Use CORS-clean hosts for in-browser data reads

[mmcky]

This site executes code cells in the reader's browser (Pyodide via pyodide_http.patch_all()), so every pandas URL read is subject to CORS on each redirect hop. The github.com/<org>/<repo>/raw/… form is a 302 whose response carries an empty access-control-allow-origin header, which browsers reject before following the redirect — so those reads fail in the one repo where the browser is the runtime. raw.githubusercontent.com and (for LFS files) media.githubusercontent.com serve access-control-allow-origin: * and work. The regression and the accompanying repoint rule are tracked in QuantEcon/data-lectures#46.

Five reads flipped, two groups:

Lecture	Read	Old host	New host
long_run_growth	mpd2020.xlsx	github.com/QuantEcon/data-lectures/raw/… (from #52)	raw.githubusercontent.com
inflation_history	longprices.xls, chapter_3.xlsx	github.com/QuantEcon/data-lectures/raw/… (from #53)	raw.githubusercontent.com
inequality	SCF_plus_mini.csv	github.com/QuantEcon/high_dim_data/raw/… (pre-existing)	media.githubusercontent.com (LFS)
mle	SCF_plus_mini_no_weights.csv	github.com/QuantEcon/high_dim_data/raw/… (pre-existing)	media.githubusercontent.com (LFS)

The {download} link and prose links keep the github.com form — they are plain navigations, CORS does not apply, and that form gives the reader the friendlier GitHub UI on click.

Verification: all five new URLs fetched from this site's own origin (quantecon.github.io) in headless Chromium — 200 with the exact expected sizes (1,765,204 / 388,608 / 73,281 / 32,853,734 / 75,902,999 bytes; the first three byte-identical to the sha256-verified data-lectures copies). The old form fails the same test with Failed to fetch. Reproduce from any quantecon.github.io page console: fetch('https://github.com/QuantEcon/data-lectures/raw/main/lectures/mpd2020.xlsx') rejects; the raw.githubusercontent.com equivalent resolves.

The data-lectures audit classifies references by org/repo across all GitHub URL forms, so these stay pattern data-lectures/unchanged — the strict migration-consistency check is unaffected.

🤖 Generated with Claude Code

[netlify]

❌ Deploy Preview for beautiful-dodol-cb9543 failed.

Name	Link
🔨 Latest commit	b8acd44
🔍 Latest deploy log	https://app.netlify.com/projects/beautiful-dodol-cb9543/deploys/6a744a858ce6440009ce2120

[Copilot]

Pull request overview

Updates a handful of in-lecture, in-browser (Pyodide) pandas data reads to use CORS-friendly GitHub content hosts so fetches succeed under browser CORS rules (especially across redirects).

Changes:

• Switched data reads in long_run_growth and inflation_history from github.com/<org>/<repo>/raw/... to raw.githubusercontent.com/... to avoid redirect/CORS failures.
• Switched LFS-backed CSV reads in inequality and mle from github.com/<org>/<repo>/raw/... to media.githubusercontent.com/... to ensure the browser receives file bytes (not a blocked redirect or LFS pointer path).

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.

File	Description
lectures/mle.md	Repoints SCF CSV read to media.githubusercontent.com for CORS-clean in-browser fetch.
lectures/long_run_growth.md	Repoints Maddison XLSX read to raw.githubusercontent.com to avoid CORS/redirect issues.
lectures/inflation_history.md	Repoints two Excel reads to raw.githubusercontent.com to keep browser fetches CORS-clean.
lectures/inequality.md	Repoints SCF CSV read to media.githubusercontent.com for CORS-clean in-browser fetch.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[github-actions]

📖 Netlify Preview Ready!

Preview URL: https://pr-54--beautiful-dodol-cb9543.netlify.app (b8acd44)

✨ Browse the preview at the URL above.