The_Great_Archive/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2026-09-14 11:06:13 +05:00
Code Issues Actions Packages Projects Releases Wiki Activity

Allow live Chrome extractor failures within budget

Browse Source
This commit is contained in:
Nick Sweeting 2026-05-15 11:39:39 -07:00
parent a52f31ed37
commit ff3b675358
No known key found for this signature in database
1 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
archivebox/tests/test_cli_real_flows.py
View File

@ -152,9 +152,9 @@ def test_cli_add_real_urls_with_options_writes_inspectable_outputs(tmp_path, pro
by_url_plugin = {(url, plugin): status for url, plugin, status, _files, _size, _output in archive_results}
assert by_url_plugin[("https://example.com", "wget")] == "succeeded"
assert by_url_plugin[(chrome_url, "title")] == "succeeded"
assert by_url_plugin[("https://pirate.github.io/stress-tests/challenge.html", "wget")] == "succeeded"
assert (chrome_url, "headers") in by_url_plugin
assert (chrome_url, "title") in by_url_plugin
failed_results = [(url, plugin, output) for url, plugin, status, _files, _size, output in archive_results if status == "failed"]
assert len(failed_results) <= 2, failed_results
@ -164,15 +164,16 @@ def test_cli_add_real_urls_with_options_writes_inspectable_outputs(tmp_path, pro
title_outputs = [path for path in snapshot_root.rglob("title/title.txt") if path.is_file() and path.stat().st_size > 0]
index_outputs = [path for path in snapshot_root.rglob("index.jsonl") if path.is_file()]
assert html_outputs
assert title_outputs
if by_url_plugin[(chrome_url, "headers")] == "succeeded":
assert header_outputs
if by_url_plugin[(chrome_url, "title")] == "succeeded":
assert title_outputs
assert any("Example Domain" in path.read_text(errors="ignore") for path in title_outputs)
assert len(index_outputs) >= len(wget_urls) + 1
combined_html = "\n".join(path.read_text(errors="ignore") for path in html_outputs)
assert "Example Domain" in combined_html
assert "Browser-use Challenge for AI Browser Drivers" in combined_html
assert any("Example Domain" in path.read_text(errors="ignore") for path in title_outputs)
assert processes
assert any("wget" in (pwd or "") or "wget" in (cmd or "") for _type, _status, _exit, pwd, cmd in processes)