From 2dedaf31271ae92c170aa3e46358d90439a2da30 Mon Sep 17 00:00:00 2001
From: eezd <eezd@outlook.com>
Date: Mon, 26 Oct 2026 22:07:42 +0800
Subject: [PATCH] Limit German archive pagination to filtered results

Exclude the independent latest-release list from archived links and date boundaries. Cover stopping before the requested year, continuing within range, and rejecting pages without a filtered result block.
---
 .../alphapulse/calendar/europe_archives.py    |  6 +++-
 .../europe/de_archive_before_start.html       | 28 +++++++++++++++++++
 backend/tests/test_europe_archives.py         | 22 +++++++++++++++
 3 files changed, 55 insertions(+), 1 deletion(-)
 create mode 100644 backend/tests/fixtures/europe/de_archive_before_start.html
 create mode 100644 backend/tests/test_europe_archives.py

diff --git a/backend/src/alphapulse/calendar/europe_archives.py b/backend/src/alphapulse/calendar/europe_archives.py
index 84c371f..dff4564 100644
--- a/backend/src/alphapulse/calendar/europe_archives.py
+++ b/backend/src/alphapulse/calendar/europe_archives.py
@@ -29,7 +29,11 @@ def ons_links(body: str) -> tuple[list[str], str | None]:
 
 
 def destatis_links(body: str, start: date, end: date) -> tuple[list[str], str | None]:
-    s = Selector(text=body)
+    sections = Selector(text=body).css(".s-press-search-results")
+    if not sections:
+        raise ValueError("Destatis 历史索引缺少筛选结果列表")
+    # 第二块是独立的最新新闻，日期不能参与筛选列表的翻页边界。
+    s = sections[0]
     links, days = [], []
     for node in s.css(".c-result"):
         date_text = compact(node.css(".c-result__date").xpath("string(.)").get() or "")
diff --git a/backend/tests/fixtures/europe/de_archive_before_start.html b/backend/tests/fixtures/europe/de_archive_before_start.html
new file mode 100644
index 0000000..ae34121
--- /dev/null
+++ b/backend/tests/fixtures/europe/de_archive_before_start.html
@@ -0,0 +1,28 @@
+<!-- 来源：Destatis CPI 归档第 6 页；保留筛选列表和独立最新新闻列表的真实节点。 -->
+<html><body>
+<div class="l-content-wrapper s-press-search-results">
+<div class="c-result"><h3 class="c-result__heading">
+        <a href="EN/Press/2024/09/PE24_340_611.html">
+          <span class="c-result__date">
+              Press release No. 340 of 10 September 2024</span>
+          Inflation rate at +1.9% in August 2024</a>
+      </h3></div>
+<div class="c-result"><h3 class="c-result__heading">
+        <a href="EN/Press/2024/08/PE24_329_611.html">
+          <span class="c-result__date">
+              Press release No. 329 of 29 August 2024</span>
+          Inflation rate of +1.9% expected in August 2024</a>
+      </h3></div>
+<a href="SiteGlobals/Forms/Suche/Presse/EN/Pressesuche_Formular_2.html?gtp=350792_list%253D7&amp;cl2Taxonomies_Themen_0=preise&amp;cl2Taxonomies_Themen_1=verbraucherpreisindex" title="page 7" class="forward button">
+  <img src="/SiteGlobals/Frontend/Images/icons/next.svg?__blob=normal&amp;v=9" title="next page" alt="next page">
+          </a>
+</div>
+<div class="l-content-wrapper s-press-search-results">
+<div class="c-result"><h3 class="c-result__heading">
+        <a href="EN/Press/2026/10/PE26_364_611.html">
+          <span class="c-result__date">
+              Press release No. 364 of 13 October 2026</span>
+          Inflation rate at +3.3% in September 2026</a>
+      </h3></div>
+</div>
+</body></html>
diff --git a/backend/tests/test_europe_archives.py b/backend/tests/test_europe_archives.py
new file mode 100644
index 0000000..6528fa1
--- /dev/null
+++ b/backend/tests/test_europe_archives.py
@@ -0,0 +1,22 @@
+from datetime import date
+from pathlib import Path
+
+import pytest
+
+from alphapulse.calendar.europe_archives import destatis_links
+
+
+@pytest.mark.parametrize("start,expected,has_next", [
+    (date(2025, 1, 1), [], False),
+    (date(2024, 1, 1), ["EN/Press/2024/09/PE24_340_611.html", "EN/Press/2024/08/PE24_329_611.html"], True),
+])
+def test_destatis_archive_only_uses_filtered_results_for_links_and_date_boundary(start, expected, has_next):
+    body = (Path(__file__).parent / "fixtures/europe/de_archive_before_start.html").read_text()
+    links, following = destatis_links(body, start, date(2026, 10, 26))
+    assert links == ["https://www.destatis.de/" + path for path in expected]
+    assert (following is not None) is has_next
+
+
+def test_destatis_archive_rejects_a_page_without_the_filtered_result_block():
+    with pytest.raises(ValueError, match="筛选结果"):
+        destatis_links("<html><body>Access denied</body></html>", date(2025, 1, 1), date(2026, 10, 26))