mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-09 12:21:57 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

enforce utf8 on literally all file operations because windows sucks

Browse source
This commit is contained in:
Nick Sweeting 2021-03-27 01:01:29 -04:00
parent 185d2f9f9b
commit bd6d9c165b
9 changed files with 29 additions and 28 deletions
Download patch file Download diff file Expand all files Collapse all files

6
tests/test_add.py
View file

@ -33,7 +33,7 @@ def test_depth_flag_0_crawls_only_the_arg_page(tmp_path, process, disable_extrac
)
archived_item_path = list(tmp_path.glob('archive/**/*'))[0]
with open(archived_item_path / "index.json", "r") as f:
with open(archived_item_path / "index.json", "r", encoding='utf-8') as f:
output_json = json.load(f)
assert output_json["base_url"] == "127.0.0.1:8080/static/example.com.html"
@ -79,7 +79,7 @@ def test_add_updates_history_json_index(tmp_path, process, disable_extractors_di
archived_item_path = list(tmp_path.glob('archive/**/*'))[0]
with open(archived_item_path / "index.json", "r") as f:
with open(archived_item_path / "index.json", "r", encoding="utf-8") as f:
output_json = json.load(f)
assert output_json["history"] != {}
@ -90,4 +90,4 @@ def test_extract_input_uses_only_passed_extractors(tmp_path, process):
archived_item_path = list(tmp_path.glob('archive/**/*'))[0]
assert (archived_item_path / "warc").exists()
assert not (archived_item_path / "singlefile.html").exists()
assert not (archived_item_path / "singlefile.html").exists()