Repository navigation
chore: merge main into next - #1667
Conversation
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7r7LjBn2Pj9k3RHUWF
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7ng9L3hDYRnUZpQPbN
Bumps the cargo-minor-patch group in /packages/eql with 1 update: [uuid](https://github.com/uuid-rs/uuid). Updates `uuid` from 1.28.0 to 1.29.0 - [Release notes](https://github.com/uuid-rs/uuid/releases) - [Commits](uuid-rs/uuid@v1.28.0...v1.29.0) --- updated-dependencies: - dependency-name: uuid dependency-version: 1.29.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps the impact-gate-minor-patch group in /.github/impact-gate with 1 update: [lizard](https://github.com/terryyin/lizard). Updates `lizard` from 1.24.1 to 1.24.3 - [Release notes](https://github.com/terryyin/lizard/releases) - [Changelog](https://github.com/terryyin/lizard/blob/master/CHANGELOG.md) - [Commits](terryyin/lizard@1.24.1...1.24.3) --- updated-dependencies: - dependency-name: lizard dependency-version: 1.24.3 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: impact-gate-minor-patch ... Signed-off-by: dependabot[bot] <support@github.com>
Bumps the cargo-minor-patch group in /languages/typescript/packages/protect-ffi with 2 updates: [uuid](https://github.com/uuid-rs/uuid) and [tokio](https://github.com/tokio-rs/tokio). Updates `uuid` from 1.28.0 to 1.29.0 - [Release notes](https://github.com/uuid-rs/uuid/releases) - [Commits](uuid-rs/uuid@v1.28.0...v1.29.0) Updates `tokio` from 1.53.3 to 1.54.1 - [Release notes](https://github.com/tokio-rs/tokio/releases) - [Commits](tokio-rs/tokio@tokio-1.53.3...tokio-1.54.1) --- updated-dependencies: - dependency-name: uuid dependency-version: 1.29.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch - dependency-name: tokio dependency-version: 1.54.1 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch ... Signed-off-by: dependabot[bot] <support@github.com>
…th 3 updates Bumps the cargo-minor-patch group with 3 updates in the / directory: [tokio](https://github.com/tokio-rs/tokio), [uuid](https://github.com/uuid-rs/uuid) and [open](https://github.com/Byron/open-rs). Bumps the cargo-minor-patch group with 3 updates in the /languages/golang/auth/guest directory: [tokio](https://github.com/tokio-rs/tokio), [uuid](https://github.com/uuid-rs/uuid) and [open](https://github.com/Byron/open-rs). Bumps the cargo-minor-patch group with 3 updates in the /languages/golang/encrypt/guest directory: [tokio](https://github.com/tokio-rs/tokio), [uuid](https://github.com/uuid-rs/uuid) and [open](https://github.com/Byron/open-rs). Bumps the cargo-minor-patch group with 1 update in the /packages/stack-encrypt/fuzz directory: [uuid](https://github.com/uuid-rs/uuid). Updates `tokio` from 1.53.3 to 1.54.1 - [Release notes](https://github.com/tokio-rs/tokio/releases) - [Commits](tokio-rs/tokio@tokio-1.53.3...tokio-1.54.1) Updates `uuid` from 1.28.0 to 1.29.0 - [Release notes](https://github.com/uuid-rs/uuid/releases) - [Commits](uuid-rs/uuid@v1.28.0...v1.29.0) Updates `open` from 5.4.4 to 5.4.5 - [Release notes](https://github.com/Byron/open-rs/releases) - [Changelog](https://github.com/Byron/open-rs/blob/main/changelog.md) - [Commits](Byron/open-rs@v5.4.4...v5.4.5) Updates `tokio` from 1.53.3 to 1.54.1 - [Release notes](https://github.com/tokio-rs/tokio/releases) - [Commits](tokio-rs/tokio@tokio-1.53.3...tokio-1.54.1) Updates `uuid` from 1.28.0 to 1.29.0 - [Release notes](https://github.com/uuid-rs/uuid/releases) - [Commits](uuid-rs/uuid@v1.28.0...v1.29.0) Updates `open` from 5.4.4 to 5.4.5 - [Release notes](https://github.com/Byron/open-rs/releases) - [Changelog](https://github.com/Byron/open-rs/blob/main/changelog.md) - [Commits](Byron/open-rs@v5.4.4...v5.4.5) Updates `tokio` from 1.53.3 to 1.54.1 - [Release notes](https://github.com/tokio-rs/tokio/releases) - [Commits](tokio-rs/tokio@tokio-1.53.3...tokio-1.54.1) Updates `uuid` from 1.28.0 to 1.29.0 - [Release notes](https://github.com/uuid-rs/uuid/releases) - [Commits](uuid-rs/uuid@v1.28.0...v1.29.0) Updates `open` from 5.4.4 to 5.4.5 - [Release notes](https://github.com/Byron/open-rs/releases) - [Changelog](https://github.com/Byron/open-rs/blob/main/changelog.md) - [Commits](Byron/open-rs@v5.4.4...v5.4.5) Updates `uuid` from 1.28.0 to 1.29.0 - [Release notes](https://github.com/uuid-rs/uuid/releases) - [Commits](uuid-rs/uuid@v1.28.0...v1.29.0) --- updated-dependencies: - dependency-name: tokio dependency-version: 1.54.1 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch - dependency-name: uuid dependency-version: 1.29.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch - dependency-name: open dependency-version: 5.4.5 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: cargo-minor-patch - dependency-name: tokio dependency-version: 1.54.1 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch - dependency-name: uuid dependency-version: 1.29.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch - dependency-name: open dependency-version: 5.4.5 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: cargo-minor-patch - dependency-name: tokio dependency-version: 1.54.1 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch - dependency-name: uuid dependency-version: 1.29.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch - dependency-name: open dependency-version: 5.4.5 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: cargo-minor-patch - dependency-name: uuid dependency-version: 1.29.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: cargo-minor-patch ... Signed-off-by: dependabot[bot] <support@github.com>
…rough
Offline coverage in the stack package for two Protect.js migration edges:
- A Protect dataType('json') column was stored as a scalar ciphertext. The
adapter drops it when read as types.Json, so the read fails. This pins the
known gap.
- A renamed property key makes decryptModel return the raw ciphertext inside
a successful { data } rather than failing.
…ack 1.x
Every existing v2 test mints its fixtures with today's protect-ffi in v2
mode, so compatibility with ciphertext written by the old native client
(protect-ffi 0.18.1 / cipherstash-client 0.28) was argued from source and
pinned by nothing.
This suite mints items with the real @cipherstash/protect 10.2.1 and
@cipherstash/protect-dynamodb 6.0.1 (exact-pinned devDependencies of the
private e2e package) and checks them against the in-tree Stack:
- decrypt with and without storedEqlVersion: 2, single and bulk, agreeing
with Protect (bigint comes back as a native bigint); tampered ciphertext
fails
- encryptQuery(...).hm equals the stored __hmac for TextEq, IntegerEq,
BigintEq and DoubleEq; IntegerOrd and Boolean carry none; downcase is left
to the application
- the HMAC is bound to neither the table nor the DB column name
- Stack-written items still read under Protect, for rollback
- a customer-shaped acceptance run (legacy write, mixed partial update,
backfill, query, rollback), including the stale __hmac a Boolean partial
update must REMOVE
- Protect's scalar dataType('json') values, readable via client.decrypt and
backfillable to types.Json
Items are minted at test time because ciphertext and HMACs are bound to the
keyset. cipherstash-client 0.28 does not derive service hosts from the CRN,
so CS_CTS_HOST / CS_ZEROKMS_HOST are pointed at the CRN's region only while
Protect's client is built.
The suite fails rather than skips without the four CS_* credentials, so the
default e2e config excludes it. It runs from vitest.protect-compat.config.ts
as the new test:protect-compat turbo task.
turbo's strict env mode filters CS_* out of `turbo run test:e2e`, so credentialed e2e suites in this job skip. Run test:protect-compat in its own step with --env-mode=loose, as the integration-*.yml workflows do. Drop the job's ap-southeast-2 CS_ZEROKMS_HOST / CS_CTS_HOST pins: current clients derive both from the CRN, and the suite scopes its own override to Protect's client.
Earlier guidance routed Protect apps through @cipherstash/stack@0.19.0 before Stack 1.x. The direct path is now proven, so that hop is no longer required. This DEPRECATED block shows that anyone who took it lands in the same place: 0.19 re-writes the same __hmac values Protect wrote, and Stack 1.x reads and queries the result as it does the Protect item. 0.19.0 is pinned as the npm alias @cipherstash/stack-0.19. Delete the block once no customer is on 0.19.
test(dynamodb): prove protect-dynamodb 6.0.1 items read and query under Stack 1.x
… with 11 updates Bumps the production-minor-patch group with 11 updates in the / directory: | Package | From | To | | --- | --- | --- | | [posthog-node](https://github.com/PostHog/posthog-js/tree/HEAD/packages/node) | `5.55.2` | `5.57.2` | | [jose](https://github.com/panva/jose) | `6.2.12` | `6.2.13` | | [@rollup/rollup-linux-x64-gnu](https://github.com/rollup/rollup) | `4.64.5` | `4.65.0` | | [arktype](https://github.com/arktypeio/arktype/tree/HEAD/ark/type) | `2.2.8` | `2.2.9` | | [@anthropic-ai/claude-agent-sdk](https://github.com/anthropics/claude-agent-sdk-typescript) | `0.3.296` | `0.3.310` | | [@anthropic-ai/sdk](https://github.com/anthropics/anthropic-sdk-typescript) | `0.133.0` | `0.137.0` | | [@modelcontextprotocol/sdk](https://github.com/modelcontextprotocol/typescript-sdk) | `1.32.1` | `1.32.2` | | [@clerk/nextjs](https://github.com/clerk/javascript/tree/HEAD/packages/nextjs) | `7.9.14` | `7.9.20` | | [@vitest/coverage-v8](https://github.com/vitest-dev/vitest/tree/HEAD/packages/coverage-v8) | `4.1.11` | `4.1.12` | | [vitest](https://github.com/vitest-dev/vitest/tree/HEAD/packages/vitest) | `4.1.11` | `4.1.12` | | [vite](https://github.com/vitejs/vite/tree/HEAD/packages/vite) | `8.3.4` | `8.3.5` | Updates `posthog-node` from 5.55.2 to 5.57.2 - [Release notes](https://github.com/PostHog/posthog-js/releases) - [Changelog](https://github.com/PostHog/posthog-js/blob/main/packages/node/CHANGELOG.md) - [Commits](https://github.com/PostHog/posthog-js/commits/posthog-node@5.57.2/packages/node) Updates `jose` from 6.2.12 to 6.2.13 - [Release notes](https://github.com/panva/jose/releases) - [Changelog](https://github.com/panva/jose/blob/main/CHANGELOG.md) - [Commits](panva/jose@v6.2.12...v6.2.13) Updates `@rollup/rollup-linux-x64-gnu` from 4.64.5 to 4.65.0 - [Release notes](https://github.com/rollup/rollup/releases) - [Changelog](https://github.com/rollup/rollup/blob/master/CHANGELOG.md) - [Commits](rollup/rollup@v4.64.5...v4.65.0) Updates `arktype` from 2.2.8 to 2.2.9 - [Release notes](https://github.com/arktypeio/arktype/releases) - [Changelog](https://github.com/arktypeio/arktype/blob/main/ark/type/CHANGELOG.md) - [Commits](https://github.com/arktypeio/arktype/commits/arktype@2.2.9/ark/type) Updates `@anthropic-ai/claude-agent-sdk` from 0.3.296 to 0.3.310 - [Release notes](https://github.com/anthropics/claude-agent-sdk-typescript/releases) - [Changelog](https://github.com/anthropics/claude-agent-sdk-typescript/blob/main/CHANGELOG.md) - [Commits](anthropics/claude-agent-sdk-typescript@v0.3.296...v0.3.310) Updates `@anthropic-ai/sdk` from 0.133.0 to 0.137.0 - [Release notes](https://github.com/anthropics/anthropic-sdk-typescript/releases) - [Changelog](https://github.com/anthropics/anthropic-sdk-typescript/blob/main/CHANGELOG.md) - [Commits](anthropics/anthropic-sdk-typescript@sdk-v0.133.0...sdk-v0.137.0) Updates `@modelcontextprotocol/sdk` from 1.32.1 to 1.32.2 - [Release notes](https://github.com/modelcontextprotocol/typescript-sdk/releases) - [Commits](modelcontextprotocol/typescript-sdk@1.32.1...1.32.2) Updates `@clerk/nextjs` from 7.9.14 to 7.9.20 - [Release notes](https://github.com/clerk/javascript/releases) - [Changelog](https://github.com/clerk/javascript/blob/main/packages/nextjs/CHANGELOG.md) - [Commits](https://github.com/clerk/javascript/commits/@clerk/nextjs@7.9.20/packages/nextjs) Updates `@vitest/coverage-v8` from 4.1.11 to 4.1.12 - [Release notes](https://github.com/vitest-dev/vitest/releases) - [Changelog](https://github.com/vitest-dev/vitest/blob/main/docs/releases.md) - [Commits](https://github.com/vitest-dev/vitest/commits/v4.1.12/packages/coverage-v8) Updates `vitest` from 4.1.11 to 4.1.12 - [Release notes](https://github.com/vitest-dev/vitest/releases) - [Changelog](https://github.com/vitest-dev/vitest/blob/main/docs/releases.md) - [Commits](https://github.com/vitest-dev/vitest/commits/v4.1.12/packages/vitest) Updates `vite` from 8.3.4 to 8.3.5 - [Release notes](https://github.com/vitejs/vite/releases) - [Changelog](https://github.com/vitejs/vite/blob/main/packages/vite/CHANGELOG.md) - [Commits](https://github.com/vitejs/vite/commits/v8.3.5/packages/vite) --- updated-dependencies: - dependency-name: "@anthropic-ai/claude-agent-sdk" dependency-version: 0.3.310 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: production-minor-patch - dependency-name: "@anthropic-ai/sdk" dependency-version: 0.137.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: production-minor-patch - dependency-name: "@clerk/nextjs" dependency-version: 7.9.20 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: production-minor-patch - dependency-name: "@modelcontextprotocol/sdk" dependency-version: 1.32.2 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: production-minor-patch - dependency-name: "@rollup/rollup-linux-x64-gnu" dependency-version: 4.65.0 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: production-minor-patch - dependency-name: "@vitest/coverage-v8" dependency-version: 4.1.12 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: production-minor-patch - dependency-name: arktype dependency-version: 2.2.9 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: production-minor-patch - dependency-name: jose dependency-version: 6.2.13 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: production-minor-patch - dependency-name: posthog-node dependency-version: 5.57.2 dependency-type: direct:production update-type: version-update:semver-minor dependency-group: production-minor-patch - dependency-name: vite dependency-version: 8.3.5 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: production-minor-patch - dependency-name: vitest dependency-version: 4.1.12 dependency-type: direct:production update-type: version-update:semver-patch dependency-group: production-minor-patch ... Signed-off-by: dependabot[bot] <support@github.com>
…ql/cargo-minor-patch-608f5ac5b5 chore(deps): bump uuid from 1.28.0 to 1.29.0 in /packages/eql in the cargo-minor-patch group
…typescript/packages/protect-ffi/cargo-minor-patch-c3f7ddf466 chore(deps): bump the cargo-minor-patch group in /languages/typescript/packages/protect-ffi with 2 updates
…r-patch-7d0f2ed97b chore(deps): bump the cargo-minor-patch group across 4 directories with 3 updates
…duction-minor-patch-e9aacc65a9 chore(deps): bump the production-minor-patch group across 1 directory with 11 updates
…ith 2 updates Bumps the dev-dependencies group with 2 updates in the / directory: [@biomejs/biome](https://github.com/biomejs/biome/tree/HEAD/packages/@biomejs/biome) and [@clerk/backend](https://github.com/clerk/javascript/tree/HEAD/packages/backend). Updates `@biomejs/biome` from 2.5.15 to 2.6.3 - [Release notes](https://github.com/biomejs/biome/releases) - [Changelog](https://github.com/biomejs/biome/blob/main/CHANGELOG_v1.md) - [Commits](https://github.com/biomejs/biome/commits/@biomejs/biome@2.6.3/packages/@biomejs/biome) Updates `@clerk/backend` from 3.26.0 to 3.26.2 - [Release notes](https://github.com/clerk/javascript/releases) - [Changelog](https://github.com/clerk/javascript/blob/main/packages/backend/CHANGELOG.md) - [Commits](https://github.com/clerk/javascript/commits/@clerk/backend@3.26.2/packages/backend) --- updated-dependencies: - dependency-name: "@biomejs/biome" dependency-version: 2.6.3 dependency-type: direct:development update-type: version-update:semver-minor dependency-group: dev-dependencies - dependency-name: "@clerk/backend" dependency-version: 3.26.2 dependency-type: direct:development update-type: version-update:semver-patch dependency-group: dev-dependencies ... Signed-off-by: dependabot[bot] <support@github.com>
…ve trigger Dependabot moved the lizard pin to 1.24.3 but kept the 1.24.1 hashes, so `pip install --require-hashes` refused the wheel. The new hashes are the ones PyPI lists for lizard-1.24.3 (wheel and sdist), from the command in the file's header, which now names 1.24.3 too. No other package moves. lizard 1.24.3 fixes the TypeScript misread the span tests used as their trigger (a ternary returning an object literal, lizard issue #523), so four tests failed: the fixture spans were read correctly. A `{` in a regex literal still makes 1.24.3 run a span on to the next `function`, over any top-level code between them. The four fixtures now use that trigger, and the driver comment names it.
fix(deps): patch rustls to 0.23.46
fix(deps): patch fast-uri to 3.1.9
Biome 2.6 refuses a config whose $schema names an older version, and it now formats Markdown and checks YAML. Both are new to this repo's Biome surface: Markdown includes skills/ (shipped in the stash tarball) and .changeset/, and the YAML includes deliberately malformed workflow fixtures that only parse-fail. Turn Markdown formatting off and exclude YAML so the upgrade changes no checked surface; adopting either is a separate decision. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7r5oCsYG9vQSKWa3xw
…n ADR The review of #1193 kept reopening the same question: whether a forced backfill pass may resume without force. The rule was being worked out per run and per command. This records the vocabulary agreed for the follow-up design in #1217 (column migration, direction, twin, backfill pass, run, obligation, waiver, abort) and the decision that a forced pass is an obligation on the column migration, enforced by @cipherstash/migrate for every caller, CLI or embedded worker. Refs #1217
The ADR described enforcement that is not implemented yet. #1193 enforces it for the reverse direction only; the forward direction, the waiver and the obligation events arrive with #1217. Say so, and correct the event claim to what is recorded today. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MUUZYHewafFdgbmD7TVAZW
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01MUUZYHewafFdgbmD7TVAZW
… main #1217 landed after the review, so ColumnMigration now enforces the obligation, the waiver and the one-open-per-pair rule in migrate, and every event records details.force. Describe that, mark the ADR accepted, and apply the review's wording fixes: partly overwritten, abort, the twin's reference column, current phase names, and a Migrate relationship in the context map.
Match the repo's other CONTEXT.md files and ADRs, as review asked.
The README called a column migration a "lifecycle", which the package glossary lists as a term to avoid.
#1217 is the open issue; #1276 is the merged implementation (part 1 of 2), so the ADR now credits the PR. Only the four backfill events record `details.force`; the dual-write confirmation and drop events do not, and the ADR says `details` is effectively public, so the sentence has to be exact. Also say that no abort operation exists yet, align the glossary with the ADR on "dropped or aborted" and on the obligation being derived from events, and name `details.acknowledgedWith` as the legacy stored key.
The Decision still said the obligation is "recorded once, on the column migration", which contradicts the Consequences: the obligation has no events of its own and is derived from the trail. Say derived. #1193 enforced an open obligation in `runDecryptBackfill`; `ColumnMigration` did not exist until #1276, so credit each PR with what it did. Events before #1276 put `force: true` on every reverse event of a forced pass, not on `backfill_started` only (that was forward only), and `details.acknowledgedWith` recorded the up-front choice, whose `force` value opens the obligation rather than waiving it. Reflow both paragraphs.
docs(migrate): add the migrate glossary and the forced-pass obligation ADR
docs(e2e): specify the two-store v2-to-v3 guide and migration harness
refactor(stack,supabase,prisma): simplify the log-safe failure helper
…alidate UTF-8 Review follow-ups on #1553. - The device-code deadline is now fixed when `begin` receives the code and stored on `PendingDeviceCode`. Before, it was computed when polling started, so a caller that waited before polling got a fresh window. The deadline is enforced across sleeps (`sleep_until` the earlier of the next poll and the deadline) and across in-flight requests (`timeout_at`), so polling stops at the deadline and a success that arrives after it is not saved. - `load` validates the whole file as UTF-8 before parsing. `from_slice` skips the bytes of an ignored field, so it accepted invalid UTF-8 in an unknown field. An invalid file is again `ProfileError::Io` with `InvalidData`, as it was under `read_to_string`, which reverts the `Json` mapping the previous commit introduced. - `SecretBuf::reserve` uses `try_reserve_exact` rather than `Vec::with_capacity`, so an oversized input returns an error instead of aborting the process. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7qCTmGiT6hfLv8rxDP
…ing it Review follow-up on #1553. On Unix, `std::env::var` returns `VarError::NotUnicode` holding the whole value when it is not UTF-8, and `.ok()` dropped that copy of the key without wiping it. `access_key_from_env` now wipes it first. The value is still treated as unset, as before, since it cannot be a valid access key. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7qCTmGiT6hfLv8rxDP
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7qSiWtt22aUYDwvgVL
The mutants gate found that turning the poll sleep's deadline from now + interval into now - interval left every test passing: the loop then polls without waiting, and under paused time the run spins until the gate's timeout. The new test lets the pending answer through for exactly the polls that fit before the deadline, then denies access, so a loop that skips the wait fails at once rather than hanging. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7qCTmGiT6hfLv8rxDP
The mutants gate still timed out on the poll-wait mutant. With the wait gone, every test whose server keeps answering authorization_pending or slow_down spins: paused time only advances when the runtime is idle, so the deadline never arrives and the run hangs past the gate's timeout before the new test's failure can count. Each of those token mocks now allows a set number of polls and answers access_denied after that, so under the mutant all six tests fail in under a second instead of hanging. The polls each test allows are at least what the real loop makes, so the tests pass unchanged on the current code. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7qCTmGiT6hfLv8rxDP
…oc calls in the test Memory.Size reports zero for a full 4 GiB memory, which would have refused every placement; each slot is now probed with ReadUint32Le instead. The out-of-range test counts se_alloc calls on the host side and requires zero, so a refusal that allocated first no longer passes. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7qSiWtt22aUYDwvgVL
fix(golang): wipe guest buffers on failed placement and close stashgen's engine
fix(eql): block a domain compared with another domain's query operand
…h sees it With a workspace pinned, a near-expiry session for a different workspace reached AutoRefresh and was refreshed on first use, consuming a credential that belongs to that workspace. The strategy now checks the starting session at build time and fails with WorkspaceMismatch before any request. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7o5FdWpXMFP7MSoBnt
…y rotated When a refresh rotates the credential but returns a token for another workspace, the saved credential is spent. Drop the saved session and make later refreshes on the same refresher ask for a new sign-in, so a retry or a reopened profile cannot replay it and get the chain revoked. A response that did not rotate the credential leaves the saved session untouched. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7o5FdWpXMFP7MSoBnt
refactor(stack-supabase): simplify the plaintext-table path from #1544
If the issuer rotated the credential and the saved copy cannot be removed, return that error instead of the mismatch: the credential is still on disk and the next process could replay it. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7o5FdWpXMFP7MSoBnt
fix(eql): align JSON array flag with SQL and handle empty arrays
read_from resized the buffer to its whole capacity before every read, zero-filling all of the spare tail, then truncated back to the bytes read. A reader that returns short reads into a large buffer had the same tail zeroed again on every call, so the cost grew with the square of the file size. Each read now gets a window of at most one chunk. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_012Jmn7qCTmGiT6hfLv8rxDP
fix(stack-auth): honour CS_WORKSPACE_CRN with CLI sign-in
fix(stack-auth,stack-profile): clamp server expires_in and wipe secret buffers
Resolves the conflicts that block #1229. - eql-codegen generate.rs: keeps both sides of render_query_functions_file: next's optional EQL v4 query terms and main's cross-pair blockers. next's tests pass the new siblings argument; main's tests render the placeholder output as eql_v3 before reading it. - src/v3/scalars: regenerated from the merged generator. The auto-merge had carried main's literal eql_v3 names into next's placeholder files. - src/v3/json: main's fixes (cardinality for an empty array, the boolean array flag, no ignored query_json cast, schema-qualified ord operators), written with next's placeholders. These files are confined to EQL v3. - Go encrypt/transport.go: next's stepping transport, which has no transport_send; token_get already uses main's guest.Place. - Go cmd/stashgen/main_test.go: both sides' tests. main's closingEngine returns next's Check signature.
between is unpinned LANGUAGE sql so that it inlines, which means its body is parsed under the caller's search_path. Its bare >= and <= then resolve against whatever that path finds. For the ORE range the term is a composite type whose operators live in public, so with public off the path the comparison falls back to pg_catalog's record comparison and orders the ciphertext bytes, with no error. This is the bug main fixed for the other wrappers; between arrived on next and was not covered. The body now names OPERATOR(<schema>.>=) and OPERATOR(<schema>.<=), with the schema from Term::operator_schema. The catalog-wide wrapper test now checks every generated file in each EQL version it belongs to, which is what found this.
🦋 Changeset detectedLatest commit: 86a820d The changes in this PR will be included in the next version bump. This PR includes changesets to release 22 packages
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the
|
Mutation testing (cargo-mutants,
|
| caught | missed | unviable | timeout |
|---|---|---|---|
| 19 | 0 | 22 | 0 |
Every mutant in the changed lines was caught by a test.
next's gate skips the live JS jobs for a pull request that changes only docs/**. main added docs/migration/protect-dynamodb-to-stack.md and e2e/tests/protect-upgrade-guide.e2e.test.ts, which runs in e2e-tests and holds the guide's claims to the suites it describes. After the merge a pull request editing only the guide would skip the test that checks it, and tests-workflow-gate.test.mjs fails on exactly that. docs/migration/** joins the gate's direct list, so a change there runs the jobs again.
cipherstash-agent
left a comment
There was a problem hiding this comment.
Review attestation
Reviewed 86a820d3 · posted from Claude Code · claude-opus-5-5 · effort medium
Sources: code-review · claude-opus-5-5
Skills: code-review, rust-security, review-cross-check
Findings (3): 3 minor — 2 inline, 1 below.
packages/eql/src/v3/json_tree/operators.sql:171(outside diff) minor The eight v4 JSON treelt/lte/gt/gtebodies (lines 171, 191, 211, 231, 255, 276, 297, 318) still use bare comparison operators.pg_catalogdoes not protect them: their operands are theope_cllwdomain, so a<defined on that domain in any schema on the caller'ssearch_pathwins. On PG 16, that returned wrong rows with no error, while theOPERATOR(pg_catalog.<)form returned the right ones. Already onnextand listed in the PR's Review notes; recorded so the follow-up has a concrete trigger.
rust-security was applied to the Rust in the conflict resolution (eql-codegen) and found nothing; that code handles no secrets. Three more candidate findings were checked at head and dropped: the eql_v4.between operators resolve to the ORE comparators and use the index with search_path='', measured on PG 16; guest.Place is unchanged from main; the lock-capacity floor still holds at 8,557 locks per v3 uninstall.
Generated by Claude Code
| /// comparison — byte-wise ciphertext order, no error. A bare | ||
| /// operator token anywhere in a wrapper body fails here. | ||
| #[test] | ||
| fn every_wrapper_body_schema_qualifies_its_operators() { |
There was a problem hiding this comment.
minor This test only walks render_type output, and only single-line AS $$ SELECT eql_vN. bodies, so hand-written inlinable SQL (src/v3/json_tree/operators.sql, src/v3/json/operators.sql) is never scanned. The bare-operator bodies in json_tree pass CI, and so would a new hand-written v4 wrapper. Extend the scan to the hand-written operators.sql files, in both body forms, or add a v4 search_path SQLx test with a decoy operator.
Source: code-review · claude-opus-5-5
Generated by Claude Code
| // file is checked in every EQL version it belongs to. Guards the filter | ||
| // itself: a template change that stopped emitting single-line | ||
| // `AS $$ SELECT eql_vN.` bodies would otherwise pass vacuously. | ||
| assert!(bodies > 500, "only {bodies} wrapper bodies inspected"); |
There was a problem hiding this comment.
minor bodies adds up both versions, and the v3 bodies alone (1087, measured) clear 500. So a change that stopped any v4 body matching, including the eql_v4.between bodies ab4d5ff6 fixed, would still pass. Count each EqlVersion separately and assert each one clears the threshold.
Source: code-review · claude-opus-5-5
Generated by Claude Code
Summary
nexthas fallen 168 commits behindmain, and #1229 (nextintomain) shows merge conflicts. This PR mergesmainintonextwith the conflicts resolved, plus one fix the merge exposed. Merge it with a merge commit (not squash or rebase), somain's commits stay innext's history and #1229 stops conflicting.Background: EQL is the SQL that stores and queries encrypted values in PostgreSQL. On
next, the EQL SQL source is a template, rendered once aseql_v3and once aseql_v4; names are written as{{prefix}}rather thaneql_v3. Most of the conflicts come frommainediting those files with literaleql_v3names.Changes
How each conflict was resolved:
packages/eql/crates/eql-codegen/src/generate.rs. This is the generator that writes the per-type SQL. Both branches changed how it writes a query operand's functions:nextadded optional query terms for EQL v4, andmainadded blockers that raise an error when a column is compared with another type's query operand (without them PostgreSQL silently compares the raw JSON). The two do not interact, so both are kept.next's tests pass the new argument;main's tests render the template output aseql_v3before reading it, asnext's own tests already do.packages/eql/src/v3/scalars/**(about 90 files). These are generated. Rather than hand-merging them, they were regenerated from the merged generator. That also fixes 39 files Git had merged without a conflict but withmain's literaleql_v3names innext's placeholder files.packages/eql/src/v3/json/{functions,operators,types}.sql. Hand-written and confined to EQL v3. Each conflict differed only bymain's fix, somain's version is taken, written withnext's placeholders:jsonb_array_lengthreturns 0 for an empty array, the array flag must be a JSON boolean, the ignoredquery_jsoncast is removed, and the ordering comparisons namepg_catalogoperators.languages/golang/encrypt/transport.go.nextreplaced the blockingtransport_sendhost function with steps, somain's change to it has nothing to apply to.next's version is kept; itstoken_getalready usesmain'sguest.Place, which wipes a buffer it could not place.languages/golang/cmd/stashgen/main_test.go. Both sides added tests in the same place; both are kept.main's fake engine now returnsnext'sChecksignature.The fix, in its own commit:
fix(eql): schema-qualify the operators in eql_v4.between.mainfixed comparison functions whose bodies used a bare<: they are inlinable SQL functions, so the operator is looked up through the caller'ssearch_path. For ORE (order-revealing encryption) values, which are a composite type with operators inpublic, a caller withoutpublicon its path got PostgreSQL's built-in record comparison, which orders the ciphertext bytes and raises no error.eql_v4.between, which exists only onnext, had the same bare>=and<=. It now names the operator's schema.main's test that checks every generated function now checks each file in both EQL versions, which is how this was found.betweenis unreleased (it is described in the pendingeql-v4-bindingschangeset), so there is no new changeset.And one CI fix, found by the first CI run:
ci(tests): run the JS jobs when the Protect upgrade guide changes.nextadded a gate that skips the live JS test jobs for a pull request that changes onlydocs/**.mainadded the Protect upgrade guide (docs/migration/protect-dynamodb-to-stack.md) and an e2e test that checks the guide against the suites it describes. Together, a pull request editing only the guide would skip the test that checks it, andtests-workflow-gate.test.mjsfailed on exactly that.docs/migration/**is now on the gate'sdirectlist, so a change there runs the jobs.Verification
cargo test --locked -p eql-domains -p eql-codegen: passed (190 tests ineql-codegen). This includes the drift gate that compares committed generated files with the generator.eql-codegen bindingsandeql-codegen go-eql: no changes to the committed Rust bindings orlanguages/golang/encrypt/eql/eql_gen.go.tasks/build.shinpackages/eql: passed, including the installer completeness, symbol-order and v4 isolation gates, for both bundles.cargo clippy --locked --workspace --all-targets --all-features -- -D warnings(root workspace): passed.cargo nextest run -p stack-auth -p stack-profile -p stack-encrypt --all-features: 1801 passed, 2 skipped.cargo metadata --lockedfor all eight Cargo workspaces, andpnpm install --lockfile-only: every lockfile is consistent, andpnpm-lock.yamldid not change.go build ./...andgo vetpassed;TestRunClosesTheEngineand theinternal/...tests passed.pnpm run test:scripts: after the CI fix, 1504 passed. One test ineql-suite-ci.test.mjstimed out locally (about 6s against a 5s limit); it passed in CI on this branch.main's new search-path and cross-pair tests, and the Go tests that need the guests built. CI runs both.Related
Refs #1229, #1455
Review notes
generate.rsand thebetweencommit; the generated SQL follows from them.next's v4 JSON tree comparisons (packages/eql/src/v3/json_tree/operators.sql) still use bare<,<=,>,>=, the patternmainfixed in the v3 JSON SQL. Those compare OPE terms with PostgreSQL's built-inbyteaoperators, so the risk is lower than for ORE, but the two versions now differ. It can be fixed the same way in a follow-up.