Repository navigation
Add honest Kotoba v1 E.164-looking check surface - #4072
com-junkawasaki wants to merge 11 commits into
Conversation
Sibling kotoba/ tree with a plus-prefix digit-length check, two testdata accept fixtures, two reject fixtures, and fork CI for kotoba CLI 0.7.2 wasm32 i64-v1. Not a metadata/carrier/geocoding port. Co-authored-by: Jun Kawasaki <04.feasts_minded@icloud.com>
Add pull_request path filters: kotoba-v1.yml runs on kotoba/** only; java-test, cpp-test, CodeQL, and the other jobs this PR woke ignore kotoba-only diffs. Co-authored-by: Jun Kawasaki <04.feasts_minded@icloud.com>
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
…phonenumber-v1-5285
com-junkawasaki
left a comment
com-junkawasaki
left a comment
There was a problem hiding this comment.
Honest review (author self-triage)
Code review. kotoba/phonenumber.kotoba is an honest "E.164-looking" syntactic check surface — it deliberately does not claim numbering-plan validity (README and comments say so). Verified:
- Fixtures decode to exactly the strings asserted in
checks.sh(+165****0000,+442****3000), and the accept/reject rows are cross-checked against an independent predicate (regex-shaped, not a re-call of the binding), so the test is not circular. - Digit-count arithmetic (2..15 digits after
+) is internally consistent through the checked trace. kotoba-v1.ymlpins the toolchain properly:actions/checkoutandstep-security/harden-runnerare pinned to full SHAs which I verified against upstream tags viagit ls-remote(checkout v7.0.1 =3d3c42e5…, harden-runner v2.20.0 =bf7454d0…), and the kotoba CLI tarball is digest-pinned. No invented identifiers.
Fixed on our side. The branch was BEHIND master; I merged upstream/master in a worktree (4f9ce21e, no conflicts, metadata-only upstream churn) and re-ran kotoba/checks.sh locally after the merge — accept/reject fixtures, wasm emit and web run all pass. PR is now MERGEABLE.
What blocks this PR (exactly):
cla/googleis FAILURE — the Google CLA is unsigned. Signing a CLA is a legal act and is reserved for the repository owner; we will not sign it. This needs the owner to complete the Google CLA.- The fork-triggered workflow (
Kotoba v1) is inaction_required— a maintainer must approve the workflow run before CI results appear. - Required maintainer review (we have no upstream write access).
Honest Kotoba v1 sibling under
kotoba/. Plus prefix plus digit length 2..15 only. NotisValidNumber, not metadata/carrier/geocoding, not a Java/C++/JS replacement, no FFI.Fork PR: kotoba-lang#1
kotoba-oss QA observed the fork job
kotoba compile + fixtureSUCCESS. This PR does not claim a wasm byte size.Google CLA may apply; it is not filed here.
Test plan
bash kotoba/checks.sh(requireskotoba.cli/code == emittedand no wasm import section)+16502530000/+442070313000; reject spaced international and NSN