nntrivi2001
c7dbaf2d9f
Two leftovers from the rebrand, done together because they touch the same docs and manifests. **Data directory.** `agent-dashboard` could not simply be renamed in the source: it names a real directory holding the live SQLite database. Startup now moves it once, via `migrateLegacyDataDir()` in `server/lib/claude-home.js` — a same-filesystem rename, so it is instant even at multiple gigabytes and cannot leave two divergent copies. `getDataDir()` stays PURE and resolves to the legacy path until the move happens; the first version of this put the rename inside the getter, which meant `npm run test:server` relocated the developer's real database as a side effect. The call sits behind `require.main === module` and ahead of the router requires, because those pull in db.js, which opens SQLite at module load. A failed rename is silent and the old directory keeps being used: an unfashionable directory name beats a blank database beside an unreachable one. Same treatment for the port-discovery file (`.agent-dashboard.json` → `.agenthub.json`, old one dropped on write). Not renamed on purpose: the Sagi localStorage keys (renaming silently resets everyone's prefs) and the PWA manifest `id` (changing it orphans an installed app). **MCP server.** 4 tool calls in 46 days across 577,132 events, against 4,556 lines of TypeScript, a 2 MB committed build artifact, and a pre-commit guard that had already blocked commits twice. Deleted: `mcp/`, the maintainer skills and `mcp-reviewer` agent, the build check, `plugins/agenthub/.mcp.json`, `docs/MCP.md`, 12 npm scripts, 5 Makefile targets, and every doc section describing it. Kept, because they are unrelated things sharing the name: the Config Explorer's MCP tab and its API, per-lane Playwright MCP, the agenthub-config plugin's MCP audit, and the historical `mcp__plugin_agenthub_*` events that analytics must keep attributing. Verified: server 1432/1432, client 459/459, tsc clean, prettier clean. Both directions of the migration proven — a sandbox-HOME test run leaves the legacy directory untouched, and a real boot moves it with its contents intact.
66 lines
2.6 KiB
Bash
Executable File
66 lines
2.6 KiB
Bash
Executable File
#!/usr/bin/env sh
|
|
#
|
|
# Pre-commit: auto-format staged files, then run the full test suite.
|
|
# The commit is ABORTED unless every test passes.
|
|
#
|
|
# Activated via `core.hooksPath=.husky` (set by the package.json "prepare"
|
|
# script on `npm install`). No husky/lint-staged runtime required.
|
|
|
|
# Abort the commit on the first failing command (formatting error or failing test).
|
|
set -e
|
|
|
|
ROOT="$(git rev-parse --show-toplevel)"
|
|
cd "$ROOT"
|
|
|
|
# ── 1. Format staged files with Prettier, then re-stage them ────────────────
|
|
# Only Added/Copied/Modified/Renamed paths in the index. --ignore-unknown lets
|
|
# Prettier silently skip anything it can't parse, so we can hand it every staged
|
|
# path without filtering by extension. NUL-delimited (-z / -0) so paths with
|
|
# spaces are handled correctly.
|
|
#
|
|
# NOTE: this re-adds the *whole* file. If you intentionally staged only part of
|
|
# a file, review the result after the auto-format.
|
|
STAGED="$(git diff --cached --name-only --diff-filter=ACMR)"
|
|
if [ -n "$STAGED" ]; then
|
|
PRETTIER="$ROOT/node_modules/.bin/prettier"
|
|
if [ -x "$PRETTIER" ]; then
|
|
set -- "$PRETTIER"
|
|
else
|
|
set -- npx --no-install prettier
|
|
fi
|
|
|
|
echo "🎨 Formatting staged files with Prettier..."
|
|
git diff --cached --name-only --diff-filter=ACMR -z | xargs -0 "$@" --write --ignore-unknown
|
|
git diff --cached --name-only --diff-filter=ACMR -z | xargs -0 git add
|
|
else
|
|
echo "🎨 No staged files to format."
|
|
fi
|
|
|
|
# ── 2. Run tests — commit is blocked unless all pass ────────────────────────
|
|
# Each suite is retried once on failure. The full run executes dozens of test
|
|
# files concurrently, each starting its own server (plus the CLI suite's
|
|
# spawned child processes with hard kill timeouts), so a loaded machine can
|
|
# produce a one-off timing failure unrelated to the commit. A real regression
|
|
# is deterministic: it fails both runs and still blocks the commit. This keeps
|
|
# the gate strict without making commits a dice roll on machine load.
|
|
run_suite() {
|
|
# $1 = human label, $2 = npm script
|
|
echo "🧪 Running $1 tests..."
|
|
if npm run "$2"; then
|
|
return 0
|
|
fi
|
|
echo "⚠️ $1 suite failed — retrying once to rule out machine-load flakiness..."
|
|
if npm run "$2"; then
|
|
echo "✅ $1 suite passed on retry — treating the first failure as a load flake."
|
|
echo " If this keeps happening, the flaky test above deserves a real fix."
|
|
return 0
|
|
fi
|
|
echo "❌ $1 suite failed twice — genuine failure, aborting the commit."
|
|
return 1
|
|
}
|
|
|
|
run_suite backend test:server
|
|
run_suite frontend test:client
|
|
|
|
echo "✅ Formatting applied and all tests passed — proceeding with commit."
|