Add suggest_cache command and tests

scripts/analyze.py

@@ -143,5 +143,59 @@ def check_missing_domains(json_output: bool = typer.Option(False, "--json", help
             typer.echo(d)
 
 
+@app.command("suggest-cache")
+def suggest_cache(
+    threshold: int = typer.Option(
+        10, help="Minimum number of MISS entries to report"
+    ),
+    json_output: bool = typer.Option(False, "--json", help="Output results as JSON"),
+) -> None:
+    """Suggest domain/path pairs that could benefit from caching.
+
+    Paths with at least ``threshold`` ``MISS`` entries are shown for domains
+    whose server blocks lack a ``proxy_cache`` directive.
+    """
+
+    # Discover domains without explicit proxy_cache
+    paths = nginx_config.discover_configs()
+    servers = nginx_config.parse_servers(paths)
+    no_cache: Set[str] = set()
+    for server in servers:
+        if "proxy_cache" in server:
+            continue
+        for name in server.get("server_name", "").split():
+            if name:
+                no_cache.add(name)
+
+    conn = _connect()
+    cur = conn.cursor()
+    cur.execute(
+        """
+        SELECT host,
+               substr(request, instr(request, ' ')+1,
+                      instr(request, ' HTTP') - instr(request, ' ') - 1) AS path,
+               COUNT(*) AS miss_count
+        FROM logs
+        WHERE cache_status = 'MISS'
+        GROUP BY host, path
+        HAVING miss_count >= ?
+        ORDER BY miss_count DESC
+        """,
+        (threshold,),
+    )
+
+    rows = [r for r in cur.fetchall() if r[0] in no_cache]
+    conn.close()
+
+    if json_output:
+        result = [
+            {"host": host, "path": path, "misses": count} for host, path, count in rows
+        ]
+        typer.echo(json.dumps(result))
+    else:
+        for host, path, count in rows:
+            typer.echo(f"{host} {path} {count}")
+
+
 if __name__ == "__main__":
     app()