAKAI23-345 (1 commit)
  • SAKAI23-324 (4 commits)
  • SAK-48907 (1 commit)

I think these commits account for the entirety for the Lessons subpage navigation feature that we have working locally. However, if you observe that there are significant missing pieces, let me know and I'll investigate.

Once you have the code copied, let me know and I'll proceed to remove the branch from my repo.

I hope this helps.

hornersa avatar Aug 02 '24 18:08 hornersa

@ern - During our phone conversation I think I may have mis-spoken about the "two sqlRead calls" per Sakai site per page load. Since then I've revisited the pertinent code for SimplePageToolDaoImpl.getLessonSubPageJSON. While the "two sqlRead calls" was a problem at some point, I now think that with one of the "SAKAI23-324" commits I had replaced both calls with logic relying on objects already cached by hibernate.

That said, it would still make sense to profile the code, especially the extant calls to getLessonSubPageJSON.

hornersa avatar Aug 16 '24 20:08 hornersa