#!/usr/bin/env python3
"""Fake ollama for end-to-end tests of the sjcode daemon + client.

Serves a scripted conversation: every /api/generate call returns the next
reply from SCRIPT. Each prompt it receives is appended to prompts.log so
the test can assert what the client actually fed back to the model.

Usage: python3 fake_ollama.py PORT PROMPT_LOG
"""
import json
import sys
from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer

PORT = int(sys.argv[1])
LOG = sys.argv[2]

RUN_REPLY = (
    "I will create the file.\n"
    "```run\n"
    "printf 'hi' > hello.txt\n"
    "```\n"
)
READ_REPLY = (
    "Checking it.\n"
    "```sj:read\n"
    '{"path": "hello.txt"}\n'
    "```\n"
)
DONE_REPLY = "The file is in place.\nDone."

SCRIPT = [RUN_REPLY, READ_REPLY, DONE_REPLY]
calls = {"n": 0}


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

    def _json(self, obj):
        body = json.dumps(obj).encode()
        self.send_response(200)
        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):
        if self.path == "/api/tags":
            self._json({"models": [{"name": "fake-model"}]})
        else:
            self._json({"error": "not found"})

    def do_POST(self):
        if self.path != "/api/generate":
            self._json({"error": "not found"})
            return
        length = int(self.headers.get("Content-Length", "0"))
        payload = json.loads(self.rfile.read(length).decode())
        with open(LOG, "a") as fh:
            fh.write(json.dumps(payload.get("prompt", "")) + "\n")
        i = min(calls["n"], len(SCRIPT) - 1)
        calls["n"] += 1
        self._json({"model": "fake-model", "response": SCRIPT[i], "done": True})


if __name__ == "__main__":
    ThreadingHTTPServer(("127.0.0.1", PORT), Handler).serve_forever()