mosaicstack/stack
25
0
Fork 1
Code Issues 340 Pull Requests 13 Actions Packages Projects Releases 15 Wiki Activity

feat(queue): Piece D, reviews as issue comments, raw per-seat token helper (row 12, #1508)

Browse Source 
queue move ID in-review posts the review request as a Gitea comment and
review record reads verdicts back, so reviews stop being files in
docs/plans/reviews/. On a comment round, in-review to waiting-on-jason
now needs every listed reviewer's approval for the current round, the
same as in-review to done (Filbert r1 C1). scripts/gitea-api.sh reads
the raw per-seat token files (lead decisions 37 to 39): config built and
checked before curl starts, export attribute cleared, fixed base URL.
test-queue.sh skips its live checks outside the canonical root.

Darkwing authored. Filbert approved D r2 (cf1d3fd0) after r1 (a2dc2302)
and corrected the plan (293747cd). Rocko reviewed the helper (e896192f,
2096b0a3), and Sage's lead check passed under decision 38. Manifest
b402fb38, 19 files.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
This commit is contained in:
jason.woltje
2026-09-27 10:07:29 -05:00
co-authored by Claude Opus 5.5
parent cdcedb2741
commit f539466fcb
19 changed files with 2541 additions and 78 deletions
Download Patch File Download Diff File Expand all files Collapse all files
agents/darkwing/work/queue-d/build.md
+247
View File
@@ -0,0 +1,247 @@
# Queue D (#1508, row 12), candidate for review, round 2
Darkwing, 2026-09-27. Piece D is review requests, per plan section 8.9
(`agents/filbert/work/queue-as-data-plan-2026-09-26.md`). Moving a row
with reviewers to in-review posts one request comment on its issue as the
acting seat, and the log records what happened. The row closes on the
reviewers' recorded verdicts. The candidate also carries the DEFERRED
item Sage attached to rows 12 and 13: `test-queue.sh` skips its live checks
outside the canonical root.
Base is 8ffbd73b. The patch applies unchanged to HEAD cdcedb27, which
since then has changed only `lead-decisions.md`, `QUEUE.md` and
`queue.json`. Filbert reviews D. The helper patch is separate
(`helper.md`); Sage approved it in lead decision 39, and it ships in the
D commit. Sage commits. Nothing is committed, staged or pushed.
Round 2 fixes Filbert's C1
(`agents/filbert/work/queue-d-review-r1-2026-09-27.md`): on a request
round, in-review→waiting-on-jason now makes the same checks as
in-review→done. See "Waiting-on-jason on a request round" below.
## Files
`build.patch` (sha256
`28d0790e797e1f24f295f70b2e0162bb1b4b0a154b220c0527f7d6624f0aa1c8`) changes 7 files and adds 3.
`build-manifest.sha256` (sha256
`af57ead2697a88d74a9214783dfa35ba3bd171aa32a594584f018181e261e956`) pins all 10 after the
patch. In a fresh clone at cdcedb27 the patch applies and the result
matches the manifest 10/10.
- `packages/queue/src/review.mjs` (new): the credential check, the request
body, the helper call under the deadline, and the reading of each
answer.
- `packages/queue/src/queue.mjs`: `SEMANTICS` 2, the five review verbs,
the round, attempt and receipt shapes, and the rules for moves and done.
- `packages/queue/src/store.mjs`: the request step after the move is
logged, the resolve check, `review verify-commit`, and the retry and
settle hints.
- `packages/queue/src/cli.mjs`: the `review` subcommands.
- `packages/queue/README.md`: a "Review requests" section, verbs, exit 3,
known limits, tests.
- `scripts/test-queue.sh`: the live checks skip outside the canonical root.
- Tests: `review.test.mjs` (new, 20 tests), `fixtures/fake-gitea.mjs`
(new), `fixtures/kill-at.mjs` (a step can be `NAME#N`, the Nth time it is
reached), `store.test.mjs` (two A2 tests set row 9's reviewers to none
first, so their moves post nothing; a note there moves from filbert to
sage, since filbert is no longer a reviewer). 122 tests at A2, 142 now.
Outside the patch:
- `helper.patch` (sha256 `48edd46b93c54908e9d59737aab78a33e007ddfafa1e6ad4d34333015bb75430`): raw per-seat token files in
`scripts/gitea-api.sh`, lead decision 37. Rocko reviewed two rounds,
and Sage approved the result (decisions 38 and 39); notes in
`helper.md`. D's tests don't need it, but the live round does.
- `tools-md.patch` (sha256 `d30d65b842bd56077ccd9164eccd946f2aa9f53290d02eba61f284960d230e14`): `docs/TOOLS.md`, the review
commands and the raw token file. It is Sage's file, so it's a proposal.
## How a request goes
1. **Intent.** `move ID in-review --candidate C --op OP` on a row with
reviewers logs the round and its first attempt, `requesting`, under the
lock, like any other op. If that write doesn't finish, nothing is sent.
2. **Pre-send.** Without the lock: the credential file check (`lstat`
only), then `GET user`, which must return the acting seat's login, or
`jarvis` for sage. A failure here is `failed`, and the POST never runs.
3. **POST** one comment on the round's issue through
`scripts/gitea-api.sh`, under `timeout -s KILL 30`. The body carries
two markers: `<!-- mosaic-queue-op: OP -->` and
`<!-- mosaic-queue-round: row=N round=R candidate=DIGEST -->`.
4. **Outcome.** A second entry, `OP.outcome`, under the lock. It holds the
HTTP status, the comment id and a fixed detail string. Nothing from the
response body is logged or printed.
Exit 0 is posted, 1 is failed (400, 401, 403, 404, 422, or pre-send), and
3 is uncertain (anything else). A same-op retry prints the logged state and
sends nothing. An uncertain or `requesting` attempt prints where to look
and the two commands that settle it.
## Choices I made
- **The CLI shape of `resolve`.** The plan has
`review resolve ID --attempt REQOP --posted <id>`. I made it
`review resolve ID REQOP --comment N`, and abandon takes the attempt the
same way. The attempt is always required, and `--comment` is the flag
name `record` uses for a comment id too.
- **Resolve checks the author.** The plan lists issue, marker, round and
candidate. I added the comment's author: it must be the requester's
login, fetched with the resolver's own token. Otherwise anyone could
copy the markers into a comment of their own.
- **Done on a request round** needs an approval recorded by every listed
reviewer in the current round, and refuses `--evidence`. If the reviewers
were removed after the round opened, done refuses until a privileged
actor sets them. A round from before D, or on a row with no reviewers,
closes with `--evidence` as in A2.
- **Waiting-on-jason on a request round** (round 2, C1). The plan's table
let the owner move in-review→waiting-on-jason with no other check, so a
Jason-gated row could reach Jason, and then close, with no reviewer's
verdict. The move now refuses while a request is unresolved, and on a
request round it needs an approval recorded by every listed reviewer.
`requireApprovals` in `queue.mjs` holds the approval checks that done
and this move share. A round with no request is unchanged, so v1 replay
is too. Filbert approved round 2 and asked for one more check (n1): an
approval from an earlier round doesn't count in the current one. The
Jason-gated test now has rocko approve round 2 first, and the move
refuses until filbert approves round 2 as well. Jason gets no exemption, because the owner makes this move.
In-review→in-progress, the changes path, is unchanged.
- **The owner records no verdict,** even when listed as a reviewer.
- **A late outcome on a done row.** `review-outcome` is the one entry a
done row accepts, so a POST that answers after the row closed is still
recorded. It can only set `conflict`; nothing reopens the row, and
`review resolve` on it refuses without fetching the comment.
- **A late `uncertain` after a resolve** keeps `posted`: the resolve saw
the comment. A late `failed` after a resolve is `conflict`, because the
server said it refused a comment someone found.
- **Semantics per entry.** Every log entry already records `semantics`.
Entries at 1 replay under A2's rules, so the live log (rev 13, all
semantics 1) loads unchanged. Review verbs need 2, checked in the entry
shape and again in apply.
- **Body limit.** A body over 60,000 bytes is a pre-send `failed`, and
nothing is sent. A 700-line manifest is enough to reach it.
- **`verify-commit` on a commit candidate** compares every path the
candidate changed against its first parent, by blob and mode, and
requires the paths it deleted to be absent in REF. A `diff-tree` line it
can't parse refuses with exit 1 instead of being skipped.
- **The claim refusal message.** `ownerOrPriv` read "darkwing cannot
resolve a request on". It now names the owner and the privileged actors
in every case, with the claim first when there is one. No A2 test
matched the old text; the resolve test pins the new one.
- **`test-queue.sh`.** After genesis it reads `canonicalRoot` from
`HEAD:docs/plans/queue.json` and compares it with the real path of the
toplevel. If they differ, it prints
`skip queue verify and render --check: this checkout (TOP) is not the queue's canonical root (CANON)`
and passes. An empty `canonicalRoot` is a failure.
## Tests
`review.test.mjs` runs each case in a scratch repository with genesis
committed. `scripts/gitea-api.sh` there is a stub that runs
`fixtures/fake-gitea.mjs`: same argv and output as the helper, calls
logged to a file, rules from a scenario file, and posted comments served
back by id. The token files are dummies under the scratch directory. No
test reads a real token or opens `~/.mosaic` or `~/.t3`.
Against the test list in 8.9:
| 8.9 asks for | Test |
|---|---|
| a kill before the POST, after it, at the outcome write, while retaking the lock | "a same-op retry after a kill sends nothing, even with a stale view" (steps `pre-send`, `posted`, `outcome`, `locked#2`), and "a held lock at the outcome exits 3" |
| posted; each listed 4xx; 5xx; request failed; timeout; 201 without an id | "each transport answer maps to posted, failed or uncertain" (deadline 300 ms) |
| a new op while `requesting` or `uncertain` refuses | "an unresolved request blocks a new request, a new round, waiting-on-jason and done", which also reaches waiting-on-jason with a late POST still out, and Jason's close refuses |
| (round 2, C1) a Jason-gated row skips its reviewers | "a Jason-gated row reaches waiting-on-jason only on every reviewer's approval": Filbert's steps refuse, then pass after both reviewers approve a later round; reviewers removed refuse too |
| a same-op retry after a stale view sends nothing | the kill test |
| a late POST after abandon; a late outcome after resolve, same id and another | "late outcomes": rounds 1 to 3, plus a late 500 (stays posted) and a late 422 (conflict); "a late POST on a closed row" (abandoned, approved and closed while the POST was out) |
| resolve with the wrong issue, marker, round or candidate | "resolve checks the comment", which adds author, id, 404, 500, a transport failure and the wrong credential |
| `GET user` mismatch and timeout | "the pre-send checks" |
| the lead posts as jarvis; `sage` as a login refuses | "the pre-send checks" and "the lead's request refuses a token for login sage" |
| the credential checks: mode, another seat's path, the default file | "the credential file" (also unset, relative, missing, symlink, a linked directory, and jarvis's path for sage) |
| request, changes, new candidate, approval, pinned | "request, changes, a new candidate, approval", which also checks no file lands under `docs/plans/reviews/` or `agents/*/work/` |
Every test that retries, kills or answers late counts the POSTs the fake
saw. None sees a second POST for one attempt.
## Mutations
61 hand-written mutants: 29 in `queue.mjs`, 16 in `store.mjs`, 16 in
`review.mjs`. Each runs alone against the whole queue test directory.
Against the round 2 tests, 59 are killed, the same result as round 1.
Round 2 adds six in `queue.mjs` for C1, run against the round 2 tests.
All six are killed:
- the round 1 branch restored (waiting-on-jason checks only the actor);
- the approval check dropped from waiting-on-jason;
- the unresolved check dropped from waiting-on-jason;
- the unresolved check dropped from waiting-on-jason→done;
- the no-reviewers check dropped from `requireApprovals`;
- the approval check applied to every round, not only request rounds.
Seven survived earlier runs. Five of them now have a test that kills them:
a second outcome for one attempt, the owner recording a verdict, done on a
row whose reviewers were removed, a same-op resolve retry that fetched the
comment again, and a resolve on a done row that fetched the comment before
the log refused it. The last one is the new test "a late POST on a closed
row".
Two survive, and each is equivalent while the other stays:
- the semantics check in `applyEntry` dropped;
- the semantics check in the entry shape dropped.
Every entry passes the shape check before it is applied, so either check
alone refuses a review verb at semantics 1. With both dropped, the
semantics test fails.
## Suites
In the scratch clone with the patch and the helper patch applied: config
24/0, task 90/0, foundation 44/0, conductor 17/0, release 14/0, auth 15/0,
discord 64/0, extension-package 18/0, queue 27/0. `node --test
packages/queue/tests/` passes 142, and `node --test packages/ledger/tests/`
passes 58.
The scratch clone isn't the canonical root, so the queue suite printed its
skip line for the two live checks. In the canonical tree they run.
In a fresh clone at cdcedb27 with `build.patch` applied: the manifest
matches 10/10, the queue tests pass 142, the queue suite 27/0. With
`helper.patch` applied on top, the ledger tests pass 58.
## Known limits
- **Scope of darkwing's and dewey's tokens.** They hold
`write:repository`. If that doesn't cover an issue comment, the live
POST gets 403, which is `failed` with nothing posted (8.9 says so). The
scope question then goes to Sage.
- **Post to outcome.** A kill between the POST and the outcome entry
leaves the attempt `requesting` with a comment on the issue. The queue
never posts again on its own; a person resolves it.
- **Verdict comments aren't fetched.** `review record` takes the
reviewer's comment id on trust, as the queue takes `--by`.
- **Commit candidates** stay checkable only while a ref keeps the commit.
- The helper's own limits are in `helper.md`.
## After approval (Sage)
1. Apply `build.patch` to the canonical tree and check it against
`build-manifest.sha256`. Run `scripts/test-queue.sh` and
`node --test packages/queue/tests/`. Commit the 10 files by path.
2. Apply `helper.patch` for the same commit (decision 39). The live
round needs it: without it the helper refuses a raw token file.
3. `tools-md.patch` if you want it.
4. Row 12 lists no reviewers, so moving it to in-review now would open a
round that posts nothing. Set them first:
`scripts/mosaic queue set 12 reviewers filbert --op OP --by sage`, and
add rocko if the helper counts as part of this round.
The live round, which is mine:
5. `MOSAIC_GITEA_CREDENTIAL_FILE=~/.mosaic/fleet/agents/darkwing/secrets/gitea-mosaicstack-darkwing.token
scripts/mosaic queue move 12 in-review --candidate <D's commit> --op OP --by darkwing`.
The queue `lstat`s the file, the helper reads it, and `GET user` must
answer `darkwing`. The request goes on #1508.
6. Exit 0 means posted. Exit 1 with HTTP 403 is the scope question above.
Exit 3 means I look on #1508 for the marker and resolve, or ask you to
abandon.
7. Filbert posts a verdict on #1508 and runs `review record 12`. Then
`move 12 done`, and `queue-commit.sh` for the queue ops.
agents/darkwing/work/queue-d/helper.md
+184
View File
@@ -0,0 +1,184 @@
# Raw per-seat token files in `scripts/gitea-api.sh` (row 12, #1508)
Darkwing, 2026-09-27, after round 2 (lead decision 38). Lead decision 37 approved this patch. It ships next
to the Piece D candidate but is a separate patch, and Rocko reviews it
because it handles a credential. Filbert reviews D. Sage commits both.
Nothing is committed, staged or pushed.
`helper.patch` (sha256 `48edd46b93c54908e9d59737aab78a33e007ddfafa1e6ad4d34333015bb75430`)
changes `scripts/gitea-api.sh` and adds
`packages/ledger/tests/gitea-helper-raw.test.mjs`. It applies to HEAD
8efc0ff3 by itself. With only this patch applied, `node --test
packages/ledger/tests/` passes 58/58. D does not depend on it, and it does
not depend on D.
## Why
Each seat's Gitea token is at
`~/.mosaic/fleet/agents/<seat>/secrets/gitea-mosaicstack-<seat>.token`.
These files hold the bare token, not the JSON `mosaic.gitea.json` shape
that the helper reads today. Without this change the helper refuses them,
so D's live round can't post as the seat.
## What changed
Both old node snippets, the one that printed the base URL and the one that
wrote the curl config, are now one script, `CRED_JS`, run in two modes:
`base` and `cfg`. Each mode runs every check before it prints anything.
1. `lstat`: a regular file, not a symlink, no group or other bits.
Unchanged.
2. Read the file. A read error exits 3. It used to be caught by the JSON
parse's catch, with the same result.
3. `JSON.parse`. If the text parses, the JSON path runs unchanged: `url`
must be `https://git.mosaicstack.dev` (trailing slashes stripped), and
`api_token` must be a nonempty string. A non-`SyntaxError`, such as
`null.mosaicstack`, exits 3.
4. The raw path runs only on a `SyntaxError`. It accepts the file only if
all of these hold:
- `stat` size is 40 or 41;
- the byte length of the text equals the `stat` size;
- the text matches `^[0-9a-f]{40}\n?$`.
Then the base URL is the fixed string `https://git.mosaicstack.dev`,
and the token is the first 40 characters. Anything else exits 3
before any request.
5. The second read runs, and must succeed, before curl starts. Round 1
ran it inside `curl -K <(…)`, where bash drops its exit status, so a
refusal gave curl an empty config and the request still went out.
Now `CFG="$(node -e "$CRED_JS" cfg)" || exit 3` and `[ -n "$CFG" ] ||
exit 3` run first, and curl reads `-K <(printf '%s\n' "$CFG")`.
`printf` is a builtin, so the token is in no argv. Both curl branches,
with and without a body, use it.
6. `export -n CFG` runs right after the checked assignment, before any
child starts. A `CFG` inherited from the caller's environment keeps its
export attribute when assigned, and `SHELLOPTS=allexport` in the
environment exports every assignment. Either way the config would reach
curl's environment, and the body file's `mktemp` and `chmod` too.
## Decision 37, condition by condition
1. **The JSON path is unchanged.** The last test runs a good JSON file,
four refused ones, and content that parses as JSON but would pass the
raw pattern: 40 decimal digits, with and without a newline. Those take
the JSON path and refuse.
2. **One line of token characters.** The format is 40 lowercase hex
characters. I checked by `stat` that the real files are 40 bytes (jarvis)
or 41 (the others), all mode 600. I then ran the patched `CRED_JS` in
`base` mode against my own file only. It printed
`https://git.mosaicstack.dev` and exited 0, so darkwing's file matches
the pattern. Nothing printed the token, and I read no other seat's file.
The other seats' files are unconfirmed beyond size and mode. If one
doesn't match, the helper exits 3 before any request. The test refuses
16 bad contents before curl runs: empty, 39 characters, 41 characters,
upper case, CRLF, a trailing CR, a trailing space, a trailing tab, two
newlines, a leading space, a trailing space at 40, a second line, a
quote, non-hex, non-ASCII, and 80 characters.
3. **Fixed base URL.** The raw path assigns the literal. The test sets
`MOSAIC_GITEA_URL`, `GITEA_URL` and `MOSAIC_GITEA_BASE_URL` to another
host, and curl still gets `https://git.mosaicstack.dev/api/v1/user`.
4. **File checks and the config stream.** Modes 640, 604, 660, 644, 000
and 200, a symlink, a missing file and a directory all refuse before
curl runs. 000 and 200 pass the group and other check, and the read
refuses them. The stub curl records its argv and its `-K` stream. The
token is in the stream only, never in argv, stdout or stderr. Every
token in the test is a dummy that the test writes.
5. **Only the acting seat's file.** That is D's job, not the helper's. D
refuses unless `MOSAIC_GITEA_CREDENTIAL_FILE` resolves to
`…/agents/<login>/secrets/gitea-mosaicstack-<login>.token` for the
acting seat, or jarvis when Sage acts, and before posting it checks that
`GET user` returns that login. See `review.mjs` `credCheck` and
`checkUser` in the D candidate.
6. This review.
## Round 1 and what changed for round 2
Rocko's round 1 (`agents/rocko/work/gitea-helper-raw-r1-review-2026-09-27.md`)
found one blocker: a second read that refused still let curl run with an
empty config, and a 2xx then exited 0. My round 1 notes called that a 401,
which the server doesn't guarantee. Sage agreed it belongs in this patch.
Step 5 above is the fix.
The new test, "a file that changes between the two reads refuses before
curl runs", uses the git stub, which runs between the two reads, to
change a valid dummy file to invalid text, to `{}`, to mode 644, to a
symlink, and to missing. It runs each change for `GET user` and for a POST
with a dummy body. Every case exits 3 with no curl call and nothing on
stdout. It also changes a valid JSON file to invalid text. As a control,
both calls reach curl with the right config when nothing changes. The
first test now also checks that the token is not in the environment curl
gets.
Rocko's round 2 (`agents/rocko/work/gitea-helper-raw-r2-review-2026-09-27.md`)
closed that blocker and found another: the inherited export attribute in
step 6. Lead decision 38 took the fix, `export -n CFG`, with no round 3;
Sage checks it with Rocko's reproducer. The test "the token reaches no
child environment, even with an inherited CFG or SHELLOPTS=allexport" runs
raw and JSON dummy files, GET and POST, with an exported harmless `CFG`,
with `SHELLOPTS=allexport`, and with both. Each call must exit 0 with the
token in curl's config stream and not in its environment, argv, stdout or
stderr. I found the `SHELLOPTS` route while checking the fix; `export -n`
covers it too, so I added no second line for it.
Not in this patch: the predictable response path
`/tmp/gitea-api-response.$$`. Sage has it on DEFERRED as "Gitea helper
response-file hardening".
## What I'd like you to look at
- **The second read now re-validates.** Before, `gen_curl_cfg` parsed the
file again without the `lstat` checks. Now both reads run every check.
So a file swapped for a symlink or a wider mode between the two reads
refuses. On the JSON path this is a tightening.
- **The token now sits in a shell variable** for the rest of the script.
`export -n` keeps it out of every child's environment, and nothing
prints it. Before, it existed only in the pipe.
- **`lstat` then `readFile` is still not atomic.** A swap between those two
calls inside one read is the pre-existing race Rocko noted. This patch
doesn't claim to close it.
- **`text.slice(0, 40)`** relies on the pattern having matched. The only
accepted texts are the token alone or the token plus `\n`.
## Mutations
25 mutants, each run alone against `gitea-helper-raw.test.mjs`: 19 of
`CRED_JS` and 6 of the gate and the export. 15 are killed:
- uppercase allowed, any trailing whitespace allowed, the `^` anchor
dropped;
- a raw file refused outright (the old behaviour);
- an env override of the raw base URL;
- the token taken with its newline;
- the mode check dropped;
- the JSON path's host check or token check dropped, or its `|| {}`;
- `cfg` output in `base` mode;
- the read's own `try` dropped (killed by the 000 and 200 modes; I added
those after this mutant first survived);
- both gate checks dropped, and the round 1 code (no gate, curl reading
`<(node … cfg)`), killed by the between-reads test;
- `export -n CFG` dropped, killed by the inherited-environment test.
Ten survive, and each is equivalent while the other checks stay:
- `\n*$` for `\n?$`: the size check caps the file at 41 bytes;
- the size check dropped: the pattern caps the length;
- the byte-length check dropped: the pattern is ASCII, so a match means
bytes equal characters; the check differs only if the file changes
between `lstat` and the read;
- the `SyntaxError` test dropped: the only non-`SyntaxError` is JSON
`null`, which then fails the raw pattern;
- `text.trim()` for `text.slice(0, 40)`: same result on every accepted
text;
- the `isSymbolicLink` test dropped: `lstat` of a symlink is never
`isFile`;
- `!== "base"` for `=== "cfg"`: the script calls only these two modes;
- `|| true` for `|| exit 3` on the `CFG` line: node prints nothing when it
refuses, so `[ -n "$CFG" ]` refuses;
- `[ -n "$CFG" ]` dropped: `|| exit 3` refuses first. Dropping
`|| exit 3` outright is the same, since `set -e` exits on the failed
assignment;
- `export CFG=…` for `CFG=…`: `export -n` on the next line clears it. In
round 2 this one was killed; the fix makes it equivalent.
I kept the size, byte-length and symlink checks anyway. Decision 37 asks
for size plus pattern, and the other two cost nothing.