Allow website to fetch ideas from Info Repo#558
Conversation
|
@M4dhav is attempting to deploy a commit to the AOSSIE Team on Vercel. A member of the Team first needs to authorize it. |
📝 WalkthroughWalkthroughThe changes implement a GitHub-backed GSoC Ideas system with automated rebuild triggers. A new workflow orchestrates rebuilds on ideas updates. Dependencies are expanded to support enhanced Markdown processing. Core functionality includes fetching ideas from the GitHub API, rendering dynamic detail pages for each idea, and updating the ideas index to use current-year data from GitHub instead of local sources. Changes
Sequence DiagramsequenceDiagram
participant GH as GitHub API
participant Build as Build Process
participant Pages as Static Pages
participant Deploy as GitHub Pages
Build->>GH: Fetch ideas list<br/>(GSoC-Ideas/{year})
GH-->>Build: Return .md files
Build->>GH: Fetch each idea content<br/>(raw.githubusercontent)
GH-->>Build: Return Markdown content
Build->>Build: Extract title & description<br/>via fetchGitHubIdeas helpers
Build->>Pages: Generate static paths<br/>(ideas/[year]/[slug])
Build->>Pages: Generate static props<br/>(idea details)
Pages->>Pages: Render Markdown with<br/>ReactMarkdown + rehype plugins
Build->>Deploy: Upload artifact
Deploy->>Deploy: Deploy to GitHub Pages
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
🤖 Fix all issues with AI agents
In @.github/workflows/rebuild-on-ideas-update.yml:
- Around line 9-12: The workflow currently grants overly broad repo permissions
by setting permissions.contents: write; change this to permissions.contents:
read to follow least-privilege practice because the job only checks out the repo
and uploads an artifact for actions/deploy-pages rather than modifying
repository contents—locate the permissions block with the "permissions" key and
replace the "contents: write" entry with "contents: read", leaving other
permissions (pages, id-token) as-is.
In `@package.json`:
- Around line 43-47: Package.json currently pins both "remark-gfm": "3.0.1" and
an alias "remark-gfm-4": "npm:remark-gfm@^4.0.0" causing v3/v4 mismatch; replace
the v3 entry by removing "remark-gfm": "3.0.1" and ensure a single dependency
"remark-gfm": "^4.0.0" (remove the alias key or rename the alias back to
remark-gfm), then update imports in next.config.mjs to reference the standard
"remark-gfm" (v4) instead of any v3 import so both Next.js config and React
component use remark-gfm v4 consistently.
In `@src/helper/fetchGitHubIdeas.js`:
- Around line 117-125: The fetchIdeasFromGitHub function builds an
unauthenticated request to contentsUrl which can be rate-limited; update the
fetch call in fetchIdeasFromGitHub to include an Authorization: `token
${process.env.GITHUB_TOKEN}` header when process.env.GITHUB_TOKEN is defined
(keep the existing Accept header) so server-side builds use authenticated GitHub
API requests and avoid the 60 requests/hour limit tied to
GITHUB_API_BASE/REPO_OWNER/REPO_NAME/IDEAS_PATH calls for getCurrentYear; ensure
the token is only read from environment (no client-side exposure) and the logic
gracefully falls back to the existing unauthenticated headers if the token is
missing.
In `@src/pages/ideas/`[year]/[slug].jsx:
- Around line 204-219: getStaticPaths currently only builds paths for the single
year returned by getCurrentYear() which causes 404s for other years; update
getStaticPaths to either (A) iterate all supported years (e.g., a list or
function like getSupportedYears()) and call fetchIdeasFromGitHub(year) for each
year to aggregate paths for every idea slug, or (B) switch the returned object
to fallback: true so Next.js will attempt rendering unmapped year/slug combos at
request time; modify the code inside getStaticPaths (references: getCurrentYear,
fetchIdeasFromGitHub, getStaticPaths, and fallback) accordingly and ensure the
final paths array contains entries for every target year if you choose option A.
🧹 Nitpick comments (3)
src/pages/ideas/index.jsx (1)
115-119: Consider an empty-state message when no ideas are returned.If the GitHub fetch fails, the grid renders blank. A short empty-state makes this more user-friendly.
💡 Example empty-state
<Grid container spacing={4} sx={{ justifyContent: 'center' }}> - {articles.map((article) => ( - <Article key={article.slug} article={article} currentYear={currentYear} /> - ))} + {articles.length === 0 ? ( + <Typography sx={{ mt: 4, width: '100%', textAlign: 'center' }}> + No ideas are available right now. Please check back later. + </Typography> + ) : ( + articles.map((article) => ( + <Article key={article.slug} article={article} currentYear={currentYear} /> + )) + )} </Grid>src/pages/ideas/[year]/[slug].jsx (1)
15-22: Remove unreachablerouter.isFallbackcheck withfallback: false.With
fallback: falseingetStaticPaths, unmapped routes return 404 pages instead of rendering fallback states. Therouter.isFallbackbranch will never execute and should be removed.🧹 Remove unreachable fallback UI
- if (router.isFallback) { - return ( - <Container className="mt-16 mb-20"> - <div className="text-center"> - <p className="text-lg text-zinc-600 dark:text-zinc-400">Loading...</p> - </div> - </Container> - ); - }src/helper/fetchGitHubIdeas.js (1)
141-155: UsePromise.allSettledto handle per-idea fetch failures gracefully.Currently,
Promise.allwithout acontentResponse.okcheck means any single network failure discards all results. Additionally, if a fetch succeeds but returns an error status, the response body is parsed as content. Node.js 18.17.0+ (required by this project) fully supportsPromise.allSettled, allowing you to skip failed fetches and return partial results instead of dropping everything.🧯 More resilient fetch aggregation
const ideasPromises = mdFiles.map(async (file) => { const contentResponse = await fetch(file.download_url); const content = await contentResponse.text(); return { slug: file.name.replace(/\.md$/, ''), title: formatTitle(file.name), description: extractDescription(content), downloadUrl: file.download_url, year: year, }; }); const ideas = await Promise.all(ideasPromises); + + const ideaResults = await Promise.allSettled( + mdFiles.map(async (file) => { + const contentResponse = await fetch(file.download_url); + if (!contentResponse.ok) { + throw new Error(`Idea fetch failed: ${contentResponse.status} ${contentResponse.statusText}`); + } + const content = await contentResponse.text(); + return { + slug: file.name.replace(/\.md$/, ''), + title: formatTitle(file.name), + description: extractDescription(content), + downloadUrl: file.download_url, + year: year, + }; + }) + ); + + return ideaResults + .filter((r) => r.status === 'fulfilled') + .map((r) => r.value);
| export async function getStaticPaths() { | ||
| // For static export, we need to generate all paths at build time | ||
| const currentYear = getCurrentYear(); | ||
| const ideas = await fetchIdeasFromGitHub(currentYear); | ||
|
|
||
| const paths = ideas.map((idea) => ({ | ||
| params: { | ||
| year: currentYear.toString(), | ||
| slug: idea.slug, | ||
| }, | ||
| })); | ||
|
|
||
| return { | ||
| paths, | ||
| fallback: false, | ||
| }; |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Look for year-specific idea links/pages that imply multi-year support
echo "=== Searching for year-specific idea references ==="
rg -n "ideas/20\d{2}" src --type jsx --type ts --type tsx || echo "No matches found"
echo -e "\n=== Files in ideas directory structure ==="
find src -type d -name "ideas" | head -20
echo -e "\n=== Checking for [year] directory patterns ==="
fd -a --type d "\[year\]" src/pages
echo -e "\n=== Checking for any hardcoded years in src ==="
rg -n "202[0-9]" src --type jsx --type ts --type tsx -A 2 -B 2 | head -50Repository: AOSSIE-Org/website
Length of output: 403
🏁 Script executed:
#!/bin/bash
echo "=== Reading the file under review ==="
cat -n src/pages/ideas/[year]/[slug].jsx | head -250
echo -e "\n=== Finding getCurrentYear function ==="
rg -n "getCurrentYear" src -A 5
echo -e "\n=== Finding fetchIdeasFromGitHub function ==="
rg -n "fetchIdeasFromGitHub" src -A 10
echo -e "\n=== Searching for navigation/links to ideas ==="
rg -n "ideas/" src --type js --type ts -B 2 -A 2 | head -60Repository: AOSSIE-Org/website
Length of output: 21253
Generate paths for all supported years or clearly constrain to current year only.
The getStaticPaths() with fallback: false only generates paths for currentYear. Although navigation currently avoids this by linking to dedicated year pages (e.g., /ideas/2024/index.jsx), the dynamic route structure /ideas/[year]/[slug] can technically accept any year. Attempting to access a historical year directly via the dynamic route (e.g., /ideas/2023/some-idea) will 404. Either generate static paths for all years with ideas (2022, 2023, 2024, etc.), or set fallback: true to handle unmapped years dynamically, or document that the route is current-year-only.
🤖 Prompt for AI Agents
In `@src/pages/ideas/`[year]/[slug].jsx around lines 204 - 219, getStaticPaths
currently only builds paths for the single year returned by getCurrentYear()
which causes 404s for other years; update getStaticPaths to either (A) iterate
all supported years (e.g., a list or function like getSupportedYears()) and call
fetchIdeasFromGitHub(year) for each year to aggregate paths for every idea slug,
or (B) switch the returned object to fallback: true so Next.js will attempt
rendering unmapped year/slug combos at request time; modify the code inside
getStaticPaths (references: getCurrentYear, fetchIdeasFromGitHub,
getStaticPaths, and fallback) accordingly and ensure the final paths array
contains entries for every target year if you choose option A.
|
@M4dhav |
We aren't storing the ideas in this Repository anymore, and as mentioned in the PR Description, Ideas are now fetched from AOSSIE-Org/Info . So the existing folders here are for ideas from previous years |
2 participants
Summary by CodeRabbit
Release Notes
New Features
Chores