Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
58 changes: 0 additions & 58 deletions quest/m0/ffi-shutdown-backport.md

This file was deleted.

23 changes: 23 additions & 0 deletions quest/m0/test-lints-hang.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,23 @@
# [XS] test-lints.test.py stalls just check on a busy host

## Goal

`python3 sh/rs/test-lints.test.py` finishes in seconds however loaded the
host is, so `just gh test`, which every `just check` runs, never stalls on it.

## Plan

Seen 2026-11-11: on `origin/main` (f190d90b34) the script prints one `.` and
then makes no progress for 5 minutes; inside `just check` it ran 80 minutes
before being killed. The same day another `just check` passed it on a quieter
host.

Decided 2026-11-11 (maintainer, /quest-unblock pick of
[moq-dev/moq#8517](https://github.com/moq-dev/moq/pull/8517)): ✅ keep this
quest, with this diagnosis (rejected: dropping it). The `.` is
`test_clippy_warning_fails_test_build`, which sets `LOCAL_SUPERVISED=1`.
`test_fail_fast_environment_reaches_nextest` does not, so each of its 20
runs (10 through `just rs test`, 10 calling `sh/rs/nextest.sh` directly) goes
through `sh/local.sh` to `quest gate` and waits for host room; on a loaded host that never comes in useful time. Set
`LOCAL_SUPERVISED=1` in its env as the clippy test does, and confirm the
script finishes while other gated work holds the host.
6 changes: 5 additions & 1 deletion quest/m2/cluster-routing/routes/playwright-pin.md
Original file line number Diff line number Diff line change
Expand Up @@ -42,4 +42,8 @@ rebuild and pin browser binaries outside nixpkgs.
[NixOS/nixpkgs#585789](https://github.com/NixOS/nixpkgs/pull/585789) (it also
fixes `update.sh` for 1.65's SHA-pinned driver version and rebases the
python driver patch). Advance it by answering that PR's review; then wait
for it to reach `nixpkgs-unstable`.
for it to reach `nixpkgs-unstable`. Decided 2026-11-11 (maintainer,
/quest-unblock): an agent drafts each review fix on the fork's branch with
an `Assisted-by:` trailer, as nixpkgs' CONTRIBUTING.md requires, plus a separate AI-disclosure line in
the reply itself, which that policy requires apart from commits; the
maintainer reviews the diff and reply before anything is pushed or posted.
Comment thread
kixelated marked this conversation as resolved.
Resolved
16 changes: 11 additions & 5 deletions quest/m3/release/dart-publish.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,15 +9,21 @@ A `moq-dart-v*` tag publishes `moq` to pub.dev with no manual step, as

Both packages are on pub.dev (checked 2026-10-25). `moq_ffi` publishes from
every `moq-ffi-v*` tag (0.4.3 through 0.4.18, the latest 2026-10-23), so its
tag-driven publish works. `moq` has only 0.1.0 (2026-09-25), the one
`moq-dart-v*` tag so far, so whether its tag publishes unattended through
trusted publishing is still unproven.
tag-driven publish works. `moq` has 0.1.0 (2026-09-25) and 0.1.1
(2026-11-11, published by a rerun of its tag's workflow, below), so whether a
tag publishes unattended through trusted publishing, with no rerun, is still
unproven.

Trusted-publishing configuration and checks of already-published artifacts
can proceed independently; cutting the tag cannot.

- Confirm trusted publishing is configured for `moq` (repository
`moq-dev/moq`, tag pattern `moq-dart-v{{version}}`).
- Trusted publishing for `moq` is fixed: its repository field read
`moq/moq-dev` (owner and name swapped), so pub.dev refused the
`moq-dart-v0.1.1` tag's token. On 2026-11-11 the maintainer set it to
`moq-dev/moq` (tag pattern `moq-dart-v{{version}}`, environment `pub.dev`,
push events), and a rerun of that tag's
[release-dart run](https://github.com/moq-dev/moq/actions/runs/43107848443)
published 0.1.1. The next tag shows whether it publishes with no rerun.
Comment thread
kixelated marked this conversation as resolved.
Resolved
- Raise `dart/moq/pubspec.yaml`'s `moq_ffi` floor to the release the wrapper now needs (the media namespace and codec types), since `^0.4.3` resolves bindings without them.
- Cut the next `moq-dart-v*` release with the binding breaks and confirm the workflow publishes it.
- Verify the published `moq_ffi` resolves its native asset from a clean
Expand Down
16 changes: 7 additions & 9 deletions quest/m4/oss-fuzz.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,15 +7,6 @@ clock, keeping its own corpus. A crash it finds reaches the maintainer
privately, and its reproducer lands in the target's `regressions/` so
`just check` replays it.

## Waiting

- Maintainer: approve the application PR to
[google/oss-fuzz](https://github.com/google/oss-fuzz) and name the contact
address it lists. An agent cannot take this: it posts outside the
`kixelated` and `moq-dev` orgs, the address receives private security
reports, and accepting commits the project to OSS-Fuzz's 90-day disclosure
deadline. Done when the application PR is open.

## Plan

Decided 2026-11-10 (maintainer, /quest-plan of the follow-ups from
Expand All @@ -26,6 +17,13 @@ Decided 2026-11-10 (maintainer, /quest-plan of the follow-ups from
- ✅ m4, after the whole-session targets exist, since stateful targets gain
the most from long runs.

Decided 2026-11-11 (maintainer, /quest-unblock): ✅ the project files list
`security@moq.dev`, the address in `SECURITY.md` (rejected: a personal
address), and ✅ the agent opens the google/oss-fuzz PR from the
maintainer's account once the maintainer has reviewed its draft files and PR
text (rejected: waiting to decide). The maintainer accepts OSS-Fuzz's 90-day
disclosure deadline.

Nightly gives each of its 22 codec targets five minutes and the two
whole-session targets (`session_lite`, `session_ietf`) fifteen. That finds
shallow bugs and replays the corpus; it rarely reaches deep states, and the
Expand Down