Refactor and deduplicate HTTP requests code in listers

Numerous listers were using the same page_request method or equivalent
in their implementation so prefer to deduplicate that code by adding
an http_request method in base lister class: swh.lister.pattern.Lister.

That method simply wraps a call to requests.Session.request and logs
some useful info for debugging and error reporting, also an HTTPError
will be raised if a request ends up with an error.

All listers using that new method now benefit of requests retry when
an HTTP error occurs thanks to the use of the http_retry decorator.
This commit is contained in:
Antoine Lambert 2022-09-21 19:53:22 +02:00
parent 9c55acd286
commit db6ce12e9e
28 changed files with 174 additions and 449 deletions

View file

@ -10,14 +10,10 @@ import logging
from typing import Any, Dict, Iterator, List, Optional, Tuple
import iso8601
import requests
from tenacity import before_sleep_log
from swh.lister.utils import http_retry
from swh.scheduler.interface import SchedulerInterface
from swh.scheduler.model import ListedOrigin
from .. import USER_AGENT
from ..pattern import CredentialsType, Lister
logger = logging.getLogger(__name__)
@ -59,10 +55,7 @@ class GolangLister(Lister[GolangStateType, GolangPageType]):
credentials=credentials,
)
self.session = requests.Session()
self.session.headers.update(
{"Accept": "application/json", "User-Agent": USER_AGENT}
)
self.session.headers.update({"Accept": "application/json"})
self.incremental = incremental
def state_from_dict(self, d: Dict[str, Any]) -> GolangStateType:
@ -87,24 +80,8 @@ class GolangLister(Lister[GolangStateType, GolangPageType]):
):
self.updated = True
@http_retry(
before_sleep=before_sleep_log(logger, logging.WARNING),
)
def api_request(self, url: str) -> List[str]:
logger.debug("Fetching URL %s", url)
response = self.session.get(url)
if response.status_code not in (200, 304):
# Log response content to ease debugging
logger.warning(
"Unexpected HTTP status code %s for URL %s",
response.status_code,
response.url,
)
response.raise_for_status()
response = self.http_request(url)
return response.text.split()
def get_single_page(

View file

@ -98,11 +98,12 @@ def _generate_responses(datadir, requests_mock):
def test_golang_lister(swh_scheduler, mocker, requests_mock, datadir):
# first listing, should return one origin per package
lister = GolangLister(scheduler=swh_scheduler)
# Exponential retries take a long time, so stub time.sleep
mocked_sleep = mocker.patch.object(lister.api_request.retry, "sleep")
mocked_sleep = mocker.patch.object(GolangLister.http_request.retry, "sleep")
# first listing, should return one origin per package
lister = GolangLister(scheduler=swh_scheduler)
_generate_responses(datadir, requests_mock)
@ -131,7 +132,7 @@ def test_golang_lister(swh_scheduler, mocker, requests_mock, datadir):
# doing it all again (without incremental) should give us the same result
lister = GolangLister(scheduler=swh_scheduler)
mocked_sleep = mocker.patch.object(lister.api_request.retry, "sleep")
_generate_responses(datadir, requests_mock)
stats = lister.run()