You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Use Arena API to display Processing feed on Community page #1615
Instead of calling the API on every deploy, it should have a scheduled (ie. daily) update via Github Actions to call the API again and create a new PR for it (if no outstanding PR exists). This is to prevent blocking deploy from failed API calls.
Hi @doradocodes@Nwakaego-Ego, I'd like to help with this. I've been working on the related #1547, where I have an Are.na fetch script (open in #1552), so I wanted to check how you'd like to split the work before anyone duplicates it.
What I found on main:
The Community page has no Are.na section yet, so this would be a new section in CommunityLayout.astro plus a jump link.
src/api/OpenProcessing.ts already reads committed JSON from src/cached-data/ instead of calling the API at build time. Following that pattern would meet the requirement that a failed API call can't block a deploy. The scheduled job would refresh the cached file and open a PR.
There's no scheduled workflow in .github/workflows yet, so the daily Action that opens a PR (skipped if one is already open) would be new.
Proposal: I can generalize my script to take a channel slug and an output path, so #1547 and #1615 share one implementation, and I can write the scheduled workflow. The Community page UI can stay with whoever is already on it. If either of you has already started on the script, I'm happy to build on yours instead.
Questions:
Should the cached data be JSON in src/cached-data/, like the OpenProcessing data, or a content collection?
Should we continue pulling content from Open Processing for the Community page, or replace it entirely with content from the Arena board? I think only pulling from Arena makes more sense.
The Community the page currently includes multiple sections (Sketches, Libraries, and Events). Would these all be coming from the Arena board as well? If so, we will need to organize the content coming from the Arena board into these sections.
@ksen0, I also have a few questions we may need to answer before we start working on this:
Should the Sketches section include only sketches that can run, or every project made with p5.js? Basically, what should count as a sketch?
Should projects from Are.na also appear under “All Sketches,” or is this only for the Community page?
The Community page currently shows eight projects. Should we show the newest projects added to Are.na, projects selected manually, or a mix of both?
The person who adds a project to the Are.na channel may not be the person who created it, so how should we identify the actual creator of each project?
The project thumbnail and description may not give us meaningful alt text for the image. How should that be handled?
Activity
ksen0 commentedon Oct 8, 2026
Related: #1547
NisargJasani0602 commentedon Oct 9, 2026
Hi @doradocodes @Nwakaego-Ego, I'd like to help with this. I've been working on the related #1547, where I have an Are.na fetch script (open in #1552), so I wanted to check how you'd like to split the work before anyone duplicates it.
What I found on
main:CommunityLayout.astroplus a jump link.src/api/OpenProcessing.tsalready reads committed JSON fromsrc/cached-data/instead of calling the API at build time. Following that pattern would meet the requirement that a failed API call can't block a deploy. The scheduled job would refresh the cached file and open a PR..github/workflowsyet, so the daily Action that opens a PR (skipped if one is already open) would be new.Proposal: I can generalize my script to take a channel slug and an output path, so #1547 and #1615 share one implementation, and I can write the scheduled workflow. The Community page UI can stay with whoever is already on it. If either of you has already started on the script, I'm happy to build on yours instead.
Questions:
src/cached-data/, like the OpenProcessing data, or a content collection?#p5jsv1/#p5jsv2or#p5v1/#p5v2? My parser accepts both for now.By:format? I've only tested Are.na education resources page (step 1: structured by-line/tag parsing) #1552 against synthetic data.doradocodes commentedon Oct 9, 2026
@ksen0 I have questions along with @NisargJasani0602's above:
@NisargJasani0602 Happy to assign this to you as well, since you're working on Education Resources in the same way.
Nwakaego-Ego commentedon Oct 13, 2026
@ksen0, I also have a few questions we may need to answer before we start working on this:
cc @doradocodes