Mirrors/kleinanzeigen-bot
1
0
Fork 0
mirror of https://github.com/Second-Hand-Friends/kleinanzeigen-bot.git synced 2026-03-12 02:31:45 +01:00
Code Issues Packages Projects Releases Wiki Activity

fix: continue own-ad extraction when links are incomplete (#854)

Browse Source
This commit is contained in:
Jens
2026-03-02 06:05:21 +01:00
committed by GitHub
parent ed6137c8ae
commit c4a2d1c4f5
3 changed files with 77 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

52
tests/unit/test_extract.py
View File

@@ -703,6 +703,58 @@ class TestAdExtractorNavigation:
# Pagination should stop (TimeoutError in callback returns True)
assert refs == []
@pytest.mark.asyncio
async def test_extract_own_ads_urls_skips_single_item_timeout(self, test_extractor:extract_module.AdExtractor) -> None:
"""Timeout on one ad item should skip that item but keep extracting others."""
ad_list_container_mock = MagicMock()
first_item = MagicMock()
second_item = MagicMock()
valid_link = MagicMock()
valid_link.attrs = {"href": "/s-anzeige/ok/999"}
with (
patch.object(test_extractor, "web_open", new_callable = AsyncMock),
patch.object(test_extractor, "web_sleep", new_callable = AsyncMock),
patch.object(test_extractor, "web_scroll_page_down", new_callable = AsyncMock),
patch.object(test_extractor, "web_find_all", new_callable = AsyncMock, return_value = [first_item, second_item]),
patch.object(
test_extractor,
"web_find",
new_callable = AsyncMock,
side_effect = [ad_list_container_mock, TimeoutError(), ad_list_container_mock, TimeoutError(), valid_link],
),
):
refs = await test_extractor.extract_own_ads_urls()
assert refs == ["/s-anzeige/ok/999"]
@pytest.mark.asyncio
async def test_extract_own_ads_urls_skips_single_item_without_href(self, test_extractor:extract_module.AdExtractor) -> None:
"""Anchor without href should be skipped instead of adding a 'None' entry."""
ad_list_container_mock = MagicMock()
first_item = MagicMock()
second_item = MagicMock()
missing_href_link = MagicMock()
missing_href_link.attrs = {}
valid_link = MagicMock()
valid_link.attrs = {"href": "/s-anzeige/ok/999"}
with (
patch.object(test_extractor, "web_open", new_callable = AsyncMock),
patch.object(test_extractor, "web_sleep", new_callable = AsyncMock),
patch.object(test_extractor, "web_scroll_page_down", new_callable = AsyncMock),
patch.object(test_extractor, "web_find_all", new_callable = AsyncMock, return_value = [first_item, second_item]),
patch.object(
test_extractor,
"web_find",
new_callable = AsyncMock,
side_effect = [ad_list_container_mock, TimeoutError(), ad_list_container_mock, missing_href_link, valid_link],
),
):
refs = await test_extractor.extract_own_ads_urls()
assert refs == ["/s-anzeige/ok/999"]
@pytest.mark.asyncio
async def test_extract_own_ads_urls_generic_exception_in_callback(self, test_extractor:extract_module.AdExtractor) -> None:
"""Test that generic Exception in extract_page_refs callback continues pagination."""