#!/usr/bin/env python3
"""Fake provider servers for the CLI and Windows tests, all on 127.0.0.1.

  ollama    GET /api/tags   tagged names only, as the real server lists them
            GET /api/ps     qwen3-coder:latest loaded with a 4096-token context
            POST /api/show  a 262144-token maximum
  lmstudio  GET /api/v0/models
  keyed     GET /v1/models, only with `Bearer sk-real` or `x-api-key: sk-real`
  proxy     404 for everything - a Messages-only proxy with no model list

Ports are chosen by the OS. The first line printed is a JSON map of kind to
port, which the harness reads before running anything.
"""

import json
import sys
import threading
from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer


class Handler(BaseHTTPRequestHandler):
    def log_message(self, *args):
        pass

    def send(self, code, obj):
        body = json.dumps(obj).encode()
        self.send_response(code)
        self.send_header("content-type", "application/json")
        self.send_header("content-length", str(len(body)))
        self.end_headers()
        self.wfile.write(body)

    def do_GET(self):
        kind, path = self.server.kind, self.path
        if kind == "ollama" and path == "/api/tags":
            return self.send(200, {"models": [
                {"name": "qwen3-coder:latest", "model": "qwen3-coder:latest",
                 "details": {"parameter_size": "30.5B", "quantization_level": "Q4_K_M", "family": "qwen3moe"}},
                {"name": "glm-4.7:cloud", "model": "glm-4.7:cloud", "details": {}}]})
        if kind == "ollama" and path == "/api/ps":
            return self.send(200, {"models": [
                {"name": "qwen3-coder:latest", "model": "qwen3-coder:latest", "context_length": 4096}]})
        if kind == "lmstudio" and path == "/api/v0/models":
            return self.send(200, {"data": [
                {"id": "qwen3-coder-30b", "state": "loaded", "max_context_length": 262144},
                {"id": "gemma-small", "state": "not-loaded", "max_context_length": 8192}]})
        if kind == "keyed" and path == "/v1/models":
            if self.headers.get("authorization") == "Bearer sk-real" or self.headers.get("x-api-key") == "sk-real":
                return self.send(200, {"data": [{"id": "gw/model-a"}, {"id": "gw/model-b"}]})
            return self.send(401, {"error": "unauthorized"})
        return self.send(404, {"error": "not found"})

    def do_POST(self):
        self.rfile.read(int(self.headers.get("content-length") or 0))
        if self.server.kind == "ollama" and self.path == "/api/show":
            return self.send(200, {"model_info": {"general.architecture": "qwen3moe",
                                                  "qwen3moe.context_length": 262144}})
        return self.send(404, {"error": "not found"})


def main():
    ports = {}
    for kind in ("ollama", "lmstudio", "keyed", "proxy"):
        server = ThreadingHTTPServer(("127.0.0.1", 0), Handler)
        server.kind = kind
        ports[kind] = server.server_address[1]
        threading.Thread(target=server.serve_forever, daemon=True).start()
    print(json.dumps(ports), flush=True)
    threading.Event().wait()


if __name__ == "__main__":
    try:
        main()
    except KeyboardInterrupt:
        sys.exit(0)