feat: 军事科技每日资讯推送系统 - Docker部署 + 日志系统 + 数据目录重组
This commit is contained in:
poiuy
@@ -0,0 +1 @@
|
||||
# pytest tests for daily-news project
|
||||
@@ -0,0 +1,154 @@
|
||||
"""pytest fixtures for daily-news project"""
|
||||
|
||||
import json
|
||||
import pytest
|
||||
import sys
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from pathlib import Path
|
||||
from unittest.mock import MagicMock
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def temp_dir(tmp_path):
|
||||
"""临时目录fixture"""
|
||||
return tmp_path
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_config():
|
||||
"""示例配置"""
|
||||
return {
|
||||
"sources": {
|
||||
"base_opml": "resources/rss.opml",
|
||||
"add": [],
|
||||
"block": [],
|
||||
"block_domains": ["*.substack.com"],
|
||||
},
|
||||
"filter": {
|
||||
"min_score": 60,
|
||||
"hot_threshold": 90,
|
||||
"context_days": 2,
|
||||
"keep_days": 7,
|
||||
},
|
||||
"schedule": {
|
||||
"fetch_interval_minutes": 30,
|
||||
"push_cron": ["0 8 * * *", "0 17 * * *"],
|
||||
"timezone_hours": 8,
|
||||
},
|
||||
"llm": {
|
||||
"provider": "openai",
|
||||
"model": "gpt-4o-mini",
|
||||
"baseUrl": "https://api.openai.com/v1",
|
||||
"apiKeyName": "OPENAI_API_KEY",
|
||||
"max_prompt_chars": 10000,
|
||||
"max_concurrent_batches": 3,
|
||||
},
|
||||
"push": {
|
||||
"discord": {
|
||||
"enabled": True,
|
||||
"webhook_url": "https://discord.com/api/webhooks/test/abc",
|
||||
},
|
||||
"feishu": {"enabled": False, "apiKeyName": "FEISHU_WEBHOOK_URL"},
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_opml(temp_dir):
|
||||
"""示例OPML文件"""
|
||||
opml_content = """<?xml version="1.0"?>
|
||||
<opml version="2.0">
|
||||
<body>
|
||||
<outline title="Feed1" xmlUrl="http://feed1.com/rss" type="rss" category="tech"/>
|
||||
<outline title="Feed2" xmlUrl="http://feed2.com/rss" type="rss" category="ai"/>
|
||||
</body>
|
||||
</opml>"""
|
||||
opml_file = temp_dir / "test.opml"
|
||||
opml_file.write_text(opml_content)
|
||||
return str(opml_file)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_entry():
|
||||
"""示例新闻条目"""
|
||||
return {
|
||||
"title": "Test Article Title",
|
||||
"link": "https://example.com/article",
|
||||
"published": datetime.now(timezone.utc).isoformat(),
|
||||
"fetched_at": datetime.now(timezone.utc).isoformat(),
|
||||
"source": "Test Source",
|
||||
"content": "<p>Test content</p>",
|
||||
"summary": "Test summary",
|
||||
"tags": ["AI", "Tech"],
|
||||
"score": 85,
|
||||
}
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_entries():
|
||||
"""示例新闻条目列表"""
|
||||
now = datetime.now(timezone.utc)
|
||||
return [
|
||||
{
|
||||
"title": "Article 1",
|
||||
"link": "https://example.com/1",
|
||||
"published": now.isoformat(),
|
||||
"fetched_at": now.isoformat(),
|
||||
"source": "Source1",
|
||||
"content": "Content 1",
|
||||
"summary": "Summary 1",
|
||||
"tags": ["AI"],
|
||||
"score": 85,
|
||||
},
|
||||
{
|
||||
"title": "Article 2",
|
||||
"link": "https://example.com/2",
|
||||
"published": (now - timedelta(hours=1)).isoformat(),
|
||||
"fetched_at": now.isoformat(),
|
||||
"source": "Source2",
|
||||
"content": "Content 2",
|
||||
"summary": "Summary 2",
|
||||
"tags": ["Tech"],
|
||||
"score": 70,
|
||||
},
|
||||
{
|
||||
"title": "Article 3",
|
||||
"link": "https://example.com/3",
|
||||
"published": (now - timedelta(hours=2)).isoformat(),
|
||||
"fetched_at": now.isoformat(),
|
||||
"source": "Source3",
|
||||
"content": "Content 3",
|
||||
"summary": "Summary 3",
|
||||
"tags": ["News"],
|
||||
"score": 55,
|
||||
},
|
||||
]
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def sample_fetch_json(temp_dir, sample_entries):
|
||||
"""示例fetch JSON文件"""
|
||||
data = {
|
||||
"meta": {"date": datetime.now().date().isoformat()},
|
||||
"entries": sample_entries,
|
||||
}
|
||||
json_file = temp_dir / "fetch-test.json"
|
||||
json_file.write_text(json.dumps(data, ensure_ascii=False, indent=2))
|
||||
return str(json_file)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def mock_httpx_session():
|
||||
"""Mock httpx/aiohttp session"""
|
||||
mock_response = MagicMock()
|
||||
mock_response.status = 200
|
||||
mock_response.text = "<rss></rss>"
|
||||
|
||||
mock_session = MagicMock()
|
||||
mock_session.__aenter__ = MagicMock(return_value=mock_session)
|
||||
mock_session.__aexit__ = MagicMock(return_value=None)
|
||||
mock_session.get = MagicMock(return_value=mock_response)
|
||||
|
||||
return mock_session
|
||||
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
@@ -0,0 +1,249 @@
|
||||
"""配置模块测试"""
|
||||
|
||||
import json
|
||||
import pytest
|
||||
import sys
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from config import (
|
||||
load_config,
|
||||
parse_opml,
|
||||
merge_sources,
|
||||
get_timezone,
|
||||
)
|
||||
|
||||
|
||||
class TestLoadConfig:
|
||||
"""测试配置加载"""
|
||||
|
||||
def test_load_valid_config(self, temp_dir):
|
||||
config = {
|
||||
"sources": {"base_opml": "test.opml", "add": [], "block": []},
|
||||
"filter": {"min_score": 60},
|
||||
"schedule": {"fetch_interval_minutes": 30},
|
||||
"llm": {"provider": "groq", "model": "moonshotai/kimi-k2-instruct"},
|
||||
"push": {"discord": {"enabled": False}},
|
||||
}
|
||||
config_file = temp_dir / "config.json"
|
||||
config_file.write_text(json.dumps(config))
|
||||
|
||||
result = load_config(str(config_file))
|
||||
assert result["filter"]["min_score"] == 60
|
||||
assert result["llm"]["provider"] == "groq"
|
||||
|
||||
def test_load_missing_file(self):
|
||||
with pytest.raises(FileNotFoundError):
|
||||
load_config("nonexistent.json")
|
||||
|
||||
def test_load_invalid_json(self, temp_dir):
|
||||
config_file = temp_dir / "config.json"
|
||||
config_file.write_text("invalid json")
|
||||
with pytest.raises(json.JSONDecodeError):
|
||||
load_config(str(config_file))
|
||||
|
||||
def test_load_config_with_sources(self, temp_dir, sample_opml):
|
||||
config = {
|
||||
"sources": {
|
||||
"base_opml": sample_opml,
|
||||
"add": [
|
||||
{
|
||||
"title": "Add1",
|
||||
"xmlUrl": "http://add1.com/rss",
|
||||
"category": "test",
|
||||
}
|
||||
],
|
||||
"block": [],
|
||||
},
|
||||
"filter": {"min_score": 60},
|
||||
"schedule": {"fetch_interval_minutes": 30, "timezone_hours": 8},
|
||||
"llm": {"provider": "test"},
|
||||
"push": {"discord": {"enabled": False}},
|
||||
}
|
||||
config_file = temp_dir / "config.json"
|
||||
config_file.write_text(json.dumps(config))
|
||||
|
||||
result = load_config(str(config_file))
|
||||
assert len(result["sources"]["add"]) == 1
|
||||
|
||||
|
||||
class TestParseOpml:
|
||||
"""测试OPML解析"""
|
||||
|
||||
def test_parse_valid_opml(self, temp_dir):
|
||||
opml_content = """<?xml version="1.0"?>
|
||||
<opml version="2.0">
|
||||
<body>
|
||||
<outline title="Feed1" xmlUrl="http://feed1.com/rss" type="rss"/>
|
||||
</body>
|
||||
</opml>"""
|
||||
opml_file = temp_dir / "test.opml"
|
||||
opml_file.write_text(opml_content)
|
||||
|
||||
feeds = parse_opml(str(opml_file))
|
||||
assert len(feeds) == 1
|
||||
assert feeds[0]["title"] == "Feed1"
|
||||
assert feeds[0]["xmlUrl"] == "http://feed1.com/rss"
|
||||
|
||||
def test_parse_missing_file(self):
|
||||
feeds = parse_opml("nonexistent.opml")
|
||||
assert feeds == []
|
||||
|
||||
def test_parse_opml_with_category(self, temp_dir):
|
||||
opml_content = """<?xml version="1.0"?>
|
||||
<opml version="2.0">
|
||||
<body>
|
||||
<outline title="TechFeed" xmlUrl="http://tech.com/rss" type="rss" category="技术"/>
|
||||
<outline title="AIFeed" xmlUrl="http://ai.com/rss" type="rss" category="AI"/>
|
||||
</body>
|
||||
</opml>"""
|
||||
opml_file = temp_dir / "test.opml"
|
||||
opml_file.write_text(opml_content)
|
||||
|
||||
feeds = parse_opml(str(opml_file))
|
||||
assert len(feeds) == 2
|
||||
assert feeds[0]["category"] == "技术"
|
||||
assert feeds[1]["category"] == "AI"
|
||||
|
||||
def test_parse_opml_empty_body(self, temp_dir):
|
||||
opml_content = """<?xml version="1.0"?>
|
||||
<opml version="2.0">
|
||||
<body>
|
||||
</body>
|
||||
</opml>"""
|
||||
opml_file = temp_dir / "test.opml"
|
||||
opml_file.write_text(opml_content)
|
||||
|
||||
feeds = parse_opml(str(opml_file))
|
||||
assert feeds == []
|
||||
|
||||
|
||||
class TestMergeSources:
|
||||
"""测试源合并"""
|
||||
|
||||
def test_merge_base_and_add(self, sample_opml):
|
||||
config = {
|
||||
"base_opml": sample_opml,
|
||||
"add": [
|
||||
{"title": "Feed3", "xmlUrl": "http://feed3.com/rss", "category": "test"}
|
||||
],
|
||||
"block": [],
|
||||
}
|
||||
sources = merge_sources(config)
|
||||
assert len(sources) == 3
|
||||
|
||||
def test_block_by_xmlUrl(self, sample_opml):
|
||||
config = {
|
||||
"base_opml": sample_opml,
|
||||
"add": [],
|
||||
"block": [{"title": "Block1", "xmlUrl": "http://feed1.com/rss"}],
|
||||
}
|
||||
sources = merge_sources(config)
|
||||
assert all(s["xmlUrl"] != "http://feed1.com/rss" for s in sources)
|
||||
assert len(sources) == 1
|
||||
|
||||
def test_deduplicate_by_xmlUrl(self, sample_opml):
|
||||
config = {
|
||||
"base_opml": sample_opml,
|
||||
"add": [
|
||||
{
|
||||
"title": "Duplicate",
|
||||
"xmlUrl": "http://feed1.com/rss",
|
||||
"category": "test",
|
||||
}
|
||||
],
|
||||
"block": [],
|
||||
}
|
||||
sources = merge_sources(config)
|
||||
urls = [s["xmlUrl"] for s in sources]
|
||||
assert len(urls) == len(set(urls))
|
||||
assert len(sources) == 2
|
||||
|
||||
def test_block_domains_wildcard(self, temp_dir):
|
||||
opml_content = """<?xml version="1.0"?>
|
||||
<opml version="2.0">
|
||||
<body>
|
||||
<outline title="Substack" xmlUrl="https://tech.substack.com/rss" type="rss"/>
|
||||
<outline title="Blog" xmlUrl="https://tech.blog/rss" type="rss"/>
|
||||
</body>
|
||||
</opml>"""
|
||||
opml_file = temp_dir / "test.opml"
|
||||
opml_file.write_text(opml_content)
|
||||
|
||||
config = {
|
||||
"base_opml": str(opml_file),
|
||||
"add": [],
|
||||
"block": [],
|
||||
"block_domains": ["*.substack.com"],
|
||||
}
|
||||
sources = merge_sources(config)
|
||||
assert len(sources) == 1
|
||||
assert sources[0]["xmlUrl"] == "https://tech.blog/rss"
|
||||
|
||||
def test_block_domains_exact(self, temp_dir):
|
||||
opml_content = """<?xml version="1.0"?>
|
||||
<opml version="2.0">
|
||||
<body>
|
||||
<outline title="YouTube" xmlUrl="https://youtube.com/feed" type="rss"/>
|
||||
<outline title="Blog" xmlUrl="https://tech.blog/rss" type="rss"/>
|
||||
</body>
|
||||
</opml>"""
|
||||
opml_file = temp_dir / "test.opml"
|
||||
opml_file.write_text(opml_content)
|
||||
|
||||
config = {
|
||||
"base_opml": str(opml_file),
|
||||
"add": [],
|
||||
"block": [],
|
||||
"block_domains": ["youtube.com"],
|
||||
}
|
||||
sources = merge_sources(config)
|
||||
assert len(sources) == 1
|
||||
assert sources[0]["xmlUrl"] == "https://tech.blog/rss"
|
||||
|
||||
def test_block_domains_subdomain(self, temp_dir):
|
||||
opml_content = """<?xml version="1.0"?>
|
||||
<opml version="2.0">
|
||||
<body>
|
||||
<outline title="Substack1" xmlUrl="https://substack.com/feed" type="rss"/>
|
||||
<outline title="Substack2" xmlUrl="https://ai.substack.com/feed" type="rss"/>
|
||||
<outline title="Blog" xmlUrl="https://tech.blog/rss" type="rss"/>
|
||||
</body>
|
||||
</opml>"""
|
||||
opml_file = temp_dir / "test.opml"
|
||||
opml_file.write_text(opml_content)
|
||||
|
||||
config = {
|
||||
"base_opml": str(opml_file),
|
||||
"add": [],
|
||||
"block": [],
|
||||
"block_domains": ["*.substack.com"],
|
||||
}
|
||||
sources = merge_sources(config)
|
||||
assert len(sources) == 1
|
||||
assert sources[0]["xmlUrl"] == "https://tech.blog/rss"
|
||||
|
||||
|
||||
class TestGetTimezone:
|
||||
"""测试时区获取"""
|
||||
|
||||
def test_get_timezone_from_config(self, sample_config):
|
||||
tz = get_timezone(sample_config)
|
||||
assert isinstance(tz, timezone)
|
||||
assert tz.utcoffset(datetime.now()).total_seconds() == 8 * 3600
|
||||
|
||||
def test_get_timezone_none_config(self):
|
||||
tz = get_timezone(None)
|
||||
assert isinstance(tz, timezone)
|
||||
|
||||
def test_get_timezone_no_timezone_hours(self):
|
||||
config = {"schedule": {}}
|
||||
tz = get_timezone(config)
|
||||
assert isinstance(tz, timezone)
|
||||
|
||||
def test_get_timezone_custom_hours(self):
|
||||
config = {"schedule": {"timezone_hours": -5}}
|
||||
tz = get_timezone(config)
|
||||
assert tz.utcoffset(datetime.now()).total_seconds() == -5 * 3600
|
||||
@@ -0,0 +1,190 @@
|
||||
"""RSS抓取模块测试"""
|
||||
|
||||
import pytest
|
||||
import sys
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from pathlib import Path
|
||||
from unittest.mock import MagicMock, patch, AsyncMock
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from fetcher import (
|
||||
parse_entry_time,
|
||||
fetch_single_feed_async,
|
||||
fetch_all_feeds,
|
||||
DEFAULT_FEED_TIMEOUT,
|
||||
)
|
||||
|
||||
|
||||
class TestParseEntryTime:
|
||||
"""测试时间解析"""
|
||||
|
||||
def test_parse_published_parsed(self):
|
||||
entry = MagicMock()
|
||||
entry.published_parsed = (2024, 1, 15, 10, 30, 0, 0, 0, 0)
|
||||
|
||||
result = parse_entry_time(entry)
|
||||
assert result is not None
|
||||
assert result.year == 2024
|
||||
assert result.month == 1
|
||||
assert result.day == 15
|
||||
assert result.tzinfo == timezone.utc
|
||||
|
||||
def test_parse_updated_parsed(self):
|
||||
entry = MagicMock()
|
||||
entry.published_parsed = None
|
||||
entry.updated_parsed = (2024, 1, 15, 10, 30, 0, 0, 0, 0)
|
||||
|
||||
result = parse_entry_time(entry)
|
||||
assert result is not None
|
||||
assert result.year == 2024
|
||||
assert result.month == 1
|
||||
assert result.day == 15
|
||||
|
||||
def test_parse_no_time(self):
|
||||
entry = MagicMock()
|
||||
entry.published_parsed = None
|
||||
entry.updated_parsed = None
|
||||
|
||||
result = parse_entry_time(entry)
|
||||
assert result is None
|
||||
|
||||
|
||||
class TestFetchSingleFeedAsync:
|
||||
"""测试单源抓取"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_fetch_success(self, temp_dir):
|
||||
rss_content = """<?xml version="1.0"?>
|
||||
<rss version="2.0">
|
||||
<channel>
|
||||
<title>Test Feed</title>
|
||||
<item>
|
||||
<title>Article 1</title>
|
||||
<link>https://example.com/1</link>
|
||||
<pubDate>Mon, 15 Jan 2024 10:00:00 GMT</pubDate>
|
||||
<description>Test description</description>
|
||||
</item>
|
||||
</channel>
|
||||
</rss>"""
|
||||
|
||||
feed_info = {"title": "Test Feed", "xmlUrl": "http://test.com/rss"}
|
||||
cutoff = datetime(2024, 1, 1, tzinfo=timezone.utc)
|
||||
|
||||
mock_response = MagicMock()
|
||||
mock_response.status = 200
|
||||
mock_response.text = AsyncMock(return_value=rss_content)
|
||||
|
||||
mock_session = MagicMock()
|
||||
mock_session.__aenter__ = AsyncMock(return_value=mock_session)
|
||||
mock_session.__aexit__ = AsyncMock(return_value=None)
|
||||
mock_session.get = MagicMock(
|
||||
return_value=AsyncMock(
|
||||
__aenter__=AsyncMock(return_value=mock_response),
|
||||
__aexit__=AsyncMock(return_value=None),
|
||||
)
|
||||
)
|
||||
|
||||
with patch("aiohttp.ClientSession", return_value=mock_session):
|
||||
entries = await fetch_single_feed_async(
|
||||
feed_info, cutoff, session=mock_session
|
||||
)
|
||||
|
||||
assert len(entries) == 1
|
||||
assert entries[0]["title"] == "Article 1"
|
||||
assert entries[0]["link"] == "https://example.com/1"
|
||||
assert entries[0]["source"] == "Test Feed"
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_fetch_http_error(self):
|
||||
feed_info = {"title": "Test Feed", "xmlUrl": "http://test.com/rss"}
|
||||
cutoff = datetime(2024, 1, 1, tzinfo=timezone.utc)
|
||||
|
||||
mock_response = MagicMock()
|
||||
mock_response.status = 404
|
||||
|
||||
mock_session = MagicMock()
|
||||
mock_session.get = MagicMock(
|
||||
return_value=AsyncMock(
|
||||
__aenter__=AsyncMock(return_value=mock_response),
|
||||
__aexit__=AsyncMock(return_value=None),
|
||||
)
|
||||
)
|
||||
|
||||
entries = await fetch_single_feed_async(feed_info, cutoff, session=mock_session)
|
||||
assert entries == []
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_fetch_timeout(self):
|
||||
feed_info = {"title": "Test Feed", "xmlUrl": "http://test.com/rss"}
|
||||
cutoff = datetime(2024, 1, 1, tzinfo=timezone.utc)
|
||||
|
||||
import aiohttp
|
||||
|
||||
mock_session = MagicMock()
|
||||
mock_session.get = MagicMock(side_effect=aiohttp.ServerTimeoutError())
|
||||
|
||||
entries = await fetch_single_feed_async(feed_info, cutoff, session=mock_session)
|
||||
assert entries == []
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_fetch_cutoff_filter(self):
|
||||
feed_info = {"title": "Test Feed", "xmlUrl": "http://test.com/rss"}
|
||||
cutoff = datetime(2024, 1, 10, tzinfo=timezone.utc)
|
||||
|
||||
import feedparser
|
||||
|
||||
with patch(
|
||||
"fetcher.fetch_single_feed_async", new_callable=AsyncMock
|
||||
) as mock_fetch:
|
||||
mock_fetch.return_value = [
|
||||
{
|
||||
"title": "New",
|
||||
"link": "https://example.com/new",
|
||||
"published": datetime(2024, 1, 15, tzinfo=timezone.utc),
|
||||
}
|
||||
]
|
||||
|
||||
result = await mock_fetch(feed_info, cutoff)
|
||||
|
||||
assert len(result) == 1
|
||||
|
||||
|
||||
class TestFetchAllFeeds:
|
||||
"""测试并发抓取"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_concurrent_limit(self):
|
||||
feeds = [
|
||||
{"title": f"Feed{i}", "xmlUrl": f"http://feed{i}.com/rss"}
|
||||
for i in range(20)
|
||||
]
|
||||
cutoff = datetime(2024, 1, 1, tzinfo=timezone.utc)
|
||||
|
||||
with patch(
|
||||
"fetcher.fetch_single_feed_async", new_callable=AsyncMock
|
||||
) as mock_fetch:
|
||||
mock_fetch.return_value = []
|
||||
await fetch_all_feeds(feeds, cutoff, max_workers=5)
|
||||
|
||||
assert mock_fetch.call_count == 20
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_empty_feeds(self):
|
||||
cutoff = datetime(2024, 1, 1, tzinfo=timezone.utc)
|
||||
entries = await fetch_all_feeds([], cutoff)
|
||||
assert entries == []
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_default_timeout(self):
|
||||
feed_info = {"title": "Test", "xmlUrl": "http://test.com"}
|
||||
cutoff = datetime.now(timezone.utc)
|
||||
|
||||
with patch(
|
||||
"fetcher.fetch_single_feed_async", new_callable=AsyncMock
|
||||
) as mock_fetch:
|
||||
mock_fetch.return_value = []
|
||||
|
||||
await mock_fetch(feed_info, cutoff, timeout=None)
|
||||
|
||||
mock_fetch.assert_called_once()
|
||||
@@ -0,0 +1,330 @@
|
||||
"""LLM模块测试"""
|
||||
|
||||
import json
|
||||
import pytest
|
||||
import sys
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock, MagicMock
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from llm import (
|
||||
load_prompt,
|
||||
_parse_llm_json_response,
|
||||
_split_entries_for_batch,
|
||||
_build_batch_prompt,
|
||||
_merge_scores,
|
||||
_score_single_batch,
|
||||
call_llm,
|
||||
check_llm_available,
|
||||
generate_immediate_push,
|
||||
score_batch,
|
||||
)
|
||||
|
||||
|
||||
class TestLoadPrompt:
|
||||
"""测试提示词加载"""
|
||||
|
||||
def test_load_prompt_basic(self, temp_dir):
|
||||
prompt_file = temp_dir / "test.txt"
|
||||
prompt_file.write_text("Hello {name}!")
|
||||
|
||||
result = load_prompt(str(prompt_file), name="World")
|
||||
assert result == "Hello World!"
|
||||
|
||||
def test_load_prompt_missing_file(self):
|
||||
with pytest.raises(FileNotFoundError):
|
||||
load_prompt("nonexistent.txt")
|
||||
|
||||
def test_load_prompt_with_braces(self, temp_dir):
|
||||
prompt_file = temp_dir / "test.txt"
|
||||
prompt_file.write_text("Hello {name}, curly braces: { }")
|
||||
|
||||
result = load_prompt(str(prompt_file), name="World")
|
||||
assert result == "Hello World, curly braces: { }"
|
||||
|
||||
def test_load_prompt_multiple_vars(self, temp_dir):
|
||||
prompt_file = temp_dir / "test.txt"
|
||||
prompt_file.write_text("{greeting} {name}, you have {count} messages")
|
||||
|
||||
result = load_prompt(str(prompt_file), greeting="Hi", name="Alice", count=5)
|
||||
assert result == "Hi Alice, you have 5 messages"
|
||||
|
||||
|
||||
class TestParseLlmJsonResponse:
|
||||
"""测试LLM响应解析"""
|
||||
|
||||
def test_parse_json_array(self):
|
||||
response = '[{"link": "https://example.com", "score": 80}]'
|
||||
result = _parse_llm_json_response(response)
|
||||
assert len(result) == 1
|
||||
assert result[0]["link"] == "https://example.com"
|
||||
assert result[0]["score"] == 80
|
||||
|
||||
def test_parse_with_markdown_codeblock(self):
|
||||
response = """```json
|
||||
[{"link": "https://example.com", "score": 80}]
|
||||
```"""
|
||||
result = _parse_llm_json_response(response)
|
||||
assert len(result) == 1
|
||||
|
||||
def test_parse_with_codeblock(self):
|
||||
response = """```
|
||||
[{"link": "https://example.com", "score": 80}]
|
||||
```"""
|
||||
result = _parse_llm_json_response(response)
|
||||
assert len(result) == 1
|
||||
|
||||
def test_parse_invalid_response(self):
|
||||
response = "This is not JSON at all"
|
||||
with pytest.raises(ValueError):
|
||||
_parse_llm_json_response(response)
|
||||
|
||||
|
||||
class TestSplitEntriesForBatch:
|
||||
"""测试条目分批"""
|
||||
|
||||
def test_split_empty(self):
|
||||
result = _split_entries_for_batch([])
|
||||
assert result == []
|
||||
|
||||
def test_split_single_batch(self):
|
||||
entries = [
|
||||
{
|
||||
"link": f"https://example.com/{i}",
|
||||
"title": f"Title{i}",
|
||||
"content": "x" * 100,
|
||||
}
|
||||
for i in range(5)
|
||||
]
|
||||
result = _split_entries_for_batch(entries, max_prompt_chars=10000)
|
||||
assert len(result) == 1
|
||||
|
||||
def test_split_multiple_batches(self):
|
||||
entries = [
|
||||
{
|
||||
"link": f"https://example.com/{i}",
|
||||
"title": f"Title{i}",
|
||||
"content": "x" * 5000,
|
||||
}
|
||||
for i in range(10)
|
||||
]
|
||||
result = _split_entries_for_batch(entries, max_prompt_chars=10000)
|
||||
assert len(result) > 1
|
||||
|
||||
|
||||
class TestBuildBatchPrompt:
|
||||
"""测试构建批量提示词"""
|
||||
|
||||
def test_build_batch_prompt_basic(self):
|
||||
entries = [
|
||||
{
|
||||
"link": "https://example.com/1",
|
||||
"title": "Title1",
|
||||
"source": "Source1",
|
||||
"published": "2024-01-15",
|
||||
"content": "Content",
|
||||
}
|
||||
]
|
||||
result = _build_batch_prompt(entries)
|
||||
assert "Title1" in result
|
||||
assert "https://example.com/1" in result
|
||||
|
||||
|
||||
class TestMergeScores:
|
||||
"""测试评分合并"""
|
||||
|
||||
def test_merge_scores_basic(self):
|
||||
entries = [
|
||||
{"link": "https://example.com/1", "title": "Title1"},
|
||||
{"link": "https://example.com/2", "title": "Title2"},
|
||||
]
|
||||
scores = [
|
||||
{
|
||||
"link": "https://example.com/1",
|
||||
"score": 85,
|
||||
"tags": ["AI"],
|
||||
"summary": "Summary1",
|
||||
},
|
||||
{
|
||||
"link": "https://example.com/2",
|
||||
"score": 70,
|
||||
"tags": ["Tech"],
|
||||
"summary": "Summary2",
|
||||
},
|
||||
]
|
||||
result = _merge_scores(entries, scores)
|
||||
|
||||
assert result[0]["score"] == 85
|
||||
assert result[0]["tags"] == ["AI"]
|
||||
assert result[1]["score"] == 70
|
||||
|
||||
def test_merge_scores_partial(self):
|
||||
entries = [
|
||||
{"link": "https://example.com/1", "title": "Title1", "score": 50},
|
||||
{"link": "https://example.com/2", "title": "Title2", "score": 60},
|
||||
]
|
||||
scores = [{"link": "https://example.com/1", "score": 85}]
|
||||
result = _merge_scores(entries, scores)
|
||||
|
||||
assert result[0]["score"] == 85
|
||||
assert result[1]["score"] == 60
|
||||
|
||||
|
||||
class TestCallLlm:
|
||||
"""测试LLM调用"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_call_llm_success(self):
|
||||
config = {
|
||||
"model": "gpt-4",
|
||||
"baseUrl": "https://api.openai.com/v1",
|
||||
"apiKeyName": "OPENAI_API_KEY",
|
||||
}
|
||||
|
||||
with patch("llm.call_llm", new_callable=AsyncMock) as mock_call:
|
||||
mock_call.return_value = "Test response"
|
||||
|
||||
result = await mock_call("Test prompt", config)
|
||||
|
||||
assert result == "Test response"
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_call_llm_missing_key(self):
|
||||
config = {"model": "gpt-4", "apiKeyName": "MISSING_KEY"}
|
||||
|
||||
with pytest.raises(ValueError, match="未设置MISSING_KEY"):
|
||||
await call_llm("Test prompt", config)
|
||||
|
||||
|
||||
class TestLlmHealthCheck:
|
||||
"""测试LLM可用性检查"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_check_llm_available_success(self, sample_config):
|
||||
with patch("llm.call_llm", new_callable=AsyncMock) as mock_call:
|
||||
mock_call.return_value = "OK"
|
||||
|
||||
result = await check_llm_available(sample_config["llm"])
|
||||
|
||||
assert result == "OK"
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_check_llm_available_empty_response(self, sample_config):
|
||||
with patch("llm.call_llm", new_callable=AsyncMock) as mock_call:
|
||||
mock_call.return_value = " "
|
||||
|
||||
with pytest.raises(RuntimeError, match="返回空响应"):
|
||||
await check_llm_available(sample_config["llm"])
|
||||
|
||||
|
||||
class TestImmediatePush:
|
||||
"""测试即时推送生成"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_generate_immediate_push_failure_returns_error(
|
||||
self, sample_entries, sample_config
|
||||
):
|
||||
with patch("llm.load_prompt", return_value="prompt"), patch(
|
||||
"llm.call_llm", new_callable=AsyncMock
|
||||
) as mock_call:
|
||||
mock_call.side_effect = RuntimeError("boom")
|
||||
|
||||
content, error = await generate_immediate_push(
|
||||
sample_entries[:1], sample_config["llm"], recent_push_context=""
|
||||
)
|
||||
|
||||
assert content == ""
|
||||
assert error == "生成即时推送失败: boom"
|
||||
|
||||
|
||||
class TestScoreBatch:
|
||||
"""测试批量评分"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_score_batch_empty(self, sample_config):
|
||||
result, errors = await score_batch([], sample_config["llm"])
|
||||
assert result == []
|
||||
assert errors == []
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_score_batch_single(self, sample_entries, sample_config):
|
||||
entries = sample_entries[:1]
|
||||
|
||||
mock_scores = [
|
||||
{
|
||||
"link": entries[0]["link"],
|
||||
"score": 85,
|
||||
"tags": ["AI"],
|
||||
"summary": "Test summary",
|
||||
}
|
||||
]
|
||||
|
||||
with patch("llm._score_single_batch", new_callable=AsyncMock) as mock_score:
|
||||
mock_score.return_value = (mock_scores, [])
|
||||
result, errors = await score_batch(entries, sample_config["llm"])
|
||||
|
||||
assert len(result) == 1
|
||||
assert result[0]["score"] == 85
|
||||
assert errors == []
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_score_single_batch_failure_returns_empty_results(
|
||||
self, sample_entries, sample_config
|
||||
):
|
||||
with patch("llm.call_llm", new_callable=AsyncMock) as mock_call:
|
||||
mock_call.side_effect = RuntimeError("boom")
|
||||
results, errors = await _score_single_batch(
|
||||
sample_entries[:2], sample_config["llm"]
|
||||
)
|
||||
|
||||
assert results == []
|
||||
assert errors == ["批次1 评分失败: boom"]
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_score_single_batch_reconcile_partial_results(
|
||||
self, sample_entries, sample_config
|
||||
):
|
||||
entries = sample_entries[:2]
|
||||
llm_results = [
|
||||
{
|
||||
"link": entries[0]["link"],
|
||||
"score": 91,
|
||||
"tags": ["AI"],
|
||||
"summary": "Matched result",
|
||||
}
|
||||
]
|
||||
|
||||
with patch("llm.call_llm", new_callable=AsyncMock) as mock_call:
|
||||
mock_call.return_value = json.dumps(llm_results, ensure_ascii=False)
|
||||
results, errors = await _score_single_batch(entries, sample_config["llm"])
|
||||
|
||||
assert len(results) == 1
|
||||
assert results[0]["score"] == 91
|
||||
assert len(errors) == 1
|
||||
assert "评分结果异常" in errors[0]
|
||||
assert "输入2" in errors[0]
|
||||
assert "返回1" in errors[0]
|
||||
assert "匹配1" in errors[0]
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_score_single_batch_keeps_full_results(self, sample_entries, sample_config):
|
||||
entries = sample_entries[:3]
|
||||
llm_results = [
|
||||
{
|
||||
"link": entry["link"],
|
||||
"score": 88,
|
||||
"tags": ["AI"],
|
||||
"summary": f"Summary for {index}",
|
||||
}
|
||||
for index, entry in enumerate(entries, start=1)
|
||||
]
|
||||
|
||||
with patch("llm.call_llm", new_callable=AsyncMock) as mock_call:
|
||||
mock_call.return_value = json.dumps(llm_results, ensure_ascii=False)
|
||||
results, errors = await _score_single_batch(entries, sample_config["llm"])
|
||||
|
||||
assert len(results) == 3
|
||||
assert [result["link"] for result in results] == [entry["link"] for entry in entries]
|
||||
assert errors == []
|
||||
@@ -0,0 +1,101 @@
|
||||
"""测试新增 LLM 函数 (summarize_github_trending 等)"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from llm import summarize_github_trending, select_ai_related_hn, summarize_hackernews
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_summarize_github_trending_happy_path(tmp_path):
|
||||
prompt_path = tmp_path / "section_github.md"
|
||||
prompt_path.write_text("Repos: {repos_json}\nmax_items={max_items}", encoding="utf-8")
|
||||
|
||||
config = {
|
||||
"model": "x",
|
||||
"baseUrl": "http://x",
|
||||
"apiKeyName": "DEEPSEEK_API_KEY",
|
||||
"prompts": {"section_github": str(prompt_path)},
|
||||
"sections": {"github_trending": {"max_items": 3}},
|
||||
}
|
||||
enriched = [{"full_name": "o/r", "readme_excerpt": "rm"}]
|
||||
|
||||
with patch("llm.call_llm", new=AsyncMock(return_value="## md")):
|
||||
md, err = await summarize_github_trending(enriched, config)
|
||||
|
||||
assert md == "## md"
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_summarize_github_trending_llm_failure_returns_error(tmp_path):
|
||||
prompt_path = tmp_path / "section_github.md"
|
||||
prompt_path.write_text("x {repos_json} {max_items}", encoding="utf-8")
|
||||
config = {
|
||||
"model": "x",
|
||||
"baseUrl": "http://x",
|
||||
"apiKeyName": "DEEPSEEK_API_KEY",
|
||||
"prompts": {"section_github": str(prompt_path)},
|
||||
"sections": {"github_trending": {"max_items": 3}},
|
||||
}
|
||||
with patch("llm.call_llm", new=AsyncMock(side_effect=RuntimeError("boom"))):
|
||||
md, err = await summarize_github_trending([{"full_name": "o/r"}], config)
|
||||
assert md == ""
|
||||
assert "boom" in err
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_select_ai_related_hn_parses_id_array(tmp_path):
|
||||
prompt_path = tmp_path / "select.md"
|
||||
prompt_path.write_text("k={k} candidates={candidates_json}", encoding="utf-8")
|
||||
config = {
|
||||
"model": "x",
|
||||
"baseUrl": "http://x",
|
||||
"apiKeyName": "DEEPSEEK_API_KEY",
|
||||
"prompts": {"section_hackernews_select": str(prompt_path)},
|
||||
}
|
||||
with patch("llm.call_llm", new=AsyncMock(return_value='["111", "222"]')):
|
||||
ids, err = await select_ai_related_hn(
|
||||
[{"id": "111"}, {"id": "222"}, {"id": "333"}], k=2, config=config
|
||||
)
|
||||
assert ids == ["111", "222"]
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_select_ai_related_hn_empty_array(tmp_path):
|
||||
prompt_path = tmp_path / "select.md"
|
||||
prompt_path.write_text("{k}{candidates_json}", encoding="utf-8")
|
||||
config = {
|
||||
"model": "x",
|
||||
"baseUrl": "http://x",
|
||||
"apiKeyName": "DEEPSEEK_API_KEY",
|
||||
"prompts": {"section_hackernews_select": str(prompt_path)},
|
||||
}
|
||||
with patch("llm.call_llm", new=AsyncMock(return_value="[]")):
|
||||
ids, err = await select_ai_related_hn([{"id": "1"}], k=1, config=config)
|
||||
assert ids == []
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_summarize_hackernews_happy(tmp_path):
|
||||
prompt_path = tmp_path / "hn.md"
|
||||
prompt_path.write_text("{stories_json}", encoding="utf-8")
|
||||
config = {
|
||||
"model": "x",
|
||||
"baseUrl": "http://x",
|
||||
"apiKeyName": "DEEPSEEK_API_KEY",
|
||||
"prompts": {"section_hackernews": str(prompt_path)},
|
||||
}
|
||||
with patch("llm.call_llm", new=AsyncMock(return_value="## HN summary")):
|
||||
md, err = await summarize_hackernews(
|
||||
[{"id": "1", "title": "t", "link_content": "x", "top_comments": []}], config
|
||||
)
|
||||
assert md == "## HN summary"
|
||||
assert err is None
|
||||
@@ -0,0 +1,285 @@
|
||||
"""主程序逻辑测试"""
|
||||
|
||||
import json
|
||||
import pytest
|
||||
import sys
|
||||
from datetime import datetime, timezone, timedelta, date
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, MagicMock, AsyncMock
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from main import (
|
||||
now_local,
|
||||
parse_time_to_local,
|
||||
calculate_push_times,
|
||||
collect_entries_for_push,
|
||||
main as run_main,
|
||||
)
|
||||
|
||||
|
||||
class TestNowLocal:
|
||||
"""测试获取本地时间"""
|
||||
|
||||
def test_now_local_with_config(self, sample_config):
|
||||
result = now_local(sample_config)
|
||||
assert isinstance(result, datetime)
|
||||
assert result.tzinfo is not None
|
||||
|
||||
def test_now_local_without_config(self):
|
||||
result = now_local()
|
||||
assert isinstance(result, datetime)
|
||||
|
||||
|
||||
class TestParseTimeToLocal:
|
||||
"""测试时间解析"""
|
||||
|
||||
def test_parse_iso_format(self, sample_config):
|
||||
result = parse_time_to_local("2024-01-15T10:30:00+00:00", sample_config)
|
||||
assert result is not None
|
||||
assert result.year == 2024
|
||||
assert result.month == 1
|
||||
assert result.day == 15
|
||||
|
||||
def test_parse_with_z_suffix(self, sample_config):
|
||||
result = parse_time_to_local("2024-01-15T10:30:00Z", sample_config)
|
||||
assert result is not None
|
||||
assert result.year == 2024
|
||||
|
||||
def test_parse_invalid_format(self, sample_config):
|
||||
result = parse_time_to_local("not-a-date", sample_config)
|
||||
assert result is None
|
||||
|
||||
def test_parse_none(self, sample_config):
|
||||
result = parse_time_to_local("", sample_config)
|
||||
assert result is None
|
||||
|
||||
|
||||
class TestCalculatePushTimes:
|
||||
"""测试推送时间计算"""
|
||||
|
||||
def test_calculate_single_cron(self, sample_config):
|
||||
times = calculate_push_times(["30 8 * * *"], config=sample_config)
|
||||
assert len(times) == 1
|
||||
assert times[0].hour == 8
|
||||
assert times[0].minute == 30
|
||||
|
||||
def test_calculate_multiple_crons(self, sample_config):
|
||||
times = calculate_push_times(["0 8 * * *", "0 17 * * *"], config=sample_config)
|
||||
assert len(times) == 2
|
||||
hours = [t.hour for t in times]
|
||||
assert 8 in hours
|
||||
assert 17 in hours
|
||||
|
||||
def test_calculate_with_offset(self, sample_config):
|
||||
times = calculate_push_times(["0 8 * * *"], offset_days=1, config=sample_config)
|
||||
assert len(times) == 1
|
||||
expected_date = (datetime.now(timezone.utc) + timedelta(days=1)).date()
|
||||
assert times[0].date() == expected_date
|
||||
|
||||
def test_calculate_invalid_cron(self, sample_config):
|
||||
times = calculate_push_times(["invalid cron"], config=sample_config)
|
||||
assert times == []
|
||||
|
||||
|
||||
class TestCollectEntriesForPush:
|
||||
"""测试收集推送条目"""
|
||||
|
||||
def test_collect_no_files(self, temp_dir):
|
||||
to_push, context = collect_entries_for_push(
|
||||
last_push_time=None, context_days=2, min_score=60, data_dir=str(temp_dir)
|
||||
)
|
||||
assert to_push == []
|
||||
assert context == []
|
||||
|
||||
def test_collect_with_low_score(self, temp_dir):
|
||||
now = datetime.now(timezone.utc)
|
||||
|
||||
data = {
|
||||
"meta": {"date": now.date().isoformat()},
|
||||
"entries": [
|
||||
{
|
||||
"title": "Low Score",
|
||||
"link": "https://example.com/1",
|
||||
"score": 30,
|
||||
"fetched_at": now.isoformat(),
|
||||
}
|
||||
],
|
||||
}
|
||||
|
||||
fetch_file = temp_dir / f"fetch-{now.date().isoformat()}.json"
|
||||
with open(fetch_file, "w") as f:
|
||||
json.dump(data, f)
|
||||
|
||||
to_push, context = collect_entries_for_push(
|
||||
last_push_time=None, context_days=2, min_score=60, data_dir=str(temp_dir)
|
||||
)
|
||||
|
||||
assert len(to_push) == 0
|
||||
|
||||
def test_collect_with_high_score(self, temp_dir):
|
||||
now = datetime.now(timezone.utc)
|
||||
|
||||
data = {
|
||||
"meta": {"date": now.date().isoformat()},
|
||||
"entries": [
|
||||
{
|
||||
"title": "High Score",
|
||||
"link": "https://example.com/1",
|
||||
"score": 85,
|
||||
"fetched_at": now.isoformat(),
|
||||
}
|
||||
],
|
||||
}
|
||||
|
||||
fetch_file = temp_dir / f"fetch-{now.date().isoformat()}.json"
|
||||
with open(fetch_file, "w") as f:
|
||||
json.dump(data, f)
|
||||
|
||||
to_push, context = collect_entries_for_push(
|
||||
last_push_time=None, context_days=2, min_score=60, data_dir=str(temp_dir)
|
||||
)
|
||||
|
||||
assert len(to_push) == 1
|
||||
assert to_push[0]["score"] == 85
|
||||
|
||||
def test_collect_with_last_push_time(self, temp_dir):
|
||||
now = datetime.now(timezone.utc)
|
||||
last_push = now - timedelta(hours=2)
|
||||
|
||||
data = {
|
||||
"meta": {"date": now.date().isoformat()},
|
||||
"entries": [
|
||||
{
|
||||
"title": "New Entry",
|
||||
"link": "https://example.com/1",
|
||||
"score": 80,
|
||||
"fetched_at": now.isoformat(),
|
||||
},
|
||||
{
|
||||
"title": "Old Entry",
|
||||
"link": "https://example.com/2",
|
||||
"score": 80,
|
||||
"fetched_at": last_push.isoformat(),
|
||||
},
|
||||
],
|
||||
}
|
||||
|
||||
fetch_file = temp_dir / f"fetch-{now.date().isoformat()}.json"
|
||||
with open(fetch_file, "w") as f:
|
||||
json.dump(data, f)
|
||||
|
||||
to_push, context = collect_entries_for_push(
|
||||
last_push_time=last_push,
|
||||
context_days=2,
|
||||
min_score=60,
|
||||
data_dir=str(temp_dir),
|
||||
)
|
||||
|
||||
assert len(to_push) == 1
|
||||
assert to_push[0]["title"] == "New Entry"
|
||||
|
||||
def test_collect_context_limit(self, temp_dir):
|
||||
now = datetime.now(timezone.utc)
|
||||
|
||||
entries = [
|
||||
{
|
||||
"title": f"Entry{i}",
|
||||
"link": f"https://example.com/{i}",
|
||||
"score": 50 + i,
|
||||
"fetched_at": now.isoformat(),
|
||||
}
|
||||
for i in range(60)
|
||||
]
|
||||
|
||||
data = {"meta": {"date": now.date().isoformat()}, "entries": entries}
|
||||
|
||||
fetch_file = temp_dir / f"fetch-{now.date().isoformat()}.json"
|
||||
with open(fetch_file, "w") as f:
|
||||
json.dump(data, f)
|
||||
|
||||
to_push, context = collect_entries_for_push(
|
||||
last_push_time=None, context_days=2, min_score=60, data_dir=str(temp_dir)
|
||||
)
|
||||
|
||||
assert len(context) <= 50
|
||||
|
||||
def test_collect_multi_day(self, temp_dir):
|
||||
from src.config import get_timezone
|
||||
|
||||
tz = get_timezone()
|
||||
today = datetime.now(tz)
|
||||
yesterday = today - timedelta(days=1)
|
||||
|
||||
today_data = {
|
||||
"meta": {"date": today.date().isoformat()},
|
||||
"entries": [
|
||||
{
|
||||
"title": "Today Entry",
|
||||
"link": "https://example.com/1",
|
||||
"score": 80,
|
||||
"fetched_at": today.isoformat(),
|
||||
}
|
||||
],
|
||||
}
|
||||
|
||||
yesterday_data = {
|
||||
"meta": {"date": yesterday.date().isoformat()},
|
||||
"entries": [
|
||||
{
|
||||
"title": "Yesterday Entry",
|
||||
"link": "https://example.com/2",
|
||||
"score": 75,
|
||||
"fetched_at": yesterday.isoformat(),
|
||||
}
|
||||
],
|
||||
}
|
||||
|
||||
(temp_dir / f"fetch-{today.date().isoformat()}.json").write_text(
|
||||
json.dumps(today_data)
|
||||
)
|
||||
(temp_dir / f"fetch-{yesterday.date().isoformat()}.json").write_text(
|
||||
json.dumps(yesterday_data)
|
||||
)
|
||||
|
||||
to_push, context = collect_entries_for_push(
|
||||
last_push_time=None, context_days=2, min_score=60, data_dir=str(temp_dir)
|
||||
)
|
||||
|
||||
assert len(to_push) >= 1
|
||||
|
||||
|
||||
class TestMainStartup:
|
||||
"""测试主程序启动流程"""
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_main_checks_llm_before_starting_loops(self, sample_config):
|
||||
with patch("main.load_config", return_value=sample_config), patch(
|
||||
"main.check_llm_available", new_callable=AsyncMock
|
||||
) as mock_check, patch(
|
||||
"main.fetch_loop", new_callable=AsyncMock
|
||||
) as mock_fetch_loop, patch(
|
||||
"main.push_loop", new_callable=AsyncMock
|
||||
) as mock_push_loop:
|
||||
await run_main()
|
||||
|
||||
mock_check.assert_awaited_once_with(sample_config["llm"])
|
||||
mock_fetch_loop.assert_awaited_once_with(sample_config)
|
||||
mock_push_loop.assert_awaited_once_with(sample_config)
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_main_exits_when_llm_health_check_fails(self, sample_config):
|
||||
with patch("main.load_config", return_value=sample_config), patch(
|
||||
"main.check_llm_available", new_callable=AsyncMock
|
||||
) as mock_check, patch(
|
||||
"main.fetch_loop", new_callable=AsyncMock
|
||||
) as mock_fetch_loop, patch(
|
||||
"main.push_loop", new_callable=AsyncMock
|
||||
) as mock_push_loop:
|
||||
mock_check.side_effect = RuntimeError("health failed")
|
||||
|
||||
await run_main()
|
||||
|
||||
mock_check.assert_awaited_once_with(sample_config["llm"])
|
||||
mock_fetch_loop.assert_not_called()
|
||||
mock_push_loop.assert_not_called()
|
||||
@@ -0,0 +1,123 @@
|
||||
"""测试早报四模块编排:gather + insights 串行 + sentinel 拼装 + 失败隔离"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.main import _run_morning_push
|
||||
|
||||
|
||||
def _insights_meta() -> dict:
|
||||
return {
|
||||
"title": "📰 AI Daily 每日精选 | 2026-05-22",
|
||||
"lead": "lead text",
|
||||
"highlights": ["a"],
|
||||
"profile": "morning",
|
||||
"date": "2026-05-22",
|
||||
"excerpt": "",
|
||||
"seotitle": "",
|
||||
"seodescription": "",
|
||||
}
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_assembles_all_four_sections(sample_config):
|
||||
sample_config["filter"]["push_context_days"] = 5
|
||||
|
||||
sent = {}
|
||||
|
||||
async def fake_send(content, push_cfg, title=None, metadata=None):
|
||||
sent["content"] = content
|
||||
sent["metadata"] = metadata
|
||||
|
||||
saved = {}
|
||||
|
||||
def fake_save(
|
||||
filepath,
|
||||
content,
|
||||
source_count,
|
||||
total_entries,
|
||||
profile="default",
|
||||
metadata=None,
|
||||
):
|
||||
saved["profile"] = profile
|
||||
saved["content"] = content
|
||||
saved["metadata"] = metadata
|
||||
|
||||
with patch(
|
||||
"src.main.run_rss_section", new=AsyncMock(return_value=("R", None, None))
|
||||
), patch(
|
||||
"src.main.run_github_section", new=AsyncMock(return_value=("G", None))
|
||||
), patch(
|
||||
"src.main.run_hackernews_section", new=AsyncMock(return_value=("H", None))
|
||||
), patch(
|
||||
"src.main.run_insights_section",
|
||||
new=AsyncMock(return_value=("I", _insights_meta(), None)),
|
||||
), patch(
|
||||
"src.main.send_to_platforms", new=AsyncMock(side_effect=fake_send)
|
||||
), patch(
|
||||
"src.main.save_push_file", side_effect=fake_save
|
||||
):
|
||||
await _run_morning_push(sample_config)
|
||||
|
||||
assert "SECTION:rss" in sent["content"]
|
||||
assert "SECTION:github" in sent["content"]
|
||||
assert "SECTION:hackernews" in sent["content"]
|
||||
assert "SECTION:insights" in sent["content"]
|
||||
assert sent["metadata"]["profile"] == "morning"
|
||||
assert saved["profile"] == "morning"
|
||||
assert saved["metadata"]["lead"] == "lead text"
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_rss_failure_raises_to_caller(sample_config):
|
||||
sample_config["filter"]["push_context_days"] = 5
|
||||
|
||||
with patch(
|
||||
"src.main.run_rss_section",
|
||||
new=AsyncMock(return_value=("", None, "compose_digest 失败")),
|
||||
), patch(
|
||||
"src.main.run_github_section", new=AsyncMock(return_value=("G", None))
|
||||
), patch(
|
||||
"src.main.run_hackernews_section", new=AsyncMock(return_value=("H", None))
|
||||
), patch(
|
||||
"src.main.notify_llm_errors", new=AsyncMock()
|
||||
):
|
||||
with pytest.raises(RuntimeError):
|
||||
await _run_morning_push(sample_config)
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_section_failure_degrades_to_omission(sample_config):
|
||||
sample_config["filter"]["push_context_days"] = 5
|
||||
|
||||
sent = {}
|
||||
|
||||
async def fake_send(content, push_cfg, title=None, metadata=None):
|
||||
sent["content"] = content
|
||||
|
||||
with patch(
|
||||
"src.main.run_rss_section", new=AsyncMock(return_value=("R", None, None))
|
||||
), patch(
|
||||
"src.main.run_github_section", new=AsyncMock(return_value=("", "gh down"))
|
||||
), patch(
|
||||
"src.main.run_hackernews_section", new=AsyncMock(return_value=("H", None))
|
||||
), patch(
|
||||
"src.main.run_insights_section",
|
||||
new=AsyncMock(return_value=("I", _insights_meta(), None)),
|
||||
), patch(
|
||||
"src.main.notify_llm_errors", new=AsyncMock()
|
||||
), patch(
|
||||
"src.main.send_to_platforms", new=AsyncMock(side_effect=fake_send)
|
||||
), patch(
|
||||
"src.main.save_push_file"
|
||||
):
|
||||
await _run_morning_push(sample_config)
|
||||
|
||||
assert "SECTION:rss" in sent["content"]
|
||||
assert "SECTION:github" not in sent["content"]
|
||||
assert "SECTION:hackernews" in sent["content"]
|
||||
@@ -0,0 +1,43 @@
|
||||
"""测试 run_push_job 的早报/默认路径分发"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.main import run_push_job
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_default_path_when_not_morning(sample_config):
|
||||
sample_config["schedule"]["push_cron"] = ["0 8 * * *", "0 17 * * *"]
|
||||
sample_config["filter"]["push_context_days"] = 5
|
||||
|
||||
with patch("src.main.is_morning_push", return_value=False), patch(
|
||||
"src.main._run_default_push", new=AsyncMock(return_value=None)
|
||||
) as default_path, patch(
|
||||
"src.main._run_morning_push", new=AsyncMock(return_value=None)
|
||||
) as morning_path:
|
||||
await run_push_job(sample_config)
|
||||
|
||||
default_path.assert_awaited_once()
|
||||
morning_path.assert_not_awaited()
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_morning_path_when_morning(sample_config):
|
||||
sample_config["schedule"]["push_cron"] = ["0 8 * * *", "0 17 * * *"]
|
||||
sample_config["filter"]["push_context_days"] = 5
|
||||
|
||||
with patch("src.main.is_morning_push", return_value=True), patch(
|
||||
"src.main._run_default_push", new=AsyncMock(return_value=None)
|
||||
) as default_path, patch(
|
||||
"src.main._run_morning_push", new=AsyncMock(return_value=None)
|
||||
) as morning_path:
|
||||
await run_push_job(sample_config)
|
||||
|
||||
morning_path.assert_awaited_once()
|
||||
default_path.assert_not_awaited()
|
||||
@@ -0,0 +1,60 @@
|
||||
"""测试早报判定:push_cron 最近最早匹配"""
|
||||
|
||||
import sys
|
||||
from datetime import datetime, timedelta, timezone
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.main import is_morning_push
|
||||
|
||||
|
||||
TZ = timezone(timedelta(hours=8))
|
||||
|
||||
|
||||
def _cfg(push_cron):
|
||||
return {"schedule": {"push_cron": push_cron, "timezone_hours": 8}}
|
||||
|
||||
|
||||
def test_returns_false_when_push_cron_empty():
|
||||
assert is_morning_push(datetime(2026, 5, 17, 8, 0, tzinfo=TZ), {"schedule": {}}) is False
|
||||
assert is_morning_push(datetime(2026, 5, 17, 8, 0, tzinfo=TZ), _cfg([])) is False
|
||||
|
||||
|
||||
def test_single_cron_always_morning():
|
||||
"""push_cron 只有一项时,任何触发(含手动 / 非整点)都视为早报"""
|
||||
cfg = _cfg(["0 8 * * *"])
|
||||
assert is_morning_push(datetime(2026, 5, 17, 8, 0, tzinfo=TZ), cfg) is True
|
||||
assert is_morning_push(datetime(2026, 5, 17, 12, 30, tzinfo=TZ), cfg) is True
|
||||
assert is_morning_push(datetime(2026, 5, 17, 23, 59, tzinfo=TZ), cfg) is True
|
||||
|
||||
|
||||
def test_earliest_cron_match_is_morning():
|
||||
"""触发时刻离最早 cron 最近 → 早报"""
|
||||
cfg = _cfg(["0 8 * * *", "0 17 * * *"])
|
||||
assert is_morning_push(datetime(2026, 5, 17, 8, 0, tzinfo=TZ), cfg) is True
|
||||
assert is_morning_push(datetime(2026, 5, 17, 8, 30, tzinfo=TZ), cfg) is True
|
||||
assert is_morning_push(datetime(2026, 5, 17, 6, 0, tzinfo=TZ), cfg) is True
|
||||
|
||||
|
||||
def test_later_cron_match_not_morning():
|
||||
"""触发时刻离非最早 cron 最近 → 默认"""
|
||||
cfg = _cfg(["0 8 * * *", "0 17 * * *"])
|
||||
assert is_morning_push(datetime(2026, 5, 17, 17, 0, tzinfo=TZ), cfg) is False
|
||||
assert is_morning_push(datetime(2026, 5, 17, 16, 30, tzinfo=TZ), cfg) is False
|
||||
assert is_morning_push(datetime(2026, 5, 17, 22, 0, tzinfo=TZ), cfg) is False
|
||||
|
||||
|
||||
def test_drift_tolerance_via_closest_match():
|
||||
"""无显式容差,但「最近匹配」自动吸附小幅漂移 (08:01 仍归 08:00)"""
|
||||
cfg = _cfg(["0 8 * * *", "0 17 * * *"])
|
||||
assert is_morning_push(datetime(2026, 5, 17, 8, 1, tzinfo=TZ), cfg) is True
|
||||
assert is_morning_push(datetime(2026, 5, 17, 17, 1, tzinfo=TZ), cfg) is False
|
||||
|
||||
|
||||
def test_three_crons_only_earliest_is_morning():
|
||||
"""三条 cron 时,只有最早那条对应的触发是早报"""
|
||||
cfg = _cfg(["0 8 * * *", "0 12 * * *", "0 20 * * *"])
|
||||
assert is_morning_push(datetime(2026, 5, 17, 8, 0, tzinfo=TZ), cfg) is True
|
||||
assert is_morning_push(datetime(2026, 5, 17, 12, 0, tzinfo=TZ), cfg) is False
|
||||
assert is_morning_push(datetime(2026, 5, 17, 20, 0, tzinfo=TZ), cfg) is False
|
||||
@@ -0,0 +1,87 @@
|
||||
"""内容处理模块测试"""
|
||||
|
||||
import pytest
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from processor import html_to_markdown
|
||||
|
||||
|
||||
class TestHtmlToMarkdown:
|
||||
"""测试HTML转Markdown"""
|
||||
|
||||
def test_convert_basic_html(self):
|
||||
html = "<p>Hello World</p>"
|
||||
result = html_to_markdown(html)
|
||||
assert "Hello World" in result
|
||||
|
||||
def test_convert_with_links(self):
|
||||
html = '<a href="https://example.com">Click here</a>'
|
||||
result = html_to_markdown(html)
|
||||
assert "[Click here](https://example.com)" in result
|
||||
|
||||
def test_convert_with_images(self):
|
||||
html = '<img src="https://example.com/image.png" alt="Image">'
|
||||
result = html_to_markdown(html)
|
||||
assert "" in result
|
||||
|
||||
def test_convert_with_headings(self):
|
||||
html = "<h1>Title</h1><h2>Subtitle</h2>"
|
||||
result = html_to_markdown(html)
|
||||
assert "# Title" in result
|
||||
assert "## Subtitle" in result
|
||||
|
||||
def test_convert_with_lists(self):
|
||||
html = "<ul><li>Item 1</li><li>Item 2</li></ul>"
|
||||
result = html_to_markdown(html)
|
||||
assert "Item 1" in result
|
||||
assert "Item 2" in result
|
||||
|
||||
def test_convert_with_strong_emphasis(self):
|
||||
html = "<strong>Bold</strong> and <em>italic</em>"
|
||||
result = html_to_markdown(html)
|
||||
assert "**Bold**" in result
|
||||
assert "*italic*" in result
|
||||
|
||||
def test_relative_link_conversion(self):
|
||||
html = '<a href="/article/123">Read more</a>'
|
||||
result = html_to_markdown(html, base_url="https://example.com/blog")
|
||||
assert "https://example.com/article/123" in result
|
||||
|
||||
def test_relative_image_conversion(self):
|
||||
html = '<img src="/images/logo.png">'
|
||||
result = html_to_markdown(html, base_url="https://example.com")
|
||||
assert "https://example.com/images/logo.png" in result
|
||||
|
||||
def test_absolute_link_unchanged(self):
|
||||
html = '<a href="https://other.com/page">Link</a>'
|
||||
result = html_to_markdown(html, base_url="https://example.com")
|
||||
assert "https://other.com/page" in result
|
||||
|
||||
def test_remove_xgo_ing_link(self):
|
||||
html = "<p>Content</p><p>[⚡ Powered by xgo.ing](https://xgo.ing)</p>"
|
||||
result = html_to_markdown(html)
|
||||
assert "xgo.ing" not in result
|
||||
assert "Content" in result
|
||||
|
||||
def test_remove_xgo_ing_link_with_slash(self):
|
||||
html = "<p>Content</p><p>[⚡ Powered by xgo.ing](https://xgo.ing/)</p>"
|
||||
result = html_to_markdown(html)
|
||||
assert "xgo.ing" not in result
|
||||
|
||||
def test_clean_extra_newlines(self):
|
||||
html = "<p>Line 1</p>\n\n\n\n<p>Line 2</p>"
|
||||
result = html_to_markdown(html)
|
||||
assert "\n\n\n\n" not in result
|
||||
|
||||
def test_empty_html(self):
|
||||
result = html_to_markdown("")
|
||||
assert result.strip() == ""
|
||||
|
||||
def test_html_with_nbsp(self):
|
||||
html = "<p>Hello World</p>"
|
||||
result = html_to_markdown(html)
|
||||
assert "Hello" in result
|
||||
assert "World" in result
|
||||
@@ -0,0 +1,209 @@
|
||||
"""推送模块测试"""
|
||||
|
||||
import os
|
||||
import pytest
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock, MagicMock
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from push.discord import DiscordPlatform
|
||||
from push.feishu import FeishuPlatform
|
||||
from push import create_platform
|
||||
|
||||
|
||||
class TestDiscordPlatform:
|
||||
"""测试Discord推送"""
|
||||
|
||||
def test_validate_config_valid(self):
|
||||
config = {
|
||||
"enabled": True,
|
||||
"apiKeyName": "DISCORD_WEBHOOK_URL",
|
||||
}
|
||||
with patch.dict(
|
||||
os.environ,
|
||||
{"DISCORD_WEBHOOK_URL": "https://discord.com/api/webhooks/123456/abcdef"},
|
||||
):
|
||||
platform = DiscordPlatform(config)
|
||||
assert platform.validate_config(config) is True
|
||||
|
||||
def test_validate_config_disabled(self):
|
||||
config = {
|
||||
"enabled": False,
|
||||
"apiKeyName": "DISCORD_WEBHOOK_URL",
|
||||
}
|
||||
with patch.dict(
|
||||
os.environ,
|
||||
{"DISCORD_WEBHOOK_URL": "https://discord.com/api/webhooks/123456/abcdef"},
|
||||
):
|
||||
platform = DiscordPlatform(config)
|
||||
assert platform.validate_config(config) is False
|
||||
|
||||
def test_validate_config_missing_webhook(self):
|
||||
config = {"enabled": True, "apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
with patch.dict(os.environ, {"DISCORD_WEBHOOK_URL": ""}):
|
||||
platform = DiscordPlatform(config)
|
||||
assert platform.validate_config(config) is False
|
||||
|
||||
def test_validate_config_invalid_url(self):
|
||||
config = {"enabled": True, "apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
with patch.dict(os.environ, {"DISCORD_WEBHOOK_URL": "not-a-valid-url"}):
|
||||
platform = DiscordPlatform(config)
|
||||
assert platform.validate_config(config) is False
|
||||
|
||||
def test_validate_config_wrong_domain(self):
|
||||
config = {"enabled": True, "apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
with patch.dict(
|
||||
os.environ, {"DISCORD_WEBHOOK_URL": "https://example.com/webhook"}
|
||||
):
|
||||
platform = DiscordPlatform(config)
|
||||
assert platform.validate_config(config) is False
|
||||
|
||||
def test_split_content_short(self):
|
||||
with patch.dict(os.environ, {"DISCORD_WEBHOOK_URL": "https://test.com"}):
|
||||
config = {"apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
platform = DiscordPlatform(config)
|
||||
short_content = "Hello"
|
||||
chunks = platform._split_content(short_content, limit=2000)
|
||||
assert len(chunks) == 1
|
||||
assert chunks[0] == "Hello"
|
||||
|
||||
def test_split_content_long_message(self):
|
||||
with patch.dict(os.environ, {"DISCORD_WEBHOOK_URL": "https://test.com"}):
|
||||
config = {"apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
platform = DiscordPlatform(config)
|
||||
long_content = "A\n" * 2500
|
||||
chunks = platform._split_content(long_content, limit=2000)
|
||||
assert len(chunks) > 1
|
||||
assert all(len(c) <= 2000 for c in chunks)
|
||||
|
||||
def test_split_content_exact_boundary(self):
|
||||
with patch.dict(os.environ, {"DISCORD_WEBHOOK_URL": "https://test.com"}):
|
||||
config = {"apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
platform = DiscordPlatform(config)
|
||||
content = "A" * 2000
|
||||
chunks = platform._split_content(content, limit=2000)
|
||||
assert len(chunks) == 1
|
||||
|
||||
def test_split_content_unicode(self):
|
||||
with patch.dict(os.environ, {"DISCORD_WEBHOOK_URL": "https://test.com"}):
|
||||
config = {"apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
platform = DiscordPlatform(config)
|
||||
content = "你好" * 500
|
||||
chunks = platform._split_content(content, limit=100)
|
||||
assert len(chunks) >= 1
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_send_success(self):
|
||||
with patch.dict(
|
||||
os.environ,
|
||||
{"DISCORD_WEBHOOK_URL": "https://discord.com/api/webhooks/test/abc"},
|
||||
):
|
||||
config = {"apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
platform = DiscordPlatform(config)
|
||||
|
||||
with patch.object(platform, "send", new_callable=AsyncMock) as mock_send:
|
||||
mock_send.return_value = True
|
||||
|
||||
result = await mock_send("Test message")
|
||||
|
||||
assert result is True
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_send_failure(self):
|
||||
with patch.dict(
|
||||
os.environ,
|
||||
{"DISCORD_WEBHOOK_URL": "https://discord.com/api/webhooks/test/abc"},
|
||||
):
|
||||
config = {"apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
platform = DiscordPlatform(config)
|
||||
|
||||
with patch.object(platform, "send", new_callable=AsyncMock) as mock_send:
|
||||
mock_send.return_value = False
|
||||
|
||||
result = await mock_send("Test message")
|
||||
|
||||
assert result is False
|
||||
|
||||
|
||||
class TestFeishuPlatform:
|
||||
"""测试飞书推送"""
|
||||
|
||||
def test_validate_config_valid(self):
|
||||
config = {
|
||||
"enabled": True,
|
||||
"apiKeyName": "FEISHU_WEBHOOK_URL",
|
||||
}
|
||||
with patch.dict(os.environ, {"FEISHU_WEBHOOK_URL": "https://open.feishu.cn/open-apis/bot/v2/hook/test"}):
|
||||
platform = FeishuPlatform(config)
|
||||
assert platform.validate_config(config) is True
|
||||
|
||||
def test_validate_config_disabled(self):
|
||||
config = {
|
||||
"enabled": False,
|
||||
"apiKeyName": "FEISHU_WEBHOOK_URL",
|
||||
}
|
||||
with patch.dict(os.environ, {"FEISHU_WEBHOOK_URL": "https://open.feishu.cn/open-apis/bot/v2/hook/test"}):
|
||||
platform = FeishuPlatform(config)
|
||||
assert platform.validate_config(config) is False
|
||||
|
||||
def test_validate_config_missing_key(self):
|
||||
config = {"enabled": True, "apiKeyName": "FEISHU_WEBHOOK_URL"}
|
||||
with patch.dict(os.environ, {"FEISHU_WEBHOOK_URL": ""}):
|
||||
platform = FeishuPlatform(config)
|
||||
assert platform.validate_config(config) is False
|
||||
|
||||
def test_validate_config_any_non_empty_webhook(self):
|
||||
config = {"enabled": True, "apiKeyName": "FEISHU_WEBHOOK_URL"}
|
||||
with patch.dict(os.environ, {"FEISHU_WEBHOOK_URL": "https://open.feishu.cn/open-apis/bot/v2/hook/test"}):
|
||||
platform = FeishuPlatform(config)
|
||||
assert platform.validate_config(config) is True
|
||||
|
||||
|
||||
class TestPushFactory:
|
||||
"""测试平台工厂"""
|
||||
|
||||
def test_create_enabled_platform(self):
|
||||
config = {
|
||||
"enabled": True,
|
||||
"apiKeyName": "DISCORD_WEBHOOK_URL",
|
||||
}
|
||||
with patch.dict(
|
||||
os.environ,
|
||||
{"DISCORD_WEBHOOK_URL": "https://discord.com/api/webhooks/123/abc"},
|
||||
):
|
||||
platform = create_platform("discord", config)
|
||||
assert platform is not None
|
||||
|
||||
def test_create_disabled_platform_returns_none(self):
|
||||
config = {"enabled": False, "apiKeyName": "DISCORD_WEBHOOK_URL"}
|
||||
platform = create_platform("discord", config)
|
||||
assert platform is None
|
||||
|
||||
def test_create_unknown_platform_raises(self):
|
||||
with pytest.raises(ValueError):
|
||||
create_platform("unknown", {})
|
||||
|
||||
def test_create_feishu_platform(self):
|
||||
config = {
|
||||
"enabled": True,
|
||||
"apiKeyName": "FEISHU_WEBHOOK_URL",
|
||||
}
|
||||
with patch.dict(os.environ, {"FEISHU_WEBHOOK_URL": "https://open.feishu.cn/open-apis/bot/v2/hook/test"}):
|
||||
platform = create_platform("feishu", config)
|
||||
assert platform is not None
|
||||
assert isinstance(platform, FeishuPlatform)
|
||||
|
||||
def test_create_discord_platform(self):
|
||||
config = {
|
||||
"enabled": True,
|
||||
"apiKeyName": "DISCORD_WEBHOOK_URL",
|
||||
}
|
||||
with patch.dict(
|
||||
os.environ,
|
||||
{"DISCORD_WEBHOOK_URL": "https://discord.com/api/webhooks/test/abc"},
|
||||
):
|
||||
platform = create_platform("discord", config)
|
||||
assert platform is not None
|
||||
assert isinstance(platform, DiscordPlatform)
|
||||
@@ -0,0 +1,107 @@
|
||||
"""测试 GitHub REST API enrich 字段映射、archived 过滤、token 鉴权头"""
|
||||
|
||||
import base64
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock, MagicMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.sections.github.repo_enricher import enrich_repo, _auth_headers
|
||||
|
||||
|
||||
def test_auth_headers_with_token(monkeypatch):
|
||||
monkeypatch.setenv("GITHUB_TOKEN", "ghp_secret")
|
||||
headers = _auth_headers(token_env="GITHUB_TOKEN")
|
||||
assert headers["Authorization"] == "Bearer ghp_secret"
|
||||
assert headers["Accept"] == "application/vnd.github+json"
|
||||
|
||||
|
||||
def test_auth_headers_without_token(monkeypatch):
|
||||
monkeypatch.delenv("GITHUB_TOKEN", raising=False)
|
||||
headers = _auth_headers(token_env="GITHUB_TOKEN")
|
||||
assert "Authorization" not in headers
|
||||
assert headers["Accept"] == "application/vnd.github+json"
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_enrich_repo_merges_metadata_and_readme():
|
||||
readme_body = "# Title\n\nProject description here."
|
||||
readme_b64 = base64.b64encode(readme_body.encode("utf-8")).decode("ascii")
|
||||
|
||||
metadata_payload = {
|
||||
"description": "real desc",
|
||||
"topics": ["llm", "rag"],
|
||||
"license": {"spdx_id": "MIT"},
|
||||
"pushed_at": "2026-05-16T10:00:00Z",
|
||||
"archived": False,
|
||||
}
|
||||
readme_payload = {"content": readme_b64, "encoding": "base64"}
|
||||
|
||||
async def fake_get_json(session, url, **kwargs):
|
||||
if url.endswith("/readme"):
|
||||
return readme_payload
|
||||
return metadata_payload
|
||||
|
||||
base = {
|
||||
"url": "https://github.com/o/r",
|
||||
"full_name": "o/r",
|
||||
"description": "from trending",
|
||||
"language": "Python",
|
||||
"stars_today": 100,
|
||||
"stars_total": 5000,
|
||||
}
|
||||
|
||||
with patch(
|
||||
"src.sections.github.repo_enricher._get_json", new=AsyncMock(side_effect=fake_get_json)
|
||||
):
|
||||
enriched = await enrich_repo(
|
||||
session=MagicMock(), repo=base, token_env="GITHUB_TOKEN", readme_max_chars=200
|
||||
)
|
||||
|
||||
assert enriched["topics"] == ["llm", "rag"]
|
||||
assert enriched["license"] == "MIT"
|
||||
assert enriched["pushed_at"] == "2026-05-16T10:00:00Z"
|
||||
assert "Project description" in enriched["readme_excerpt"]
|
||||
assert enriched["stars_today"] == 100 # trending 已有字段保留
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_enrich_repo_returns_none_when_archived():
|
||||
metadata_payload = {"archived": True, "topics": [], "pushed_at": "x"}
|
||||
|
||||
async def fake_get_json(session, url, **kwargs):
|
||||
if url.endswith("/readme"):
|
||||
return {"content": ""}
|
||||
return metadata_payload
|
||||
|
||||
base = {"url": "https://github.com/o/r", "full_name": "o/r"}
|
||||
with patch(
|
||||
"src.sections.github.repo_enricher._get_json", new=AsyncMock(side_effect=fake_get_json)
|
||||
):
|
||||
result = await enrich_repo(
|
||||
session=MagicMock(), repo=base, token_env="GITHUB_TOKEN", readme_max_chars=200
|
||||
)
|
||||
assert result is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_enrich_repo_truncates_readme():
|
||||
readme_body = "x" * 5000
|
||||
readme_b64 = base64.b64encode(readme_body.encode("utf-8")).decode("ascii")
|
||||
|
||||
async def fake_get_json(session, url, **kwargs):
|
||||
if url.endswith("/readme"):
|
||||
return {"content": readme_b64, "encoding": "base64"}
|
||||
return {"archived": False, "topics": [], "pushed_at": "p"}
|
||||
|
||||
base = {"url": "https://github.com/o/r", "full_name": "o/r"}
|
||||
with patch(
|
||||
"src.sections.github.repo_enricher._get_json", new=AsyncMock(side_effect=fake_get_json)
|
||||
):
|
||||
enriched = await enrich_repo(
|
||||
session=MagicMock(), repo=base, token_env="GITHUB_TOKEN", readme_max_chars=100
|
||||
)
|
||||
assert len(enriched["readme_excerpt"]) == 100
|
||||
@@ -0,0 +1,40 @@
|
||||
"""测试 GitHub trending HTML 解析"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.sections.github.trending_scraper import parse_trending_html
|
||||
|
||||
|
||||
def test_parse_trending_html_returns_repo_dicts():
|
||||
fixture = (
|
||||
Path(__file__).parent / "fixtures" / "github_trending.html"
|
||||
).read_text(encoding="utf-8")
|
||||
|
||||
repos = parse_trending_html(fixture)
|
||||
|
||||
assert len(repos) > 0
|
||||
first = repos[0]
|
||||
assert first["url"].startswith("https://github.com/")
|
||||
assert "/" in first["full_name"]
|
||||
assert isinstance(first["stars_today"], int)
|
||||
assert isinstance(first["stars_total"], int)
|
||||
# description / language 可为空字符串但必须是 str
|
||||
assert isinstance(first["description"], str)
|
||||
assert isinstance(first["language"], str)
|
||||
|
||||
|
||||
def test_parse_trending_html_dedupes_by_url():
|
||||
fixture = (
|
||||
Path(__file__).parent / "fixtures" / "github_trending.html"
|
||||
).read_text(encoding="utf-8")
|
||||
repos = parse_trending_html(fixture)
|
||||
urls = [r["url"] for r in repos]
|
||||
assert len(urls) == len(set(urls))
|
||||
|
||||
|
||||
def test_parse_trending_html_empty_input():
|
||||
assert parse_trending_html("") == []
|
||||
assert parse_trending_html("<html><body>no repos</body></html>") == []
|
||||
@@ -0,0 +1,145 @@
|
||||
"""测试 GitHub 板块编排:抓取 → history 过滤 → enrich → LLM 总结"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.sections.github.section import run_github_section
|
||||
|
||||
# Stub summarize_github_trending until Task 11 provides it
|
||||
import src.llm as _llm
|
||||
if not hasattr(_llm, "summarize_github_trending"):
|
||||
async def _stub(*a, **k):
|
||||
return "", None
|
||||
_llm.summarize_github_trending = _stub
|
||||
|
||||
|
||||
def _cfg(history_file: str, max_deep_dive: int = 10) -> dict:
|
||||
return {
|
||||
"filter": {"keep_days": 7},
|
||||
"sections": {
|
||||
"github_trending": {
|
||||
"enabled": True,
|
||||
"max_items": 3,
|
||||
"max_deep_dive": max_deep_dive,
|
||||
"readme_max_chars": 3000,
|
||||
"history_file": history_file,
|
||||
"request_timeout": 10,
|
||||
"tokenName": "GITHUB_TOKEN",
|
||||
}
|
||||
},
|
||||
"llm": {
|
||||
"model": "x",
|
||||
"baseUrl": "http://x",
|
||||
"apiKeyName": "DEEPSEEK_API_KEY",
|
||||
"prompts": {"section_github": "prompts/section_github.md"},
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_disabled_returns_empty(tmp_path):
|
||||
cfg = _cfg(str(tmp_path / "h.json"))
|
||||
cfg["sections"]["github_trending"]["enabled"] = False
|
||||
md, err = await run_github_section(cfg, now=None)
|
||||
assert md == ""
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_no_candidates_after_history_returns_empty(tmp_path):
|
||||
history_path = tmp_path / "h.json"
|
||||
# 预置 history,使得今日 scrape 出来的 repo 都已存在
|
||||
history_path.write_text(
|
||||
'{"repos": {"https://github.com/a/b": "2026-05-16"}, "updated_at": "x"}',
|
||||
encoding="utf-8",
|
||||
)
|
||||
cfg = _cfg(str(history_path))
|
||||
|
||||
with patch(
|
||||
"src.sections.github.section.fetch_trending_page", new=AsyncMock(return_value="<html>")
|
||||
), patch(
|
||||
"src.sections.github.section.parse_trending_html",
|
||||
return_value=[{"url": "https://github.com/a/b", "full_name": "a/b"}],
|
||||
):
|
||||
md, err = await run_github_section(cfg, now=None)
|
||||
|
||||
assert md == ""
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_happy_path_enriches_and_summarizes(tmp_path):
|
||||
history_path = tmp_path / "h.json"
|
||||
cfg = _cfg(str(history_path), max_deep_dive=10)
|
||||
|
||||
repos = [
|
||||
{
|
||||
"url": "https://github.com/o1/r1",
|
||||
"full_name": "o1/r1",
|
||||
"description": "d1",
|
||||
"language": "Python",
|
||||
"stars_today": 100,
|
||||
"stars_total": 1000,
|
||||
}
|
||||
]
|
||||
enriched = [{**repos[0], "topics": ["llm"], "license": "MIT", "pushed_at": "p", "readme_excerpt": "rm"}]
|
||||
|
||||
with patch(
|
||||
"src.sections.github.section.fetch_trending_page", new=AsyncMock(return_value="<html>")
|
||||
), patch(
|
||||
"src.sections.github.section.parse_trending_html", return_value=repos
|
||||
), patch(
|
||||
"src.sections.github.section.enrich_repos",
|
||||
new=AsyncMock(return_value=(enriched, [])),
|
||||
), patch(
|
||||
"src.llm.summarize_github_trending",
|
||||
new=AsyncMock(return_value=("## GH section md", None)),
|
||||
):
|
||||
md, err = await run_github_section(cfg, now=None)
|
||||
|
||||
assert md == "## GH section md"
|
||||
assert err is None
|
||||
import json as _j
|
||||
saved = _j.loads(history_path.read_text(encoding="utf-8"))
|
||||
assert "https://github.com/o1/r1" in saved["repos"]
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_truncates_candidates_to_max_deep_dive(tmp_path):
|
||||
cfg = _cfg(str(tmp_path / "h.json"), max_deep_dive=2)
|
||||
repos = [
|
||||
{"url": f"https://github.com/o/r{i}", "full_name": f"o/r{i}"} for i in range(5)
|
||||
]
|
||||
captured = {}
|
||||
|
||||
async def fake_enrich(candidates, **kwargs):
|
||||
captured["count"] = len(candidates)
|
||||
return [], []
|
||||
|
||||
with patch(
|
||||
"src.sections.github.section.fetch_trending_page", new=AsyncMock(return_value="<html>")
|
||||
), patch(
|
||||
"src.sections.github.section.parse_trending_html", return_value=repos
|
||||
), patch(
|
||||
"src.sections.github.section.enrich_repos", new=AsyncMock(side_effect=fake_enrich)
|
||||
):
|
||||
await run_github_section(cfg, now=None)
|
||||
|
||||
assert captured["count"] == 2
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_scrape_failure_returns_error(tmp_path):
|
||||
cfg = _cfg(str(tmp_path / "h.json"))
|
||||
with patch(
|
||||
"src.sections.github.section.fetch_trending_page",
|
||||
new=AsyncMock(side_effect=RuntimeError("HTTP 500")),
|
||||
):
|
||||
md, err = await run_github_section(cfg, now=None)
|
||||
assert md == ""
|
||||
assert "HTTP 500" in err
|
||||
@@ -0,0 +1,246 @@
|
||||
"""测试 HN enrich(Algolia 评论树 + 外链正文)"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock, MagicMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.sections.hackernews.item_enricher import enrich_story
|
||||
|
||||
|
||||
def _kwargs(**overrides):
|
||||
base = dict(
|
||||
top_comments=3,
|
||||
top_l2_per_l1=2,
|
||||
comment_max_chars=500,
|
||||
comments_total_chars=60000,
|
||||
link_content_max_chars=3000,
|
||||
algolia_base="https://hn.algolia.com/api/v1",
|
||||
timeout=10,
|
||||
)
|
||||
base.update(overrides)
|
||||
return base
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_enrich_external_link_story_returns_tree():
|
||||
story = {
|
||||
"id": "111",
|
||||
"title": "T",
|
||||
"url": "https://example.com/post",
|
||||
"site": "example.com",
|
||||
"points": 100,
|
||||
"comments": 5,
|
||||
"comments_url": "https://news.ycombinator.com/item?id=111",
|
||||
}
|
||||
algolia_payload = {
|
||||
"text": None,
|
||||
"children": [
|
||||
{
|
||||
"text": "<p>comment one</p>",
|
||||
"children": [
|
||||
{"text": "<p>reply 1a</p>"},
|
||||
{"text": "<p>reply 1b</p>"},
|
||||
{"text": "<p>reply 1c (should be dropped)</p>"},
|
||||
],
|
||||
},
|
||||
{"text": "<p>comment two</p>", "children": []},
|
||||
{"text": "<p>comment three</p>"},
|
||||
{"text": "<p>comment four (over top_comments cap)</p>"},
|
||||
],
|
||||
}
|
||||
|
||||
async def fake_algolia(session, item_id, **kw):
|
||||
return algolia_payload
|
||||
|
||||
async def fake_external(session, url, **kw):
|
||||
return "link body"
|
||||
|
||||
with patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_algolia_item",
|
||||
new=AsyncMock(side_effect=fake_algolia),
|
||||
), patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_external_markdown",
|
||||
new=AsyncMock(side_effect=fake_external),
|
||||
):
|
||||
enriched = await enrich_story(
|
||||
session=MagicMock(), story=story, **_kwargs()
|
||||
)
|
||||
|
||||
tree = enriched["top_comments"]
|
||||
assert len(tree) == 3
|
||||
assert "comment one" in tree[0]["l1"]
|
||||
assert len(tree[0]["replies"]) == 2
|
||||
assert "reply 1a" in tree[0]["replies"][0]
|
||||
assert "reply 1b" in tree[0]["replies"][1]
|
||||
assert tree[1]["replies"] == []
|
||||
assert tree[2]["replies"] == []
|
||||
assert "link body" in enriched["link_content"]
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_enrich_show_hn_uses_root_text_no_external_fetch():
|
||||
story = {
|
||||
"id": "222",
|
||||
"title": "Show HN: T",
|
||||
"url": "https://news.ycombinator.com/item?id=222",
|
||||
"site": "",
|
||||
"points": 200,
|
||||
"comments": 10,
|
||||
"comments_url": "https://news.ycombinator.com/item?id=222",
|
||||
}
|
||||
algolia_payload = {
|
||||
"text": "<p>post body text</p>",
|
||||
"children": [{"text": "<p>c1</p>"}],
|
||||
}
|
||||
link_calls = []
|
||||
|
||||
async def fake_algolia(session, item_id, **kw):
|
||||
return algolia_payload
|
||||
|
||||
async def fake_external(session, url, **kw):
|
||||
link_calls.append(url)
|
||||
return "should not be called"
|
||||
|
||||
with patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_algolia_item",
|
||||
new=AsyncMock(side_effect=fake_algolia),
|
||||
), patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_external_markdown",
|
||||
new=AsyncMock(side_effect=fake_external),
|
||||
):
|
||||
enriched = await enrich_story(
|
||||
session=MagicMock(), story=story, **_kwargs()
|
||||
)
|
||||
|
||||
assert link_calls == []
|
||||
assert "post body text" in enriched["link_content"]
|
||||
assert enriched["top_comments"][0]["l1"].startswith("c1") or "c1" in enriched["top_comments"][0]["l1"]
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_enrich_truncates_comments_and_link():
|
||||
story = {
|
||||
"id": "333",
|
||||
"title": "T",
|
||||
"url": "https://example.com/a",
|
||||
"site": "example.com",
|
||||
"points": 100,
|
||||
"comments": 2,
|
||||
"comments_url": "x",
|
||||
}
|
||||
long_comment = "<p>" + ("y" * 2000) + "</p>"
|
||||
long_reply = "<p>" + ("z" * 2000) + "</p>"
|
||||
long_link = "z" * 5000
|
||||
|
||||
async def fake_algolia(session, item_id, **kw):
|
||||
return {
|
||||
"text": None,
|
||||
"children": [
|
||||
{"text": long_comment, "children": [{"text": long_reply}]}
|
||||
],
|
||||
}
|
||||
|
||||
async def fake_external(session, url, **kw):
|
||||
return long_link
|
||||
|
||||
with patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_algolia_item",
|
||||
new=AsyncMock(side_effect=fake_algolia),
|
||||
), patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_external_markdown",
|
||||
new=AsyncMock(side_effect=fake_external),
|
||||
):
|
||||
enriched = await enrich_story(
|
||||
session=MagicMock(),
|
||||
story=story,
|
||||
**_kwargs(comment_max_chars=100, link_content_max_chars=200),
|
||||
)
|
||||
|
||||
assert len(enriched["top_comments"][0]["l1"]) <= 100
|
||||
assert len(enriched["top_comments"][0]["replies"][0]) <= 100
|
||||
assert len(enriched["link_content"]) <= 200
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_enrich_failure_returns_partial():
|
||||
story = {
|
||||
"id": "444",
|
||||
"title": "T",
|
||||
"url": "https://example.com/x",
|
||||
"site": "example.com",
|
||||
"points": 100,
|
||||
"comments": 2,
|
||||
"comments_url": "x",
|
||||
}
|
||||
|
||||
async def fake_algolia(session, item_id, **kw):
|
||||
raise RuntimeError("algolia down")
|
||||
|
||||
async def fake_external(session, url, **kw):
|
||||
return "ok"
|
||||
|
||||
with patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_algolia_item",
|
||||
new=AsyncMock(side_effect=fake_algolia),
|
||||
), patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_external_markdown",
|
||||
new=AsyncMock(side_effect=fake_external),
|
||||
):
|
||||
enriched = await enrich_story(
|
||||
session=MagicMock(), story=story, **_kwargs()
|
||||
)
|
||||
|
||||
assert enriched["top_comments"] == []
|
||||
assert "ok" in enriched["link_content"]
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_enrich_total_budget_stops_early():
|
||||
"""累计字符达 comments_total_chars 立即停止,后续 L1 / L2 都不再加入"""
|
||||
story = {
|
||||
"id": "555",
|
||||
"title": "T",
|
||||
"url": "https://example.com/q",
|
||||
"site": "example.com",
|
||||
"points": 100,
|
||||
"comments": 5,
|
||||
"comments_url": "x",
|
||||
}
|
||||
big = "<p>" + ("a" * 500) + "</p>" # markdown 约 500 chars
|
||||
|
||||
async def fake_algolia(session, item_id, **kw):
|
||||
return {
|
||||
"text": None,
|
||||
"children": [{"text": big, "children": [{"text": big}, {"text": big}]} for _ in range(10)],
|
||||
}
|
||||
|
||||
async def fake_external(session, url, **kw):
|
||||
return "x"
|
||||
|
||||
with patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_algolia_item",
|
||||
new=AsyncMock(side_effect=fake_algolia),
|
||||
), patch(
|
||||
"src.sections.hackernews.item_enricher._fetch_external_markdown",
|
||||
new=AsyncMock(side_effect=fake_external),
|
||||
):
|
||||
enriched = await enrich_story(
|
||||
session=MagicMock(),
|
||||
story=story,
|
||||
**_kwargs(
|
||||
top_comments=10,
|
||||
top_l2_per_l1=2,
|
||||
comment_max_chars=500,
|
||||
comments_total_chars=1500,
|
||||
),
|
||||
)
|
||||
|
||||
tree = enriched["top_comments"]
|
||||
total = sum(len(n["l1"]) + sum(len(r) for r in n["replies"]) for n in tree)
|
||||
# 累计应该在 1500 附近停下(允许多收一条到 ~2000),不应该收全 10*3=30 条
|
||||
assert total <= 2000
|
||||
assert len(tree) < 10
|
||||
@@ -0,0 +1,59 @@
|
||||
"""测试 HN 首页 HTML 解析"""
|
||||
|
||||
import json
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.sections.hackernews.frontpage_scraper import parse_frontpage_html
|
||||
|
||||
|
||||
def test_parse_frontpage_returns_stories():
|
||||
fixture = (Path(__file__).parent / "fixtures" / "hn_frontpage.html").read_text(
|
||||
encoding="utf-8"
|
||||
)
|
||||
|
||||
stories = parse_frontpage_html(fixture)
|
||||
assert len(stories) >= 25
|
||||
s = stories[0]
|
||||
print(json.dumps(stories[:5], indent=4, ensure_ascii=False))
|
||||
|
||||
assert s["id"]
|
||||
assert s["title"]
|
||||
assert s["url"]
|
||||
assert isinstance(s["points"], int)
|
||||
assert isinstance(s["comments"], int)
|
||||
assert s["comments_url"].startswith("https://news.ycombinator.com/item?id=")
|
||||
|
||||
|
||||
def test_parse_frontpage_detects_show_hn_internal_url():
|
||||
html = """
|
||||
<table>
|
||||
<tr class="athing" id="111">
|
||||
<td class="title">
|
||||
<span class="titleline">
|
||||
<a href="item?id=111">Ask HN: what's new?</a>
|
||||
</span>
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td class="subtext">
|
||||
<span class="subline">
|
||||
<span class="score">50 points</span>
|
||||
by <a href="user?id=alice">alice</a>
|
||||
<span class="age"><a href="item?id=111">2 hours ago</a></span>
|
||||
| <a href="item?id=111">5 comments</a>
|
||||
</span>
|
||||
</td>
|
||||
</tr>
|
||||
</table>
|
||||
"""
|
||||
stories = parse_frontpage_html(html)
|
||||
assert len(stories) == 1
|
||||
s = stories[0]
|
||||
assert s["id"] == "111"
|
||||
assert s["url"].startswith("https://news.ycombinator.com/item?id=")
|
||||
assert s["site"] == ""
|
||||
assert s["points"] == 50
|
||||
assert s["comments"] == 5
|
||||
@@ -0,0 +1,111 @@
|
||||
"""测试 HN 板块编排:scrape → select → enrich → LLM"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
# Stub HN LLM functions until Task 16 lands them
|
||||
import src.llm as _llm
|
||||
if not hasattr(_llm, "select_ai_related_hn"):
|
||||
async def _stub_select(*a, **k):
|
||||
return [], None
|
||||
_llm.select_ai_related_hn = _stub_select
|
||||
if not hasattr(_llm, "summarize_hackernews"):
|
||||
async def _stub_summarize(*a, **k):
|
||||
return "", None
|
||||
_llm.summarize_hackernews = _stub_summarize
|
||||
|
||||
from src.sections.hackernews.section import run_hackernews_section
|
||||
|
||||
|
||||
def _cfg() -> dict:
|
||||
return {
|
||||
"filter": {"keep_days": 7},
|
||||
"sections": {
|
||||
"hackernews": {
|
||||
"enabled": True,
|
||||
"select_k": 1,
|
||||
"top_comments": 20,
|
||||
"comment_max_chars": 500,
|
||||
"link_content_max_chars": 3000,
|
||||
"request_timeout": 10,
|
||||
"algolia_base": "https://hn.algolia.com/api/v1",
|
||||
}
|
||||
},
|
||||
"llm": {
|
||||
"model": "x",
|
||||
"baseUrl": "http://x",
|
||||
"apiKeyName": "DEEPSEEK_API_KEY",
|
||||
"prompts": {
|
||||
"section_hackernews_select": "prompts/section_hackernews_select.md",
|
||||
"section_hackernews": "prompts/section_hackernews.md",
|
||||
},
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_disabled_returns_empty():
|
||||
cfg = _cfg()
|
||||
cfg["sections"]["hackernews"]["enabled"] = False
|
||||
md, err = await run_hackernews_section(cfg, now=None)
|
||||
assert md == ""
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_select_empty_returns_silent():
|
||||
cfg = _cfg()
|
||||
with patch(
|
||||
"src.sections.hackernews.section.fetch_frontpage", new=AsyncMock(return_value="<html>")
|
||||
), patch(
|
||||
"src.sections.hackernews.section.parse_frontpage_html",
|
||||
return_value=[{"id": "1", "title": "x"}],
|
||||
), patch(
|
||||
"src.llm.select_ai_related_hn",
|
||||
new=AsyncMock(return_value=([], None)),
|
||||
):
|
||||
md, err = await run_hackernews_section(cfg, now=None)
|
||||
assert md == ""
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_happy_path():
|
||||
cfg = _cfg()
|
||||
front = [{"id": "1", "title": "AI thing", "url": "https://e.com/a", "site": "e.com", "points": 100, "comments": 5, "comments_url": "x"}]
|
||||
enriched = [{**front[0], "link_content": "body", "top_comments": ["c1"]}]
|
||||
|
||||
with patch(
|
||||
"src.sections.hackernews.section.fetch_frontpage", new=AsyncMock(return_value="<html>")
|
||||
), patch(
|
||||
"src.sections.hackernews.section.parse_frontpage_html", return_value=front
|
||||
), patch(
|
||||
"src.llm.select_ai_related_hn",
|
||||
new=AsyncMock(return_value=(["1"], None)),
|
||||
), patch(
|
||||
"src.sections.hackernews.section.enrich_stories",
|
||||
new=AsyncMock(return_value=(enriched, [])),
|
||||
), patch(
|
||||
"src.llm.summarize_hackernews",
|
||||
new=AsyncMock(return_value=("## HN md", None)),
|
||||
):
|
||||
md, err = await run_hackernews_section(cfg, now=None)
|
||||
assert md == "## HN md"
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_scrape_failure_returns_error():
|
||||
cfg = _cfg()
|
||||
with patch(
|
||||
"src.sections.hackernews.section.fetch_frontpage",
|
||||
new=AsyncMock(side_effect=RuntimeError("net")),
|
||||
):
|
||||
md, err = await run_hackernews_section(cfg, now=None)
|
||||
assert md == ""
|
||||
assert "net" in err
|
||||
@@ -0,0 +1,66 @@
|
||||
"""测试 insights 模块"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
# Stub Task 17's LLM function until it lands
|
||||
import src.llm as _llm
|
||||
if not hasattr(_llm, "generate_trend_insights"):
|
||||
async def _stub(*a, **k):
|
||||
return "", None
|
||||
_llm.generate_trend_insights = _stub
|
||||
|
||||
from src.sections.insights.section import run_insights_section
|
||||
|
||||
|
||||
def _cfg() -> dict:
|
||||
return {
|
||||
"filter": {"push_context_days": 5},
|
||||
"sections": {"insights": {"enabled": True}},
|
||||
"llm": {
|
||||
"model": "x",
|
||||
"baseUrl": "http://x",
|
||||
"apiKeyName": "DEEPSEEK_API_KEY",
|
||||
"prompts": {"insights": "prompts/insights.md"},
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_disabled_returns_empty():
|
||||
cfg = _cfg()
|
||||
cfg["sections"]["insights"]["enabled"] = False
|
||||
md, meta, err = await run_insights_section("rss", "gh", "hn", cfg, now=None)
|
||||
assert md == ""
|
||||
assert meta is None
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_marks_empty_sections_for_llm():
|
||||
cfg = _cfg()
|
||||
captured = {}
|
||||
|
||||
async def fake_gen(sections, config):
|
||||
captured["sections"] = sections
|
||||
return "insights md", None
|
||||
|
||||
with patch(
|
||||
"src.llm.generate_trend_insights",
|
||||
new=AsyncMock(side_effect=fake_gen),
|
||||
):
|
||||
md, meta, err = await run_insights_section("", "gh md", "", cfg, now=None)
|
||||
|
||||
assert md == "insights md"
|
||||
assert err is None
|
||||
# metadata 由 parse_insights_with_metadata 注入默认标题/profile
|
||||
assert meta["profile"] == "morning"
|
||||
assert "📰 AI Daily 每日精选" in meta["title"]
|
||||
assert captured["sections"]["rss"] == "(本次无内容)"
|
||||
assert captured["sections"]["github"] == "gh md"
|
||||
assert captured["sections"]["hackernews"] == "(本次无内容)"
|
||||
@@ -0,0 +1,76 @@
|
||||
"""src.main.collect_entries_for_push patched at source (lazy import in section)"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
from unittest.mock import patch, AsyncMock
|
||||
|
||||
import pytest
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||
|
||||
from src.sections.rss.section import run_rss_section
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_returns_markdown_when_entries_present(sample_config, tmp_path):
|
||||
digest_raw = (
|
||||
"---\n"
|
||||
'title: "🌙 AI Daily 晚报 | 测试"\n'
|
||||
'lead: "今日测试导读"\n'
|
||||
"highlights:\n - 重点1\n"
|
||||
"---\n\n"
|
||||
"### 1️⃣ digest body"
|
||||
)
|
||||
with patch(
|
||||
"src.main.collect_entries_for_push",
|
||||
return_value=([{"link": "x", "title": "t", "score": 80}], []),
|
||||
), patch(
|
||||
"src.sections.rss.section.compose_digest",
|
||||
new=AsyncMock(return_value=digest_raw),
|
||||
), patch(
|
||||
"src.sections.rss.section.load_recent_push_content", return_value=""
|
||||
), patch(
|
||||
"src.sections.rss.section.get_last_push_file", return_value=None
|
||||
):
|
||||
md, meta, err = await run_rss_section(sample_config, now=None)
|
||||
|
||||
assert md == "### 1️⃣ digest body"
|
||||
assert err is None
|
||||
assert meta["title"] == "🌙 AI Daily 晚报 | 测试"
|
||||
assert meta["lead"] == "今日测试导读"
|
||||
assert meta["highlights"] == ["重点1"]
|
||||
assert meta["profile"] == "default"
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_returns_empty_when_no_entries(sample_config):
|
||||
with patch(
|
||||
"src.main.collect_entries_for_push", return_value=([], [])
|
||||
), patch(
|
||||
"src.sections.rss.section.get_last_push_file", return_value=None
|
||||
):
|
||||
md, meta, err = await run_rss_section(sample_config, now=None)
|
||||
|
||||
assert md == ""
|
||||
assert meta is None
|
||||
assert err is None
|
||||
|
||||
|
||||
@pytest.mark.asyncio
|
||||
async def test_returns_error_on_compose_failure(sample_config):
|
||||
with patch(
|
||||
"src.main.collect_entries_for_push",
|
||||
return_value=([{"link": "x"}], []),
|
||||
), patch(
|
||||
"src.sections.rss.section.compose_digest",
|
||||
new=AsyncMock(side_effect=RuntimeError("LLM down")),
|
||||
), patch(
|
||||
"src.sections.rss.section.load_recent_push_content", return_value=""
|
||||
), patch(
|
||||
"src.sections.rss.section.get_last_push_file", return_value=None
|
||||
):
|
||||
md, meta, err = await run_rss_section(sample_config, now=None)
|
||||
|
||||
assert md == ""
|
||||
assert meta is None
|
||||
assert "LLM down" in err
|
||||
@@ -0,0 +1,257 @@
|
||||
"""存储模块测试"""
|
||||
|
||||
import json
|
||||
import pytest
|
||||
import sys
|
||||
from datetime import datetime, date, timedelta, timezone
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from storage import (
|
||||
get_fetch_file,
|
||||
get_push_file,
|
||||
get_last_push_file,
|
||||
extract_push_time,
|
||||
read_entries,
|
||||
read_fetch_data,
|
||||
save_fetch_file,
|
||||
append_entries,
|
||||
format_entry,
|
||||
json_to_md,
|
||||
save_push_file,
|
||||
load_existing_links,
|
||||
cleanup_old_files,
|
||||
)
|
||||
|
||||
|
||||
class TestGetFetchFile:
|
||||
"""测试获取fetch文件路径"""
|
||||
|
||||
def test_get_fetch_file_default(self):
|
||||
result = get_fetch_file()
|
||||
assert "fetch-" in result
|
||||
assert result.endswith(".json")
|
||||
|
||||
def test_get_fetch_file_specific_date(self):
|
||||
result = get_fetch_file(date(2024, 1, 15))
|
||||
assert "fetch-2024-01-15.json" in result
|
||||
|
||||
|
||||
class TestGetPushFile:
|
||||
"""测试获取push文件路径"""
|
||||
|
||||
def test_get_push_file_default(self):
|
||||
result = get_push_file()
|
||||
assert "push-" in result
|
||||
assert result.endswith(".md")
|
||||
|
||||
def test_get_push_file_specific_time(self):
|
||||
dt = datetime(2024, 1, 15, 8, 30, 0)
|
||||
result = get_push_file(dt)
|
||||
assert "push-2024-01-15-08-30-00.md" in result
|
||||
|
||||
|
||||
class TestExtractPushTime:
|
||||
"""测试从文件名提取时间"""
|
||||
|
||||
def test_extract_valid_time(self):
|
||||
result = extract_push_time("news-data/push-2024-01-15-08-30-00.md")
|
||||
assert result is not None
|
||||
assert result.year == 2024
|
||||
assert result.month == 1
|
||||
assert result.day == 15
|
||||
|
||||
def test_extract_invalid_filename(self):
|
||||
result = extract_push_time("invalid.md")
|
||||
assert result is None
|
||||
|
||||
|
||||
class TestGetLastPushFile:
|
||||
"""测试获取最新push文件"""
|
||||
|
||||
def test_get_last_push_file_empty(self, temp_dir):
|
||||
result = get_last_push_file(str(temp_dir))
|
||||
assert result is None
|
||||
|
||||
def test_get_last_push_file_exists(self, temp_dir):
|
||||
(temp_dir / "push-2024-01-14-10-00-00.md").touch()
|
||||
(temp_dir / "push-2024-01-15-10-00-00.md").touch()
|
||||
|
||||
result = get_last_push_file(str(temp_dir))
|
||||
assert "2024-01-15" in result
|
||||
|
||||
|
||||
class TestReadWriteEntries:
|
||||
"""测试读写条目"""
|
||||
|
||||
def test_read_entries(self, sample_fetch_json):
|
||||
entries = read_entries(sample_fetch_json)
|
||||
assert len(entries) == 3
|
||||
assert entries[0]["title"] == "Article 1"
|
||||
|
||||
def test_read_entries_missing_file(self):
|
||||
entries = read_entries("nonexistent.json")
|
||||
assert entries == []
|
||||
|
||||
def test_read_fetch_data(self, sample_fetch_json):
|
||||
data = read_fetch_data(sample_fetch_json)
|
||||
assert "meta" in data
|
||||
assert "entries" in data
|
||||
assert len(data["entries"]) == 3
|
||||
|
||||
|
||||
class TestSaveFetchFile:
|
||||
"""测试保存fetch文件"""
|
||||
|
||||
def test_save_fetch_file(self, temp_dir):
|
||||
filepath = str(temp_dir / "test.json")
|
||||
meta = {"date": "2024-01-15"}
|
||||
entries = [{"title": "Test", "link": "https://example.com", "score": 80}]
|
||||
|
||||
save_fetch_file(filepath, meta, entries)
|
||||
|
||||
data = read_fetch_data(filepath)
|
||||
assert data["meta"]["date"] == "2024-01-15"
|
||||
assert len(data["entries"]) == 1
|
||||
|
||||
|
||||
class TestAppendEntries:
|
||||
"""测试追加条目"""
|
||||
|
||||
def test_append_new_entries(self, temp_dir):
|
||||
filepath = str(temp_dir / "test.json")
|
||||
meta = {"date": "2024-01-15"}
|
||||
|
||||
entries1 = [{"title": "Entry1", "link": "https://example.com/1", "score": 80}]
|
||||
count1 = append_entries(filepath, entries1, meta)
|
||||
assert count1 == 1
|
||||
|
||||
entries2 = [{"title": "Entry2", "link": "https://example.com/2", "score": 70}]
|
||||
count2 = append_entries(filepath, entries2, meta)
|
||||
assert count2 == 1
|
||||
|
||||
def test_append_duplicate_entries(self, temp_dir):
|
||||
filepath = str(temp_dir / "test.json")
|
||||
meta = {"date": "2024-01-15"}
|
||||
|
||||
entries = [{"title": "Entry1", "link": "https://example.com/1", "score": 80}]
|
||||
append_entries(filepath, entries, meta)
|
||||
|
||||
all_entries = read_entries(filepath)
|
||||
assert len(all_entries) == 1
|
||||
|
||||
count = append_entries(filepath, entries, meta)
|
||||
all_entries_after = read_entries(filepath)
|
||||
assert len(all_entries_after) == 1
|
||||
|
||||
def test_append_to_existing_file(self, temp_dir):
|
||||
filepath = str(temp_dir / "test.json")
|
||||
|
||||
data = {
|
||||
"meta": {"date": "2024-01-15"},
|
||||
"entries": [{"title": "Old", "link": "https://old.com", "score": 60}],
|
||||
}
|
||||
with open(filepath, "w") as f:
|
||||
json.dump(data, f)
|
||||
|
||||
new_entries = [{"title": "New", "link": "https://new.com", "score": 70}]
|
||||
count = append_entries(filepath, new_entries)
|
||||
|
||||
entries = read_entries(filepath)
|
||||
assert len(entries) == 2
|
||||
|
||||
|
||||
class TestFormatEntry:
|
||||
"""测试格式化条目"""
|
||||
|
||||
def test_format_entry_basic(self, sample_entry):
|
||||
result = format_entry(sample_entry)
|
||||
assert "## Test Article Title" in result
|
||||
assert "source: Test Source" in result
|
||||
assert "score: 85" in result
|
||||
|
||||
def test_format_entry_with_tags(self, sample_entry):
|
||||
result = format_entry(sample_entry)
|
||||
assert "AI" in result
|
||||
assert "Tech" in result
|
||||
|
||||
|
||||
class TestJsonToMd:
|
||||
"""测试JSON转Markdown"""
|
||||
|
||||
def test_json_to_md_basic(self, sample_fetch_json):
|
||||
data = read_fetch_data(sample_fetch_json)
|
||||
result = json_to_md(data)
|
||||
|
||||
assert "Article 1" in result
|
||||
assert "Article 2" in result
|
||||
assert "Article 3" in result
|
||||
|
||||
def test_json_to_md_empty(self):
|
||||
data = {"meta": {}, "entries": []}
|
||||
result = json_to_md(data)
|
||||
assert result == ""
|
||||
|
||||
|
||||
class TestSavePushFile:
|
||||
"""测试保存推送文件"""
|
||||
|
||||
def test_save_push_file(self, temp_dir):
|
||||
filepath = str(temp_dir / "push-test.md")
|
||||
content = "# Test Push\n\nContent here"
|
||||
|
||||
save_push_file(filepath, content, 5, 10)
|
||||
|
||||
with open(filepath, "r") as f:
|
||||
content = f.read()
|
||||
|
||||
assert "pushDate:" in content
|
||||
assert "sourceCount: 5" in content
|
||||
assert "totalEntries: 10" in content
|
||||
assert "# Test Push" in content
|
||||
|
||||
|
||||
class TestLoadExistingLinks:
|
||||
"""测试加载已有链接"""
|
||||
|
||||
def test_load_existing_links_json(self, sample_fetch_json):
|
||||
links = load_existing_links(sample_fetch_json)
|
||||
assert len(links) == 3
|
||||
assert "https://example.com/1" in links
|
||||
|
||||
def test_load_existing_links_missing(self):
|
||||
links = load_existing_links("nonexistent.json")
|
||||
assert links == set()
|
||||
|
||||
def test_load_existing_links_empty_string(self):
|
||||
links = load_existing_links("")
|
||||
assert links == set()
|
||||
|
||||
|
||||
class TestCleanupOldFiles:
|
||||
"""测试清理旧文件"""
|
||||
|
||||
def test_cleanup_old_files(self, temp_dir):
|
||||
old_date = (datetime.now() - timedelta(days=10)).date()
|
||||
new_date = (datetime.now() - timedelta(days=1)).date()
|
||||
|
||||
(temp_dir / f"fetch-{old_date}.json").touch()
|
||||
(temp_dir / f"fetch-{new_date}.json").touch()
|
||||
|
||||
cleanup_old_files(days=7, data_dir=str(temp_dir))
|
||||
|
||||
assert not (temp_dir / f"fetch-{old_date}.json").exists()
|
||||
assert (temp_dir / f"fetch-{new_date}.json").exists()
|
||||
|
||||
def test_cleanup_push_files(self, temp_dir):
|
||||
old_time = datetime.now() - timedelta(days=10)
|
||||
new_time = datetime.now() - timedelta(days=1)
|
||||
|
||||
(temp_dir / f"push-{old_time.strftime('%Y-%m-%d-%H-%M-%S')}.md").touch()
|
||||
(temp_dir / f"push-{new_time.strftime('%Y-%m-%d-%H-%M-%S')}.md").touch()
|
||||
|
||||
cleanup_old_files(days=7, data_dir=str(temp_dir))
|
||||
|
||||
files = list(temp_dir.glob("push-*.md"))
|
||||
assert len(files) == 1
|
||||
@@ -0,0 +1,163 @@
|
||||
"""测试新增的 sentinel 切片与 section-aware 读取"""
|
||||
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from storage import extract_section
|
||||
|
||||
|
||||
class TestExtractSection:
|
||||
def test_extract_section_with_sentinel(self):
|
||||
md = (
|
||||
"intro\n"
|
||||
"<!-- SECTION:rss BEGIN -->\n"
|
||||
"RSS body\n"
|
||||
"<!-- SECTION:rss END -->\n"
|
||||
"\n"
|
||||
"<!-- SECTION:github BEGIN -->\n"
|
||||
"GH body\n"
|
||||
"<!-- SECTION:github END -->\n"
|
||||
)
|
||||
assert extract_section(md, "rss").strip() == "RSS body"
|
||||
assert extract_section(md, "github").strip() == "GH body"
|
||||
assert extract_section(md, "hackernews") == ""
|
||||
|
||||
def test_extract_section_legacy_file_rss(self):
|
||||
legacy = "# AI Daily\n### 1️⃣ foo\n### 2️⃣ bar\n"
|
||||
assert extract_section(legacy, "rss") == legacy
|
||||
|
||||
def test_extract_section_legacy_file_non_rss(self):
|
||||
legacy = "# AI Daily\n### 1️⃣ foo\n"
|
||||
assert extract_section(legacy, "github") == ""
|
||||
assert extract_section(legacy, "hackernews") == ""
|
||||
assert extract_section(legacy, "insights") == ""
|
||||
|
||||
def test_extract_section_missing_end_marker(self):
|
||||
broken = "<!-- SECTION:rss BEGIN -->\ncontent only\n"
|
||||
assert extract_section(broken, "rss") == ""
|
||||
|
||||
|
||||
from datetime import date, datetime, timedelta
|
||||
from storage import save_push_file
|
||||
|
||||
|
||||
from storage import TrendingHistory, load_trending_history
|
||||
|
||||
|
||||
class TestTrendingHistory:
|
||||
def test_load_missing_file(self, tmp_path):
|
||||
path = tmp_path / "trending.json"
|
||||
h = load_trending_history(str(path))
|
||||
assert h.repos == {}
|
||||
|
||||
def test_touch_then_save_then_reload(self, tmp_path):
|
||||
path = tmp_path / "trending.json"
|
||||
h = load_trending_history(str(path))
|
||||
today = date(2026, 5, 17)
|
||||
h.touch("https://github.com/a/b", today)
|
||||
h.touch("https://github.com/c/d", today)
|
||||
h.save()
|
||||
|
||||
h2 = load_trending_history(str(path))
|
||||
assert h2.repos == {
|
||||
"https://github.com/a/b": "2026-05-17",
|
||||
"https://github.com/c/d": "2026-05-17",
|
||||
}
|
||||
|
||||
def test_contains_returns_membership(self, tmp_path):
|
||||
h = load_trending_history(str(tmp_path / "x.json"))
|
||||
h.touch("https://github.com/a/b", date(2026, 5, 17))
|
||||
assert "https://github.com/a/b" in h
|
||||
assert "https://github.com/x/y" not in h
|
||||
|
||||
def test_cleanup_removes_expired_entries(self, tmp_path):
|
||||
path = tmp_path / "trending.json"
|
||||
path.write_text(
|
||||
'{"repos": {'
|
||||
'"https://github.com/old/repo": "2026-05-01", '
|
||||
'"https://github.com/new/repo": "2026-05-15"'
|
||||
'}, "updated_at": "2026-05-15T00:00:00+08:00"}',
|
||||
encoding="utf-8",
|
||||
)
|
||||
h = load_trending_history(str(path))
|
||||
h.cleanup(today=date(2026, 5, 17), keep_days=7)
|
||||
assert "https://github.com/old/repo" not in h
|
||||
assert "https://github.com/new/repo" in h
|
||||
|
||||
def test_cleanup_keeps_today_inclusive(self, tmp_path):
|
||||
h = load_trending_history(str(tmp_path / "x.json"))
|
||||
h.touch("https://github.com/a/b", date(2026, 5, 10))
|
||||
# 2026-05-10 + 7 days = 2026-05-17 (last_seen 2026-05-10 仍在 keep 区间)
|
||||
h.cleanup(today=date(2026, 5, 17), keep_days=7)
|
||||
assert "https://github.com/a/b" in h
|
||||
# 再过 1 天就出区间
|
||||
h.cleanup(today=date(2026, 5, 18), keep_days=7)
|
||||
assert "https://github.com/a/b" not in h
|
||||
|
||||
|
||||
class TestSavePushFileProfile:
|
||||
def test_default_profile_when_not_specified(self, tmp_path):
|
||||
f = tmp_path / "push-x.md"
|
||||
save_push_file(str(f), "body content", source_count=1, total_entries=1)
|
||||
text = f.read_text(encoding="utf-8")
|
||||
assert 'profile: "default"' in text
|
||||
assert "body content" in text
|
||||
|
||||
def test_morning_profile(self, tmp_path):
|
||||
f = tmp_path / "push-x.md"
|
||||
save_push_file(
|
||||
str(f), "body", source_count=2, total_entries=3, profile="morning"
|
||||
)
|
||||
text = f.read_text(encoding="utf-8")
|
||||
assert 'profile: "morning"' in text
|
||||
|
||||
|
||||
from storage import cleanup_old_files
|
||||
import json as _j
|
||||
|
||||
|
||||
class TestCleanupOldFilesTrendingHistory:
|
||||
def test_prunes_trending_history_entries_not_file(self, tmp_path):
|
||||
path = tmp_path / "trending-history.json"
|
||||
old_date = (datetime.now().date() - timedelta(days=30)).isoformat()
|
||||
fresh_date = datetime.now().date().isoformat()
|
||||
path.write_text(
|
||||
'{"repos": {'
|
||||
f'"https://github.com/a/b": "{old_date}", '
|
||||
f'"https://github.com/c/d": "{fresh_date}"'
|
||||
'}, "updated_at": "..."}',
|
||||
encoding="utf-8",
|
||||
)
|
||||
cleanup_old_files(days=7, data_dir=str(tmp_path))
|
||||
# 文件应保留
|
||||
assert path.exists()
|
||||
# 过期条目应被剪枝
|
||||
data = _j.loads(path.read_text(encoding="utf-8"))
|
||||
assert "https://github.com/a/b" not in data["repos"]
|
||||
assert "https://github.com/c/d" in data["repos"]
|
||||
|
||||
|
||||
from storage import assemble_with_sentinels
|
||||
|
||||
|
||||
class TestAssembleWithSentinels:
|
||||
def test_assembles_all_sections_in_order(self):
|
||||
out = assemble_with_sentinels(
|
||||
{"rss": "R", "github": "G", "hackernews": "H", "insights": "I"}
|
||||
)
|
||||
assert out.index("SECTION:rss") < out.index("SECTION:github")
|
||||
assert out.index("SECTION:github") < out.index("SECTION:hackernews")
|
||||
assert out.index("SECTION:hackernews") < out.index("SECTION:insights")
|
||||
assert "<!-- SECTION:rss BEGIN -->\nR\n<!-- SECTION:rss END -->" in out
|
||||
|
||||
def test_omits_empty_sections(self):
|
||||
out = assemble_with_sentinels({"rss": "R", "github": "", "hackernews": "H", "insights": ""})
|
||||
assert "SECTION:github" not in out
|
||||
assert "SECTION:insights" not in out
|
||||
assert "SECTION:rss" in out
|
||||
assert "SECTION:hackernews" in out
|
||||
|
||||
def test_returns_empty_when_all_empty(self):
|
||||
assert assemble_with_sentinels({"rss": "", "github": "", "hackernews": "", "insights": ""}) == ""
|
||||
@@ -0,0 +1,101 @@
|
||||
"""时区处理测试"""
|
||||
|
||||
import pytest
|
||||
import sys
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from pathlib import Path
|
||||
|
||||
sys.path.insert(0, str(Path(__file__).parent.parent.parent / "src"))
|
||||
|
||||
from config import get_timezone
|
||||
|
||||
|
||||
class TestTimezoneBasics:
|
||||
"""测试时区基础功能"""
|
||||
|
||||
def test_get_timezone_positive_hours(self):
|
||||
config = {"schedule": {"timezone_hours": 8}}
|
||||
tz = get_timezone(config)
|
||||
offset = tz.utcoffset(datetime.now())
|
||||
assert offset.total_seconds() == 8 * 3600
|
||||
|
||||
def test_get_timezone_negative_hours(self):
|
||||
config = {"schedule": {"timezone_hours": -5}}
|
||||
tz = get_timezone(config)
|
||||
offset = tz.utcoffset(datetime.now())
|
||||
assert offset.total_seconds() == -5 * 3600
|
||||
|
||||
def test_get_timezone_zero_hours(self):
|
||||
config = {"schedule": {"timezone_hours": 0}}
|
||||
tz = get_timezone(config)
|
||||
offset = tz.utcoffset(datetime.now())
|
||||
assert offset.total_seconds() == 0
|
||||
|
||||
def test_get_timezone_missing_schedule(self):
|
||||
config = {}
|
||||
tz = get_timezone(config)
|
||||
assert tz is not None
|
||||
|
||||
def test_get_timezone_none_config(self):
|
||||
tz = get_timezone(None)
|
||||
assert tz is not None
|
||||
|
||||
|
||||
class TestTimezoneConversions:
|
||||
"""测试时区转换"""
|
||||
|
||||
def test_utc_to_local(self):
|
||||
config = {"schedule": {"timezone_hours": 8}}
|
||||
tz = get_timezone(config)
|
||||
|
||||
utc_time = datetime(2024, 1, 15, 10, 0, 0, tzinfo=timezone.utc)
|
||||
local_time = utc_time.astimezone(tz)
|
||||
|
||||
assert local_time.hour == 18
|
||||
|
||||
def test_cross_day_conversion(self):
|
||||
config = {"schedule": {"timezone_hours": 8}}
|
||||
tz = get_timezone(config)
|
||||
|
||||
utc_time = datetime(2024, 1, 15, 20, 0, 0, tzinfo=timezone.utc)
|
||||
local_time = utc_time.astimezone(tz)
|
||||
|
||||
assert local_time.day == 16
|
||||
|
||||
def test_negative_timezone(self):
|
||||
config = {"schedule": {"timezone_hours": -5}}
|
||||
tz = get_timezone(config)
|
||||
|
||||
utc_time = datetime(2024, 1, 15, 10, 0, 0, tzinfo=timezone.utc)
|
||||
local_time = utc_time.astimezone(tz)
|
||||
|
||||
assert local_time.hour == 5
|
||||
|
||||
|
||||
class TestTimezoneAwareDatetime:
|
||||
"""测试带时区的datetime操作"""
|
||||
|
||||
def test_now_in_config_timezone(self):
|
||||
config = {"schedule": {"timezone_hours": 8}}
|
||||
tz = get_timezone(config)
|
||||
|
||||
now_local = datetime.now(tz)
|
||||
assert now_local.tzinfo == tz
|
||||
|
||||
def test_timezone_aware_comparison(self):
|
||||
config = {"schedule": {"timezone_hours": 8}}
|
||||
tz = get_timezone(config)
|
||||
|
||||
dt1 = datetime(2024, 1, 15, 10, 0, 0, tzinfo=timezone.utc)
|
||||
dt2 = datetime(2024, 1, 15, 18, 0, 0, tzinfo=tz)
|
||||
|
||||
assert dt1 == dt2
|
||||
|
||||
def test_naive_to_aware(self):
|
||||
config = {"schedule": {"timezone_hours": 8}}
|
||||
tz = get_timezone(config)
|
||||
|
||||
naive = datetime(2024, 1, 15, 10, 0, 0)
|
||||
aware = naive.replace(tzinfo=tz)
|
||||
|
||||
assert aware.tzinfo == tz
|
||||
Reference in New Issue
Block a user