backup/warcprox
1
0
Fork 0
mirror of https://github.com/internetarchive/warcprox.git synced 2025-01-18 13:22:09 +01:00
Code Issues Projects Releases Activity

make test pass with py27

Browse Source
This commit is contained in:
Noah Levitt 2017-08-07 16:21:08 -07:00
parent 8a768dcd44
commit 30b69c5838
1 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
tests/test_warcprox.py
View File

@ -1490,7 +1490,8 @@ def test_crawl_log(warcprox_, http_daemon, archiving_proxies):
assert fields[10] == b'-'
assert fields[11] == b'-'
extra_info = json.loads(fields[12].decode('utf-8'))
assert extra_info.keys() == {'contentSize','warcFilename','warcFileOffset'}
assert set(extra_info.keys()) == {
'contentSize', 'warcFilename', 'warcFileOffset'}
assert extra_info['contentSize'] == 145
crawl_log_1 = open(os.path.join(
@ -1510,7 +1511,8 @@ def test_crawl_log(warcprox_, http_daemon, archiving_proxies):
assert fields[10] == b'-'
assert fields[11] == b'-'
extra_info = json.loads(fields[12].decode('utf-8'))
assert extra_info.keys() == {'contentSize','warcFilename','warcFileOffset'}
assert set(extra_info.keys()) == {
'contentSize', 'warcFilename', 'warcFileOffset'}
assert extra_info['contentSize'] == 145
# should be deduplicated
@ -1546,7 +1548,8 @@ def test_crawl_log(warcprox_, http_daemon, archiving_proxies):
assert fields[10] == b'http://example.com/seed'
assert fields[11] == b'duplicate:digest'
extra_info = json.loads(fields[12].decode('utf-8'))
assert extra_info.keys() == {'contentSize','warcFilename','warcFileOffset'}
assert set(extra_info.keys()) == {
'contentSize', 'warcFilename', 'warcFileOffset'}
assert extra_info['contentSize'] == 145
if __name__ == '__main__':