Skip to content

docs: fix case-sensitive RS BOM links breaking the link check on main - #61

Merged
ZhuYaoHui1998 merged 1 commit into
Seeed-Projects:mainfrom
Chirag6722:fix/rs-readme-link-case
Aug 22, 2026
Merged

docs: fix case-sensitive RS BOM links breaking the link check on main#61
ZhuYaoHui1998 merged 1 commit into
Seeed-Projects:mainfrom
Chirag6722:fix/rs-readme-link-case

Conversation

@Chirag6722

Copy link
Copy Markdown

Closes #60

What this changes

Two links, one line each:

File Was Now
README_Fr.md:159 ./hardware/reBot_B601_RS/readme.md ./hardware/reBot_B601_RS/README.md
README_JP.md:158 ./hardware/reBot_B601_RS/readme.md ./hardware/reBot_B601_RS/README.md

The RS hardware folder holds README.md in uppercase; the DM folder beside it uses lowercase readme.md. GitHub serves repository paths case-sensitively, so both links 404 today for anyone reading the French or Japanese README and clicking through to the RS bill of materials.

This is what turned the weekly link check red on main in run 32002320359 — the only two errors out of 443 unique links.

Why #58 missed it

Worth stating plainly, since it bears on how much weight to give that PR's verification: the checker I ran to confirm "0 broken relative links" ran on Windows, whose filesystem is case-insensitive. hardware/reBot_B601_RS/readme.md resolves to README.md there and reports as a working link. On ubuntu-latest it does not. So the relative-path claim in #58 was sound for missing files but blind to case — and the CI job added in that same PR caught it on its first scheduled run, which is a reasonable argument for the job's existence.

Verification

  • Re-ran the audit case-sensitively this time, matching every relative link against a case-exact index of real repository paths: 0 failures across all markdown, down from 2. That is the check that should have run the first time.
  • Both replacement targets exist as hardware/reBot_B601_RS/README.md.
  • No other case mismatches anywhere in the repository — these two were the whole set.

After this merges, the next scheduled Check Documentation Links run (Mondays, 06:00 UTC) should go green.

One suggestion, not included here

This failure sat red on main from Monday to Thursday without action. A scheduled check nobody looks at is close to no check. lychee documents an if: failure() step using peter-evans/create-issue-from-file that turns a scheduled failure into an issue. I have deliberately left it out of this PR — it is a separate decision about how noisy you want CI to be, and I would rather you choose it than inherit it. Say the word and I will send it.

🤖 Generated with Claude Code

The weekly link check has been failing on main since 2026-08-17. Two
links, in README_Fr.md and README_JP.md, target
./hardware/reBot_B601_RS/readme.md, but that directory holds README.md
in uppercase — the RS folder uses README.md while the DM folder beside
it uses readme.md. GitHub serves paths case-sensitively, so both 404.

These slipped through review in Seeed-Projects#58 because that verification ran on a
case-insensitive Windows filesystem, where readme.md resolves to
README.md and reports as fine. The CI job runs on ubuntu-latest, which
is why it caught them on its first scheduled run.

Closes Seeed-Projects#60

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
@ZhuYaoHui1998
ZhuYaoHui1998 merged commit c26afc6 into Seeed-Projects:main Aug 22, 2026
1 check passed
ZhuYaoHui1998 pushed a commit that referenced this pull request Sep 7, 2026
Merge commit 11f09f7 resolved README conflicts in favour of a local
branch that predated #58, #61 and #63, reverting all of their README
changes. The same merge dropped LAN-GER's community row for the Isaac
Sim-to-Real VLA course (af00e42), whose GIF is still in community/ with
nothing pointing at it. Nothing was force-pushed — the commits are all
still ancestors of main; the content was resolved away by hand across
five large files, which leaves no obvious trace.

Restored, by re-applying onto current main rather than reverting the
merge, so the newer work in it survives (the DLI course link, the Star
Arm 102 purchase links, the isaacsim wiki link):

- Feishu redirect back to the AliExpress product (4 files)
- Sensor/IMU 404 back to Sensor_accelerometer (5 files)
- shields.io badges re-escaped: the literal space and the raw emoji
- readme.md/#about-power-supply trailing slash (4 files)
- RS BOM links back to README.md, which is the real filename (2 files)
- zh: reach back to 767 mm, reSpeaker row restored, RS Isaac Sim back to
  completed; fr and jp likewise
- jp: the RS roadmap table back to Japanese from the French copy
- LAN-GER's VLA course row, in each language's own wording

Neither check could have caught this: both only run on pull_request, and
these commits were pushed straight to main. link-check would have waited
for its Monday schedule; readme-sync has no schedule, so it would never
have run at all. Both now also run on push to main, and readme-sync gets
the same weekly schedule link-check has.

Closes #79

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Link check failing on main: two RS BOM links use readme.md instead of README.md

2 participants