feat: add run-aware corpus schema and named queries

Co-Authored-By: Oz <oz-agent@warp.dev>

Author: Mjboothaus

src/pytest_just/toolkit/query_pack.py

@@ -0,0 +1,83 @@
+"""Named query packs for justfile corpus analytics."""
+
+from __future__ import annotations
+
+from pathlib import Path
+from typing import Any
+
+import duckdb
+
+NAMED_QUERIES: dict[str, str] = {
+    "top_recipe_names": """
+        SELECT
+            recipe_name,
+            COUNT(DISTINCT repo_name) AS repo_count,
+            COUNT(*) AS occurrence_count
+        FROM recipe_occurrences
+        WHERE (? IS NULL OR run_id = ?)
+        GROUP BY recipe_name
+        ORDER BY repo_count DESC, occurrence_count DESC, recipe_name ASC
+        LIMIT ?
+    """,
+    "top_reused_signatures": """
+        SELECT
+            recipe_name,
+            repo_count,
+            occurrence_count,
+            signature
+        FROM unique_recipes
+        WHERE (? IS NULL OR run_id = ?) AND repo_count > 1
+        ORDER BY repo_count DESC, occurrence_count DESC, recipe_name ASC
+        LIMIT ?
+    """,
+    "parse_failures": """
+        SELECT
+            repo_name,
+            parse_error
+        FROM repo_sources
+        WHERE (? IS NULL OR run_id = ?) AND parsed_success = FALSE
+        ORDER BY repo_name ASC
+        LIMIT ?
+    """,
+    "dependency_hotspots": """
+        SELECT
+            dependency_name,
+            COUNT(*) AS dependent_recipe_count,
+            COUNT(DISTINCT repo_name) AS repo_count
+        FROM recipe_dependencies
+        WHERE (? IS NULL OR run_id = ?)
+        GROUP BY dependency_name
+        ORDER BY dependent_recipe_count DESC, repo_count DESC, dependency_name ASC
+        LIMIT ?
+    """,
+}
+
+
+def list_named_queries() -> list[str]:
+    """Return available named query identifiers sorted by name."""
+    return sorted(NAMED_QUERIES)
+
+
+def run_named_query(
+    db_path: Path,
+    query_name: str,
+    *,
+    limit: int = 25,
+    run_id: str | None = None,
+) -> tuple[list[str], list[tuple[Any, ...]]]:
+    """Execute a named query against a corpus database."""
+    if query_name not in NAMED_QUERIES:
+        known = ", ".join(list_named_queries())
+        raise ValueError(f"Unknown query `{query_name}`. Known queries: {known}")
+    if limit < 1:
+        raise ValueError("limit must be >= 1")
+
+    con = duckdb.connect(str(db_path))
+    try:
+        relation = con.execute(NAMED_QUERIES[query_name], [run_id, run_id, limit])
+        rows = relation.fetchall()
+        columns = [item[0] for item in (relation.description or [])]
+    finally:
+        con.close()
+    return columns, rows
+