Repository navigation
docs: point citations and architecture at the published research repo - #99
Merged
Merged
Conversation
…d research repo
Eleven tracked files carried citations into `research/`, which has always been
gitignored — so every one of them pointed at a file that does not exist for
anyone who clones this repo. They now link into
NovaLux12/carelink-api-research, which is the published, scrubbed, citable
surface.
- src/{carelink/client,discovery,last-alarm,login-errors,refresh-failure,retry-policy}
and five test files: research notes -> the public repo's findings-log.
- .gitignore: explain where the research went and why the local copy stays
ignored (it carries material that must not be published).
- docs/vendor-api/README.md: new pointer page — the handful of pages a bridge
maintainer actually needs, plus a paragraph on what this bridge relies on.
- docs/ARCHITECTURE.md: add a pointer block at the top of "Live CareLink API
surface" naming what the research repo has added or MOVED since that section
was written (the unauthenticated config file, the v2 data plane, the reports
namespace relocating to the patient host, the second front-end app, five hosts
carrying the clinician bundle). The section itself is kept intact — its
bridge-specific reasoning and its deliberately-recorded corrections are worth
more than a shorter file.
No behaviour change. Suite 323/323, typecheck clean.
This was referenced Oct 11, 2026
NovaLux12
added a commit
that referenced
this pull request
Oct 11, 2026
…link (#101) #99 replaced the `research/` .gitignore rule with a comment asserting the directory 'stays gitignored' — and dropped the rule itself. Nothing was ignored any more: 1,170 files became stageable by a plain `git add -A`, including jsessionid_ and .corp hostname hits that are on the research repo's scrub list (research-repo AGENTS.md §7). Restore the rule and add the README orientation block linking the canonical map, which #99 left out of the README (it covered ARCHITECTURE, .gitignore and docs/vendor-api/README.md, but README.md still had zero references). Also adds a local research/README.md declaring the mirror non-canonical. Gitignored by design, so it is not part of this diff. Verified: research/ ignored again (0 stageable), 323/323 tests pass (identical to clean-main baseline), no .ts touched. Co-authored-by: NovaLux12 <nova@luxmail.co.uk>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Eleven tracked files cited
research/paths — andresearch/has always been gitignored, so every one of those citations pointed at a file that does not exist for anyone who clones this repo. This points them at the published research repo.Changed
src/andtest/(11 files) — source-of-truth comments now link intoNovaLux12/carelink-api-research/blob/main/findings-log/…instead of a gitignored local path..gitignore— explains where the research went and why the local copy stays ignored: the local notes carry material that must not be published (a vendor API key, internal hostnames, raw archive captures). The research repo is the scrubbed, citable surface.docs/vendor-api/README.md(new) — a pointer page: the six pages a bridge maintainer actually needs, plus a short paragraph on what this bridge relies on and why.docs/ARCHITECTURE.md— a pointer block at the top of "Live CareLink API surface", listing what the research repo has added or moved since that section was written.What the research repo says has moved
Section 5 of that doc was accurate when written. It has since been superseded in places, and it should say so:
/patient/v2/configuration/publicis unauthenticated — 69 keys, including the full device-family vocabulary/patient/v2/monitor/datais live, 405 POST-only — the v2 tree is a data plane, not just config/patient/reports/*is live oncarelink.minimed.eu— so the reports surface behind Unintegrated Medtronic reports API (commoncore + hcp/*) — likely source for the v0.2.0 deferred 780G fixtures #76 is notcommoncore-only.commoncore.medtronic.*is now CloudFront 502 "could not resolve the origin domain name", an origin DNS failure rather than an absence/patient/*endpoints not in the clinician bundleKept intact on purpose
The "Live CareLink API surface" body is not trimmed. Its bridge-specific reasoning (
dataHostCandidates(), the sibling-host rule, the November validation item) and its deliberately-recorded corrections — including the note that one row was "corrected twice in the wrong direction" — are worth more than a shorter file. The pointer block adds context; it does not replace hard-won analysis.No behaviour change.
323/323, typecheck clean.🤖 Generated with Claude Code