coderag-retrieval-console / coderag_api.py
akdybala's picture
Publish CodeRAG static Space
a39b25f verified
Raw History Blame Contribute Delete
11.1 kB
#!/usr/bin/env python
"""Local HTTP API for the CodeRAG React demo."""
import argparse
import json
import mimetypes
import os
from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
from pathlib import Path
from urllib.parse import urlparse
from coderag import build_file_index
from coderag.vector_db import CodeRagVectorDB
ROOT = Path(__file__).resolve().parent
MAX_REQUEST_BYTES = 2 * 1024 * 1024
STATIC_ROOT = ROOT / "coderag_ui" / "dist"
DEFAULT_SAMPLE_CODE = """from collections import defaultdict
def normalize_user(user):
email = user["email"].strip().lower()
return {
"id": user["id"],
"email": email,
"active": bool(user.get("active", True)),
}
def group_active_users(users):
groups = defaultdict(list)
for user in users:
normalized = normalize_user(user)
if normalized["active"]:
domain = normalized["email"].split("@")[-1]
groups[domain].append(normalized)
return dict(groups)
def find_user_by_email(users, email):
target = email.strip().lower()
for user in users:
normalized = normalize_user(user)
if normalized["email"] == target:
return normalized
return None
"""
class CodeRagApiHandler(BaseHTTPRequestHandler):
server_version = "CodeRAGApi/1.0"
def log_message(self, format_string, *args):
print(f"{self.address_string()} - {format_string % args}")
@property
def vector_db(self) -> CodeRagVectorDB:
return self.server.vector_db
def _send_json(self, status: int, payload: dict) -> None:
body = json.dumps(payload).encode("utf-8")
self.send_response(status)
self.send_header("Content-Type", "application/json; charset=utf-8")
self.send_header("Content-Length", str(len(body)))
self.send_header("Cache-Control", "no-store")
self.send_header("Access-Control-Allow-Origin", "*")
self.send_header("Access-Control-Allow-Headers", "Content-Type")
self.send_header("Access-Control-Allow-Methods", "GET, POST, OPTIONS")
self.end_headers()
self.wfile.write(body)
def do_OPTIONS(self):
self._send_json(204, {})
def do_GET(self):
path = urlparse(self.path).path
if path == "/api/health":
self._send_json(200, {"status": "ok"})
return
if path == "/api/coderag/stats":
try:
self._send_json(200, self.vector_db.stats())
except Exception as error:
self._send_json(
500,
{"error": f"Stats failed: {type(error).__name__}: {error}"},
)
return
self._serve_static(path)
def do_POST(self):
path = urlparse(self.path).path
if path not in {
"/api/coderag/search",
"/api/coderag/rebuild",
"/api/coderag/index-code",
}:
self._send_json(404, {"error": "Not found"})
return
try:
payload = self._read_payload()
except ValueError as error:
self._send_json(400, {"error": str(error)})
return
if path == "/api/coderag/rebuild":
try:
if self.server.repository_index.exists():
result = self.vector_db.build(
repository_index_path=self.server.repository_index,
synthetic_questions_path=self.server.synthetic_questions,
)
else:
sample_dir = ROOT / ".code_query_repo_cache" / "sample"
sample_dir.mkdir(parents=True, exist_ok=True)
sample_path = sample_dir / "sample_users.py"
sample_path.write_text(DEFAULT_SAMPLE_CODE, encoding="utf-8")
result = self.vector_db.build_from_file_index(
build_file_index(sample_path)
)
except Exception as error:
self._send_json(
500,
{
"error": f"Rebuild failed: "
f"{type(error).__name__}: {error}"
},
)
return
self._send_json(200, result)
return
if path == "/api/coderag/index-code":
code = payload.get("code")
filename = payload.get("filename", "pasted_code.py")
if not isinstance(code, str) or not code.strip():
self._send_json(400, {"error": "'code' must be a non-empty string"})
return
if not isinstance(filename, str) or not filename.strip():
filename = "pasted_code.py"
if not filename.endswith(".py"):
filename += ".py"
safe_name = "".join(
char if char.isalnum() or char in {"_", "-", "."} else "_"
for char in filename
)
pasted_dir = ROOT / ".code_query_repo_cache" / "pasted"
pasted_dir.mkdir(parents=True, exist_ok=True)
pasted_path = pasted_dir / safe_name
try:
pasted_path.write_text(code, encoding="utf-8")
index = build_file_index(
pasted_path,
propagation_depth=int(payload.get("propagation_depth", 2)),
max_callees=int(payload.get("max_callees", 4)),
callee_context_tokens=int(
payload.get("callee_context_tokens", 192)
),
)
result = self.vector_db.build_from_file_index(index)
except SyntaxError as error:
self._send_json(
400,
{
"error": (
f"Python parse failed at line {error.lineno}: "
f"{error.msg}"
)
},
)
return
except Exception as error:
self._send_json(
500,
{
"error": f"Indexing failed: "
f"{type(error).__name__}: {error}"
},
)
return
self._send_json(200, result)
return
query = payload.get("query")
if not isinstance(query, str) or not query.strip():
self._send_json(400, {"error": "'query' must be a non-empty string"})
return
try:
result = self.vector_db.search(
query,
top_k=int(payload.get("top_k", 8)),
per_source=int(payload.get("per_source", 5)),
sources=payload.get("sources") or None,
)
except Exception as error:
self._send_json(
500,
{"error": f"Search failed: {type(error).__name__}: {error}"},
)
return
self._send_json(200, result)
def _read_payload(self) -> dict:
try:
content_length = int(self.headers.get("Content-Length", "0"))
except ValueError as error:
raise ValueError("Invalid Content-Length") from error
if content_length < 0 or content_length > MAX_REQUEST_BYTES:
raise ValueError("Request body is too large")
if content_length == 0:
return {}
try:
return json.loads(self.rfile.read(content_length))
except (UnicodeDecodeError, json.JSONDecodeError) as error:
raise ValueError("Request body must be valid JSON") from error
def _serve_static(self, request_path: str) -> None:
if not STATIC_ROOT.exists():
self._send_json(404, {"error": "UI build not found. Run npm run build."})
return
relative = request_path.lstrip("/") or "index.html"
candidate = (STATIC_ROOT / relative).resolve()
try:
candidate.relative_to(STATIC_ROOT.resolve())
except ValueError:
self._send_json(403, {"error": "Forbidden"})
return
if not candidate.exists() or candidate.is_dir():
candidate = STATIC_ROOT / "index.html"
content_type = mimetypes.guess_type(candidate.name)[0] or "application/octet-stream"
body = candidate.read_bytes()
self.send_response(200)
self.send_header("Content-Type", content_type)
self.send_header("Content-Length", str(len(body)))
self.send_header("Cache-Control", "no-store")
self.end_headers()
self.wfile.write(body)
class CodeRagApiServer(ThreadingHTTPServer):
def __init__(
self,
server_address,
handler_class,
*,
vector_db: CodeRagVectorDB,
repository_index: Path,
synthetic_questions: Path,
):
super().__init__(server_address, handler_class)
self.vector_db = vector_db
self.repository_index = repository_index
self.synthetic_questions = synthetic_questions
def build_parser() -> argparse.ArgumentParser:
parser = argparse.ArgumentParser(description="Run the CodeRAG API server.")
parser.add_argument("--host", default="127.0.0.1")
parser.add_argument("--port", type=int, default=int(os.environ.get("PORT", "8010")))
parser.add_argument(
"--database",
type=Path,
default=ROOT / ".code_query_repo_cache" / "coderag_vectors.sqlite",
)
parser.add_argument(
"--repository-index",
type=Path,
default=ROOT / "coderag_repository_index.json",
)
parser.add_argument(
"--synthetic-questions",
type=Path,
default=ROOT / "coderag_synthetic_queries.json",
)
parser.add_argument(
"--rebuild",
action="store_true",
help="Rebuild the SQLite vector database before serving.",
)
return parser
def main() -> int:
args = build_parser().parse_args()
vector_db = CodeRagVectorDB(args.database)
vector_db.initialize()
if args.rebuild or not args.database.exists():
stats = vector_db.build(
repository_index_path=args.repository_index,
synthetic_questions_path=args.synthetic_questions,
)
print(f"Built vector database: {json.dumps(stats, indent=2)}", flush=True)
server = CodeRagApiServer(
(args.host, args.port),
CodeRagApiHandler,
vector_db=vector_db,
repository_index=args.repository_index,
synthetic_questions=args.synthetic_questions,
)
url = f"http://{args.host}:{args.port}"
print(f"CodeRAG API running at {url}", flush=True)
print("Press Ctrl+C to stop.", flush=True)
try:
server.serve_forever()
except KeyboardInterrupt:
print("\nStopping CodeRAG API.")
finally:
server.server_close()
return 0
if __name__ == "__main__":
raise SystemExit(main())