test_replays_response_from_cassette fails because of X-Amzn-Trace-Id #184
Description
$ pytest tests/integration/test_record_modes.py
.F...... [100%]
================================== FAILURES ===================================
_____________ TestRecordOnce.test_replays_response_from_cassette ______________
assert r0_headers == r1_headers
> assert r0.content == r1.content
E assert b'{\n "args"...org/get"\n}\n' == b'{\n "args"...org/get"\n}\n'
E At index 208 diff: b'd' != b'e'
E Full diff:
E (
E b'{\n "args": {}, \n "headers": {\n "Accept": "*/*", \n "Accept-Encodi'
E b'ng": "gzip, deflate", \n "Host": "httpbin.org", \n "User-Agent": "py'
E - b'thon-requests/2.23.0", \n "X-Amzn-Trace-Id": "Root=1-5e91e8b3-e94687be'
E ? ^^^^ ^^^...
E
E ...Full output truncated (9 lines hidden), use '-vv' to show
tests/integration/test_record_modes.py:43: AssertionError
The test code already removes some headers which may differ between requests but the difference shown above is in r1.content and I don't know what the best way is to ignore X-Amzn-Trace-Id there.
It seems as if httpbin.org uses Amazon's Elastic Load Balancing which adds a unique X-Amzn-Trace-Id header. That header is different for each request.