d8cabdfdb5
* fix: CI yml name * add: mypy configuraion * add: type annotation to waybackpy modules * add: type annotation to test modules * fix: mypy command * add: types-requests to dev deps * fix: disable max-line-length * fix: move pytest.ini into setup.cfg * add: urllib3 to deps * fix: Retry (ref: https://github.com/python/typeshed/issues/6893) * fix: f-string * fix: shorten long lines * add: staticmethod decorator to no-self-use methods * fix: str(headers)->headers_str * fix: error message * fix: revert "str(headers)->headers_str" and ignore assignment CaseInsensitiveDict with str * fix: mypy error
43 lines
1.3 KiB
Python
43 lines
1.3 KiB
Python
from waybackpy.cdx_api import WaybackMachineCDXServerAPI
|
|
|
|
|
|
def test_a() -> None:
|
|
user_agent = (
|
|
"Mozilla/5.0 (MacBook Air; M1 Mac OS X 11_4) AppleWebKit/605.1.15 "
|
|
"(KHTML, like Gecko) Version/14.1.1 Safari/604.1"
|
|
)
|
|
url = "https://twitter.com/jack"
|
|
|
|
wayback = WaybackMachineCDXServerAPI(
|
|
url=url,
|
|
user_agent=user_agent,
|
|
match_type="prefix",
|
|
collapses=["urlkey"],
|
|
start_timestamp="201001",
|
|
end_timestamp="201002",
|
|
)
|
|
# timeframe bound prefix matching enabled along with active urlkey based collapsing
|
|
|
|
snapshots = wayback.snapshots() # <class 'generator'>
|
|
|
|
for snapshot in snapshots:
|
|
assert snapshot.timestamp.startswith("2010")
|
|
|
|
|
|
def test_b() -> None:
|
|
user_agent = (
|
|
"Mozilla/5.0 (MacBook Air; M1 Mac OS X 11_4) "
|
|
"AppleWebKit/605.1.15 (KHTML, like Gecko) Version/14.1.1 Safari/604.1"
|
|
)
|
|
url = "https://www.google.com"
|
|
|
|
wayback = WaybackMachineCDXServerAPI(
|
|
url=url, user_agent=user_agent, start_timestamp="202101", end_timestamp="202112"
|
|
)
|
|
# timeframe bound prefix matching enabled along with active urlkey based collapsing
|
|
|
|
snapshots = wayback.snapshots() # <class 'generator'>
|
|
|
|
for snapshot in snapshots:
|
|
assert snapshot.timestamp.startswith("2021")
|