"""历史日报解析器单元测试（基于真实样例 HTML）。"""
from __future__ import annotations
from datetime import date, datetime
from pathlib import Path
import pytest
from report_import.parser import (
ReportParseError,
parse_finance_report,
parse_intl_report,
parse_report,
)
FIXTURES = Path(__file__).parent / "fixtures"
FINANCE_HTML = (FIXTURES / "finance_news_daily_20260710_0720.html").read_text(encoding="utf-8")
INTL_HTML = (FIXTURES / "intl_news_daily_20260711_070304.html").read_text(encoding="utf-8")
class TestFinanceParse:
def test_metadata(self) -> None:
r = parse_finance_report(FINANCE_HTML, "finance_news_daily_20260710_0720.html")
assert r.report_date == date(2026, 7, 10)
assert r.report_type == "finance"
assert r.file_name == "finance_news_daily_20260710_0720.html"
assert r.generated_at == datetime(2026, 7, 11, 7, 20, 27) # header"生成于"优先
def test_ai_summary_lines(self) -> None:
r = parse_finance_report(FINANCE_HTML, "finance_news_daily_20260710_0720.html")
assert r.ai_summary is not None
assert len(r.ai_summary.splitlines()) >= 5
assert "碳达峰" in r.ai_summary
def test_sections_and_ranks(self) -> None:
r = parse_finance_report(FINANCE_HTML, "finance_news_daily_20260710_0720.html")
sections = {e.section for e in r.events}
assert sections == {"xwlb", "news", "cninfo"}
assert sum(1 for e in r.events if e.section == "xwlb") == 16
assert sum(1 for e in r.events if e.section == "news") == 20
assert sum(1 for e in r.events if e.section == "cninfo") == 20
def test_event_fields(self) -> None:
r = parse_finance_report(FINANCE_HTML, "finance_news_daily_20260710_0720.html")
xwlb = next(e for e in r.events if e.section == "xwlb")
assert xwlb.importance == 4
assert xwlb.event_type == "新闻联播"
assert xwlb.sentiment == "neutral"
assert "张国清" in xwlb.title
news = next(e for e in r.events if e.section == "news" and e.source)
assert news.source # 新闻板块带来源
def test_stats_keys(self) -> None:
r = parse_finance_report(FINANCE_HTML, "finance_news_daily_20260710_0720.html")
for key in ("pipeline", "sources", "sentiment", "importance", "event_types"):
assert key in r.stats, f"缺少 stats.{key}"
assert r.stats["pipeline"]["M1 原始文章"] == 758
class TestIntlParse:
def test_metadata(self) -> None:
r = parse_intl_report(INTL_HTML, "intl_news_daily_20260711_070304.html")
assert r.report_date == date(2026, 7, 11)
assert r.report_type == "intl"
assert r.generated_at == datetime(2026, 7, 11, 7, 3, 30)
def test_events_and_source_from_small(self) -> None:
r = parse_intl_report(INTL_HTML, "intl_news_daily_20260711_070304.html")
assert len(r.events) == 19
first = r.events[0]
assert first.section == "intl"
assert first.source == "investinglive.com" # 从摘要 [来源] 提取
assert first.url.startswith("https://investinglive.com/")
assert first.importance == 4
assert first.event_type == "地缘政治"
# 摘要中不应残留 [来源] 标记
assert first.summary is not None and "[investinglive.com]" not in first.summary
def test_stats_keys(self) -> None:
r = parse_intl_report(INTL_HTML, "intl_news_daily_20260711_070304.html")
for key in ("pipeline", "sentiment", "importance", "event_types", "source_dist"):
assert key in r.stats
assert r.stats["source_dist"][0]["来源"] == "ForexLive"
class TestParseReportDispatch:
def test_dispatch_finance(self) -> None:
r = parse_report(FINANCE_HTML, "finance_news_daily_20260710_0720.html")
assert r.report_type == "finance"
def test_dispatch_intl(self) -> None:
r = parse_report(INTL_HTML, "intl_news_daily_20260711_070304.html")
assert r.report_type == "intl"
def test_bad_filename_raises(self) -> None:
with pytest.raises(ReportParseError):
parse_report("", "not_a_report.html")