build(deps): bump actions/setup-node from 6 to 7 - #10
Open
dependabot[bot] wants to merge 3 commits into
Open
Conversation
dependabot
Bot
added
dependencies
Text search previously shelled out to the `ck` CLI for every query over the filesystem export tree. On a large corpus that collapsed: an unscoped `lex` query hung for minutes with runaway RAM, and `ck --regex` over two channel roots did not finish in 120s. Cost scaled with corpus size rather than result size, because ck fans out sequentially per scope and reconciles a manifest that reached 282 MB on a 888k-file export. Search now routes through a planner (`src/lib/query-plan.ts`) across three tiers: - `fts` — a local SQLite FTS5 sidecar (`src/lib/fts.ts`) built directly from opencode.db. A `trigram` table serves literal/substring queries and a `unicode61 tokenchars '-_'` table serves `lex` BM25 ranking. - `rg` — ripgrep (`src/lib/rg.ts`), the exhaustive tier: real regular expressions, any channel including the raw replay tree, and the escalation target when the index is not authoritative. - `ck` — now reserved for `sem`/`hybrid` embedding search only, keeping the existing lazy index build/refresh. `ck` is never a fallback for regex/lex: it reads the same file tree as ripgrep, so escalating to it adds latency and no recall. Measured on a 568k-document index: literal 169ms, phrase 184ms, real regex 23ms, two-character literal 271ms, forensics over the 328k-file raw tree 285ms, peak RSS 212 MB. The raw tree previously timed out at 34s returning nothing. Also: - `grep-session` migrated from ck to ripgrep; it no longer depends on ck. - New `opencode-sessions-explorer-fts-build` command (`--reset`, `--budget-ms`), with resumable builds, dead-letter tracking for parts whose documents fail to build, and SQLite tombstone reconciliation. - Index completeness is durable and drain-based, and gates the fast tier; a partial or corrupt index falls back to ripgrep instead of silently serving a fraction of the corpus. Freshness lag is reported but never gating. - `mode:"lex"` is always text and never reinterpreted as a regular expression. - A literal without a usable 3-character trigram run is routed to ripgrep rather than degrading to a full table scan. - `timeout_ms` now bounds the whole chain, including syncs and ck index probes. - A missing or non-executable ripgrep raises `RG_NOT_FOUND`/`RG_FAILED` instead of returning an empty successful result. - Unscoped `sem`/`hybrid` that would resolve onto the raw tree is refused with an actionable `BAD_ARGS`, including the indirect route via a partial curated export. - New response fields: `backend`, `backends_tried`, `plan_reason`, `literal`, `search_duration_ms`, `partial`, `role_filter_truncated`, `scope_truncated`. - `check-deps` probes ripgrep and the sidecar; `db-stats` reports sidecar health including completeness, lag, and failure counts. `tool-output` is indexed as a 4 KB head + tail excerpt with a pointer to `get-part`, so the fast tier is not full-fidelity for that channel; full content stays searchable via ripgrep. Trigram case folding is ASCII-only. Both limitations are documented rather than papered over.
ripgrep became a runtime dependency when search moved to the three-tier query planner: it serves `regex` mode and is the exhaustive fallback tier for both `search-text` and `grep-session`. GitHub-hosted ubuntu and macOS runners do not ship it, so every rg-backed test failed with RG_NOT_FOUND. Install it explicitly per runner OS instead of relying on undocumented runner image contents, and skip the install when a usable rg is already present.
Bumps [actions/setup-node](https://github.com/actions/setup-node) from 6 to 7. - [Release notes](https://github.com/actions/setup-node/releases) - [Commits](actions/setup-node@v6...v7) --- updated-dependencies: - dependency-name: actions/setup-node dependency-version: '7' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>
dependabot
Bot
force-pushed
the
dependabot/github_actions/actions/setup-node-7
branch
from
August 6, 2026 18:06
debe38c to
5fde878
Compare
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.
Bumps actions/setup-node from 6 to 7.
Release notes
Sourced from actions/setup-node's releases.
... (truncated)
Commits
8207627Migrate to ESM and upgrade dependencies (#1574)04be95cAdd cache-primary-key and cache-matched-key as outputs (#1577)7c2c68ddocs: Update caching recommendations to mitigate cache poisoning risks (#1567)6a61c03Merge pull request #1569 from jasongin/update-actions-cache-5.1.030eb73bResolve high-severity audit issues4e1a87aUpdate dist360237fStrict equality4f8aac5Bump@actions/cacheto 5.1.0, log cache write deniedf4a67bbOnly usemirrorTokeningetManifestif it's provided (#1548)0355742Remove dummy NODE_AUTH_TOKEN export (#1558)