Add leaderboard.querygym.com Astro app + workspace

.github/workflows/leaderboard-build.yml

@@ -0,0 +1,63 @@
+name: Leaderboard Build
+
+on:
+  pull_request:
+    paths:
+      - 'reproducibility/site/**'
+      - 'reproducibility/data/**'
+      - 'web/shared/**'
+      - 'package.json'
+      - 'pnpm-workspace.yaml'
+      - 'dataset_registry.yaml'
+      - '.github/workflows/leaderboard-build.yml'
+  push:
+    branches: [main]
+    paths:
+      - 'reproducibility/site/**'
+      - 'reproducibility/data/**'
+      - 'web/shared/**'
+      - 'package.json'
+      - 'pnpm-workspace.yaml'
+
+concurrency:
+  group: leaderboard-${{ github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Node
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+
+      - name: Set up pnpm
+        # No `version:` — pnpm/action-setup reads `packageManager` from
+        # package.json. Specifying both causes ERR_PNPM_BAD_PM_VERSION.
+        uses: pnpm/action-setup@v4
+
+      - name: pnpm cache
+        uses: actions/cache@v4
+        with:
+          path: ~/.pnpm-store
+          key: pnpm-${{ runner.os }}-${{ hashFiles('**/pnpm-lock.yaml') }}
+          restore-keys: pnpm-${{ runner.os }}-
+
+      - name: Install
+        run: pnpm install --frozen-lockfile
+
+      - name: Build leaderboard data shards
+        run: pnpm -F @qg/leaderboard build:data
+
+      - name: Build leaderboard site
+        run: pnpm -F @qg/leaderboard build
+
+      - name: Upload dist artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: leaderboard-dist
+          path: reproducibility/site/dist
+          retention-days: 7

.github/workflows/reproducibility-check.yml

@@ -6,12 +6,17 @@ on:
       - 'reproducibility/**'
       - 'examples/querygym_pyserini/pipeline.py'
       - 'dataset_registry.yaml'
+      - 'pyproject.toml'
+      - '.github/workflows/reproducibility-check.yml'
   push:
     branches: [main]
     paths:
       - 'reproducibility/**'
       - 'examples/querygym_pyserini/pipeline.py'
       - 'dataset_registry.yaml'
+      - 'pyproject.toml'
+      - '.github/workflows/reproducibility-check.yml'
+  workflow_dispatch:
 
 concurrency:
   group: repro-${{ github.ref }}
@@ -36,7 +41,10 @@ jobs:
           pip install pytest-cov
 
       - name: Run repro tests
-        run: pytest reproducibility/tests -v --no-cov
+        # Use `python -m pytest` so the cwd (repo root) is on sys.path —
+        # `from reproducibility.lib import ...` needs that, and the bare
+        # `pytest` form doesn't add it on CI's Python 3.9.
+        run: python -m pytest reproducibility/tests -v --no-cov
 
       - name: Aggregator --check
         run: python -m reproducibility.scripts.aggregate_runs --check

.gitignore

@@ -1,3 +1,17 @@
+# Local Claude Code notes (not committed)
+CLAUDE.local.md
+.claude/
+
+# JavaScript / Astro / pnpm — never check in
+node_modules/
+.astro/
+**/dist/
+pnpm-lock.yaml.bak
+package-lock.json
+yarn.lock
+# Generated leaderboard data shards (built from reproducibility/data/)
+reproducibility/site/src/data/
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]
@@ -8,17 +22,20 @@ __pycache__/
 
 # Distribution / packaging
 .Python
-build/
-develop-eggs/
-dist/
-downloads/
-eggs/
-.eggs/
-lib/
-lib64/
-parts/
-sdist/
-var/
+# Anchored to repo root so these Python build dirs don't collide with project
+# directories that legitimately use the same names (e.g. reproducibility/lib/,
+# web/site/dist/).
+/build/
+/develop-eggs/
+/dist/
+/downloads/
+/eggs/
+/.eggs/
+/lib/
+/lib64/
+/parts/
+/sdist/
+/var/
 wheels/
 pip-wheel-metadata/
 share/python-wheels/
@@ -136,11 +153,11 @@ dmypy.json
 reformulated_queries.tsv
 *.run
 *.eval.txt
-out/
-results/
-indices/
-datasets/
-scratch/
+/out/
+/results/
+/indices/
+/datasets/
+/scratch/
 query_expansion_results/
 dev_scripts/
 quick_independence_test.py

MANIFEST.in

@@ -19,6 +19,12 @@ recursive-include docs *.md
 prune reproducibility
 prune web
 
+# Belt-and-suspenders: explicit exclude for workspace files that would never
+# match a default include rule but shouldn't sneak in via a future MANIFEST sweep.
+exclude package.json pnpm-workspace.yaml pnpm-lock.yaml package-lock.json yarn.lock
+global-exclude node_modules
+global-exclude .astro
+
 # Exclude development and build artifacts
 global-exclude __pycache__
 global-exclude *.py[co]

package.json

@@ -0,0 +1,15 @@
+{
+  "name": "querygym-web",
+  "version": "0.0.0",
+  "private": true,
+  "description": "Workspace root for QueryGym websites (leaderboard + marketing). Not published.",
+  "scripts": {
+    "build:leaderboard": "pnpm -F @qg/leaderboard build:data && pnpm -F @qg/leaderboard build",
+    "dev:leaderboard": "pnpm -F @qg/leaderboard dev"
+  },
+  "engines": {
+    "node": ">=20",
+    "pnpm": ">=9"
+  },
+  "packageManager": "pnpm@10.14.0"
+}