iomgaa
620b426ede
The pre-commit hook runs the whole suite, and tests/e2e/ talks to a real LLM gateway, so whether a commit is allowed depended on how fast that gateway happened to be. During the issue 14 work it blocked two commits on two different cases; both passed when rerun alone, and the suite went from 165s to 336s that hour. The wasted minutes are not the real cost. Retrying on red teaches you to read "test failed" as "gateway was slow", and a genuinely flaky bug then gets retried away too. An alarm that cries wolf stops being an alarm. test_thinking_live.py already carried the slow marker; the other three files now match it, and the release checklist gains an explicit `pytest -m slow` step so they still run where a human is watching -- without that step this change would just delete the coverage. Also raises test_flat_legacy_keys_assemble's LLM_TIMEOUT from 120 to 300, matching .env. At 120 the case allowed half of what production allows, on a gateway that needs the full 300 -- it measured 116s in a solo run. The assertion is that the flat key name parses into SourceConfig.timeout_s; the value itself was never under test.
90 lines
3.1 KiB
Python
90 lines
3.1 KiB
Python
"""真实网关 /embeddings 端点探测(M2 设计 §11.6;人类默认口径: 实现时探测)。
|
|
|
|
对 .env 的 LLM 源网关发一次真实 embeddings 请求: 支持则记录向量证据,
|
|
不支持(404/翻译为领域错误)则 skip 并把响应记录进 tests/outputs/
|
|
(降级证据)。无 EMBED scope 配置时复用 LLM 源的 base_url/api_key。
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import dataclasses
|
|
import os
|
|
from datetime import datetime
|
|
from pathlib import Path
|
|
|
|
import pytest
|
|
from dotenv import dotenv_values
|
|
|
|
from polygateway.errors import PolyGatewayError
|
|
from polygateway.transports.openai_compat import OpenAICompatTransport
|
|
from polygateway.types import SourceConfig
|
|
|
|
_ENV = {k: v for k, v in {**dotenv_values(".env"), **os.environ}.items() if v is not None}
|
|
|
|
# 真实网关调用: 与 test_thinking_live.py 同待遇标 slow(pytest addopts 默认排除,
|
|
# 显式 `pytest -m slow` 运行)。理由见 test_compat_projects.py 同处注释。
|
|
pytestmark = [
|
|
pytest.mark.slow,
|
|
pytest.mark.skipif(
|
|
"LLM__MINIMAX__1__BASE_URL" not in _ENV,
|
|
reason="缺真实网关配置(.env)",
|
|
),
|
|
]
|
|
|
|
_OUT = Path("tests/outputs/embedding")
|
|
|
|
|
|
def _record(name: str, lines: list[str]) -> Path:
|
|
_OUT.mkdir(parents=True, exist_ok=True)
|
|
path = _OUT / f"{name}_{datetime.now():%Y%m%d_%H%M%S}.md"
|
|
path.write_text("\n".join(lines) + "\n", encoding="utf-8")
|
|
return path
|
|
|
|
|
|
async def test_probe_real_gateway_embeddings():
|
|
source = SourceConfig(
|
|
name="probe_1",
|
|
provider="minimax",
|
|
base_url=_ENV["LLM__MINIMAX__1__BASE_URL"],
|
|
api_key=_ENV["LLM__MINIMAX__1__API_KEY"],
|
|
model=_ENV.get("PGW_EMBED_PROBE_MODEL", "text-embedding-v1"),
|
|
timeout_s=30.0,
|
|
est_tokens=8,
|
|
)
|
|
transport = OpenAICompatTransport()
|
|
try:
|
|
result = await transport.embed(
|
|
texts=["polygateway embedding probe"], source=source, call_id="probe"
|
|
)
|
|
except PolyGatewayError as exc:
|
|
path = _record(
|
|
"probe_unsupported",
|
|
[
|
|
"# Embedding 端点探测: 网关不支持",
|
|
f"- base_url: {source.base_url}",
|
|
f"- model: {source.model}",
|
|
f"- 错误分类: {type(exc).__name__}",
|
|
f"- status_code: {exc.status_code}",
|
|
f"- 详情: {exc}",
|
|
"",
|
|
"结论: e2e 按设计 §11.6 降级,embedding 行为由 unit 全覆盖。",
|
|
],
|
|
)
|
|
await transport.aclose()
|
|
pytest.skip(f"网关不支持 embeddings({type(exc).__name__}),证据: {path}")
|
|
else:
|
|
await transport.aclose()
|
|
assert result.dim > 0 and len(result.vectors) == 1
|
|
_record(
|
|
"probe_supported",
|
|
[
|
|
"# Embedding 端点探测: 网关支持",
|
|
f"- base_url: {source.base_url}",
|
|
f"- model: {source.model}",
|
|
f"- dim: {result.dim}",
|
|
f"- usage: {result.prompt_tokens}({result.usage_source})",
|
|
f"- 向量前 5 维: {result.vectors[0][:5]}",
|
|
f"- raw: {dataclasses.asdict(result)['raw']}",
|
|
],
|
|
)
|