PlantAI-rag-new/tests/docgen/test_docgen_conf.py

# -*- coding: utf-8 -*-
"""tests/docgen/test_docgen_conf.py —— DC-CONF-01~06（路径配置三形态 + 握手拒绝 + env + EMBED_URL）。

v1.2 签名适配（dev1 权威签名）：
  - parse_backend(vault_spec, index_spec) 两参；同形态才合法，混合抛 BackendError
  - handshake(backend) 独立函数：health + 空检索，失败抛 BackendError
  - get_fernet_key() 调用即抛（VSTORE_KEY 缺失 → RuntimeError）
  - VstorePaths(root/vault_path/index_path/embed_url/embed_model) + set_paths 注入

DC-CONF-03/04 用受控假远端（tests/docgen/fake_remote_vstore.py，本机随机端口），不碰生产端口。
执行：bash tests/docgen/run_all.sh conf
"""
from __future__ import annotations

import os
import socket
import subprocess
import sys
from pathlib import Path

import pytest

REPO = Path(__file__).resolve().parents[2]
sys.path.insert(0, str(REPO))
VENV_PY = "/home/xzq/5090/teamWorkspace/zh/plantai-rag-new-venv/bin/python"


def _importable(mod):
    try:
        __import__(mod)
        return True
    except ImportError:
        return False


def _free_port():
    with socket.socket() as s:
        s.bind(("127.0.0.1", 0))
        return s.getsockname()[1]


def _start_fake_remote(port, reject=False):
    proc = subprocess.Popen(
        [VENV_PY, str(REPO / "tests/docgen/fake_remote_vstore.py"), "--port", str(port)]
        + (["--reject"] if reject else []),
        stdout=subprocess.DEVNULL,
        stderr=subprocess.DEVNULL,
    )
    import time

    for _ in range(50):
        time.sleep(0.1)
        try:
            # 就绪探测=端口可连（reject 模式 /health 本就返回 404，不能以 200 判定）
            with socket.create_connection(("127.0.0.1", port), timeout=1):
                return proc
        except OSError:
            continue
    proc.terminate()
    pytest.skip("假远端启动失败（本组前置环境）")


# ---------------------------------------------------------------------------
# DC-CONF-01/02：裸路径与 file:// → LocalBackend（两参）
# ---------------------------------------------------------------------------

def test_DC_CONF_01_bare_path(tmp_path):
    if not _importable("vstore.backend"):
        pytest.skip("PENDING: vstore.backend 未交付")
    from vstore.backend import parse_backend

    be = parse_backend(str(tmp_path / "va"), str(tmp_path / "ix"))
    assert be.kind == "local"
    assert Path(be.vault_path) == tmp_path / "va"
    assert Path(be.index_path) == tmp_path / "ix"


def test_DC_CONF_02_file_url(tmp_path):
    if not _importable("vstore.backend"):
        pytest.skip("PENDING: vstore.backend 未交付")
    from vstore.backend import parse_backend

    be = parse_backend(
        "file://" + str(tmp_path / "va"),
        "file://" + str(tmp_path / "ix"),
    )
    assert be.kind == "local"  # file:// 不算远程
    assert str(be.vault_path) == str(tmp_path / "va")  # 前缀剥掉


def test_DC_CONF_02b_mixed_forms_rejected(tmp_path):
    """混合形态（本地 + 远程）抛 BackendError（dev1 权威签名）。"""
    if not _importable("vstore.backend"):
        pytest.skip("PENDING: vstore.backend 未交付")
    from vstore.backend import BackendError, parse_backend

    with pytest.raises(BackendError):
        parse_backend(str(tmp_path / "va"), "http://127.0.0.1:1/ix")


# ---------------------------------------------------------------------------
# DC-CONF-03：http(s):// URL → RemoteBackend + 真实受控握手通过
# ---------------------------------------------------------------------------

def test_DC_CONF_03_remote_backend_real_handshake():
    if not _importable("vstore.backend"):
        pytest.skip("PENDING: vstore.backend 未交付")
    from vstore.backend import handshake, parse_backend

    port = _free_port()
    proc = _start_fake_remote(port)
    try:
        be = parse_backend(f"http://127.0.0.1:{port}", f"http://127.0.0.1:{port}")
        assert be.kind == "remote"
        handshake(be)  # 不应抛异常（受控远端正常）
    finally:
        proc.terminate()
        proc.wait(timeout=10)


# ---------------------------------------------------------------------------
# DC-CONF-04：远程握手失败 → BackendError（拒绝启动）
# ---------------------------------------------------------------------------

def test_DC_CONF_04_handshake_reject():
    if not _importable("vstore.backend"):
        pytest.skip("PENDING: vstore.backend 未交付")
    from vstore.backend import BackendError, handshake, parse_backend

    port = _free_port()
    proc = _start_fake_remote(port, reject=True)
    try:
        be = parse_backend(f"http://127.0.0.1:{port}", f"http://127.0.0.1:{port}")
        with pytest.raises(BackendError):
            handshake(be)
    finally:
        proc.terminate()
        proc.wait(timeout=10)


# ---------------------------------------------------------------------------
# DC-CONF-05：缺 VSTORE_KEY → get_fernet_key() 调用即抛
# ---------------------------------------------------------------------------

def test_DC_CONF_05_missing_key(monkeypatch):
    if not _importable("vstore.settings"):
        pytest.skip("PENDING: vstore.settings 未交付")
    from vstore.settings import get_fernet_key

    monkeypatch.delenv("VSTORE_KEY", raising=False)
    with pytest.raises(RuntimeError) as ei:
        get_fernet_key()
    assert "VSTORE_KEY" in str(ei.value)


# ---------------------------------------------------------------------------
# DC-CONF-06：VSTORE_EMBED_URL 可配置（集群化预留：embed_batch 打到配置地址）
# ---------------------------------------------------------------------------

def test_DC_CONF_06_embed_url_configurable():
    if not _importable("vstore.embed"):
        pytest.skip("PENDING: vstore.embed 未交付")
    import json
    import threading
    from http.server import BaseHTTPRequestHandler, HTTPServer

    import vstore.embed as emod
    import vstore.settings as smod

    hits = []

    class Handler(BaseHTTPRequestHandler):
        def do_POST(self):
            hits.append(self.path)
            length = int(self.headers.get("Content-Length") or 0)
            if length:
                self.rfile.read(length)
            body = json.dumps(
                {"model": "bge-m3:567m", "embeddings": [[0.1] * 1024]}
            ).encode("utf-8")
            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 log_message(self, *args):
            pass

    srv = HTTPServer(("127.0.0.1", 0), Handler)
    port = srv.server_address[1]
    threading.Thread(target=srv.serve_forever, daemon=True).start()
    try:
        # dev1 权威注入：set_paths（VstorePaths 含 embed_url）
        smod.set_paths(
            smod.VstorePaths(
                root="/tmp/docgen-conf-root",
                vault_path="/tmp/docgen-conf-root/vault",
                index_path="/tmp/docgen-conf-root/index",
                embed_url=f"http://127.0.0.1:{port}",
                embed_model="bge-m3:567m",
            )
        )
        try:
            res = emod.embed_batch(["占位文本一", "占位文本二"])
        finally:
            smod.set_paths(smod.VstorePaths())  # 恢复默认（env 驱动）配置
        assert hits, "embed_batch 未向配置地址发出 HTTP 请求"
        assert "/api/embed" in hits[0], f"请求路径异常: {hits[0]}"
        assert all("11434" not in h for h in hits), "请求打到了默认 11434 而非配置地址"
        assert len(res) == 2
    finally:
        srv.shutdown()