ReMe/reme/reme.py
Sen Huang 8877743ca9
Some checks failed
Pre-commit / run (ubuntu-latest) (push) Waiting to run
Tests ReMe / Unit Tests - py3.12 (push) Waiting to run
Tests ReMe / Unit Tests - py3.13 (push) Waiting to run
Tests ReMe / Unit Tests - py3.11 (push) Waiting to run
Deploy Docs / deploy (push) Has been cancelled
Deploy Docs / build (push) Has been cancelled
feat(cli): route bare commands to the running server's real config (#312)
call_server now resolves the client backend/transport/host/port from the
live `reme start` process (replaying its start args through
resolve_app_config) so a bare `reme <action>` reaches the server however
it was actually launched, falling back to local config when none runs.
Explicit backend=/transport=/host=/port= still win.

Also fix as_embedding to pass `dimensions` explicitly for agentscope
>=2.0.2, and add Claude Code auto-memory/auto-dream demos to the READMEs.
2026-07-01 17:15:48 +08:00

77 lines
3.1 KiB
Python

"""ReMe memory management application entry point."""
import asyncio
import sys
from .application import Application
from .components import R
from .config import parse_args, resolve_app_config
from .enumeration import ComponentEnum
from .utils import cli_find_reme, load_env, precheck_start, running_service_config
_CLIENT_KWARGS = {"host", "port", "timeout", "transport", "command", "args"}
class ReMe(Application):
"""ReMe memory management application."""
async def call_server(action: str, **kwargs):
"""Call the running server with a client matching its *actual* service config.
The client backend and its transport/host/port are taken from the running
``reme start`` process — its start args replayed through the same
``resolve_app_config`` the server used — so a bare ``reme <action>`` reaches
the server however it was actually started (``http`` REST or ``mcp``
streamable-http / sse / stdio), including ``service.*`` overrides that never
touched the on-disk config file. Falls back to local config resolution when
no server is detected. Explicit ``backend=`` / ``transport=`` / ``host=`` /
``port=`` kwargs still win and never leak into the tool payload.
"""
# config-selecting keys steer client construction; they are not tool args.
resolve_kwargs = {}
if isinstance(kwargs.get("config"), str):
resolve_kwargs["config"] = kwargs.pop("config")
if isinstance(kwargs.get("service"), dict):
resolve_kwargs["service"] = kwargs.pop("service")
# Prefer the running server's real config; fall back to the local config file.
service = running_service_config()
if service is None:
service = resolve_app_config(**resolve_kwargs).get("service")
service = service if isinstance(service, dict) else {}
backend: str = kwargs.pop("backend", None) or service.get("backend", "http")
# Seed client kwargs from the service config only when we are actually using
# that service's backend — transport/host/port are backend-specific, so an
# explicit backend override must not inherit the other backend's settings.
seed = service if backend == service.get("backend") else {}
client_kwargs = {k: seed[k] for k in _CLIENT_KWARGS if k in seed}
client_kwargs.update({key: kwargs.pop(key) for key in list(kwargs) if key in _CLIENT_KWARGS})
client_cls = R.get(ComponentEnum.CLIENT, backend)
if client_cls is None:
raise ValueError(f"Unknown client backend: {backend!r}")
async with client_cls(**client_kwargs) as client:
async for chunk in client(action=action, **kwargs):
print(chunk, end="", flush=True)
print()
def main():
"""Parse CLI arguments and launch the appropriate mode."""
action, kwargs = parse_args(*sys.argv[1:])
if action == "start":
load_env()
kwargs = resolve_app_config(**kwargs)
if not precheck_start(kwargs.get("service")):
return
ReMe(**kwargs).run_app()
elif action == "find_reme":
cli_find_reme()
else:
asyncio.run(call_server(action, **kwargs))
if __name__ == "__main__":
main()