Files
sbnews/docker/crawler/tests/test_match_live_stream.py

250 lines
9.1 KiB
Python

import json
import sys
import unittest
from pathlib import Path
from unittest import mock
ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
sys.path.insert(0, str(ROOT))
from match_live_stream import (
_build_detail_url,
_build_run_result,
choose_next_fetch_at,
parse_match_live_detail_html,
MatchLiveStreamRepository,
run,
)
class FakeCursor:
def __init__(self, fetchall_result=None):
self.fetchall_result = list(fetchall_result or [])
self.executed = []
def __enter__(self):
return self
def __exit__(self, exc_type, exc, tb):
return False
def execute(self, sql, params):
self.executed.append((sql, params))
def fetchall(self):
return list(self.fetchall_result)
class FakeConnection:
def __init__(self, fetchall_result=None):
self.cursor_instance = FakeCursor(fetchall_result=fetchall_result)
self.commit_count = 0
def cursor(self):
return self.cursor_instance
def commit(self):
self.commit_count += 1
class MatchLiveStreamTest(unittest.TestCase):
def _build_repo(self, *, prefix="la_", fetchall_result=None):
repo = object.__new__(MatchLiveStreamRepository)
repo.prefix = prefix
repo.conn = FakeConnection(fetchall_result=fetchall_result)
return repo
def test_parse_json_ncctrials_detail_extracts_four_urls(self):
payload = {
"code": 200,
"msg": "ok",
"data": {
"stream": {
"flv": "http://a/flv",
"hdFlv": "http://a/hdflv",
"m3u8": "http://a/m3u8",
"hdM3u8": "http://a/hdm3u8",
}
},
}
parsed = parse_match_live_detail_html(f"detail({json.dumps(payload, ensure_ascii=False)})")
self.assertEqual(parsed["play_url_m3u8"], "http://a/m3u8")
self.assertEqual(parsed["play_url_hd_m3u8"], "http://a/hdm3u8")
self.assertEqual(parsed["play_url_flv"], "http://a/flv")
self.assertEqual(parsed["play_url_hd_flv"], "http://a/hdflv")
def test_parse_json_ncctrials_detail_accepts_trailing_semicolon(self):
payload = {
"code": 200,
"msg": "ok",
"data": {
"stream": {
"flv": "http://b/flv",
"hdFlv": "http://b/hdflv",
"m3u8": "http://b/m3u8",
"hdM3u8": "http://b/hdm3u8",
}
},
}
parsed = parse_match_live_detail_html(f"detail({json.dumps(payload, ensure_ascii=False)});")
self.assertEqual(parsed["play_url_m3u8"], "http://b/m3u8")
self.assertEqual(parsed["play_url_hd_m3u8"], "http://b/hdm3u8")
self.assertEqual(parsed["play_url_flv"], "http://b/flv")
self.assertEqual(parsed["play_url_hd_flv"], "http://b/hdflv")
def test_build_detail_url_derives_room_id_from_source_url(self):
detail_url = _build_detail_url(
"https://yyzb1.tv/room/7988511?scheduleId=undefined",
now_ts=123456,
)
self.assertEqual(
detail_url,
"https://json.ncctrials.com/room/7988511/detail.json?v=123456",
)
def test_build_run_result_exposes_top_level_error_for_partial_failure(self):
result = _build_run_result(candidate_count=3, success_count=2, failed_count=1)
self.assertFalse(result["success"])
self.assertEqual(result["candidate_count"], 3)
self.assertEqual(result["saved_count"], 2)
self.assertEqual(result["count"], 2)
self.assertEqual(result["failed_count"], 1)
self.assertIn("失败 1 条", result["summary_text"])
self.assertEqual(result["error"], "赛事直播流抓取存在失败线路: 失败 1 条")
def test_choose_next_fetch_at_uses_60s_for_empty_streams(self):
self.assertEqual(choose_next_fetch_at(now_ts=1000, has_existing_stream=False), 1060)
def test_choose_next_fetch_at_uses_600s_for_existing_streams(self):
self.assertEqual(choose_next_fetch_at(now_ts=1000, has_existing_stream=True), 1600)
def test_load_due_rows_orders_by_due_time_and_honors_limit(self):
expected_rows = [{"id": 12, "source_url": "https://yyzb1.tv/room/7988511"}]
repo = self._build_repo(fetchall_result=expected_rows)
rows = repo.load_due_rows(now_ts=1711111111, limit=7)
self.assertEqual(rows, expected_rows)
self.assertEqual(repo.conn.commit_count, 0)
self.assertEqual(len(repo.conn.cursor_instance.executed), 1)
sql, params = repo.conn.cursor_instance.executed[0]
self.assertIn("FROM `la_match_live`", sql)
self.assertIn("`next_fetch_at` <= %s", sql)
self.assertIn("ORDER BY `next_fetch_at` ASC, `id` ASC LIMIT %s", sql)
self.assertEqual(params, (1711111111, 7))
def test_mark_failed_only_updates_fetch_fields_and_preserves_stream_urls(self):
repo = self._build_repo()
error_message = "x" * 600
repo.mark_failed(row_id=88, error_message=error_message, now_ts=2000, has_existing_stream=True)
self.assertEqual(repo.conn.commit_count, 1)
self.assertEqual(len(repo.conn.cursor_instance.executed), 1)
sql, params = repo.conn.cursor_instance.executed[0]
self.assertIn("UPDATE `la_match_live` SET", sql)
self.assertIn("`fetch_status`=2", sql)
self.assertIn("`fetch_error`=%s, `last_fetch_at`=%s, `next_fetch_at`=%s, `update_time`=%s", sql)
self.assertNotIn("`play_url_flv`=%s", sql)
self.assertNotIn("`play_url_hd_flv`=%s", sql)
self.assertNotIn("`play_url_m3u8`=%s", sql)
self.assertNotIn("`play_url_hd_m3u8`=%s", sql)
self.assertEqual(params, ("x" * 500, 2000, 2600, 2000, 88))
def test_mark_success_updates_all_stream_urls_and_resets_fetch_fields(self):
repo = self._build_repo()
stream_urls = {
"play_url_flv": "http://stream/flv",
"play_url_hd_flv": "http://stream/hdflv",
"play_url_m3u8": "http://stream/m3u8",
"play_url_hd_m3u8": "http://stream/hdm3u8",
}
repo.mark_success(row_id=66, stream_urls=stream_urls, now_ts=3000)
self.assertEqual(repo.conn.commit_count, 1)
self.assertEqual(len(repo.conn.cursor_instance.executed), 1)
sql, params = repo.conn.cursor_instance.executed[0]
self.assertIn("UPDATE `la_match_live` SET", sql)
self.assertIn("`play_url_flv`=%s, `play_url_hd_flv`=%s, `play_url_m3u8`=%s, `play_url_hd_m3u8`=%s", sql)
self.assertIn("`fetch_status`=1, `fetch_error`='', `last_fetch_at`=%s, `next_fetch_at`=%s, `update_time`=%s", sql)
self.assertEqual(
params,
(
"http://stream/flv",
"http://stream/hdflv",
"http://stream/m3u8",
"http://stream/hdm3u8",
3000,
3600,
3000,
66,
),
)
def test_run_does_not_skip_due_rows_when_invoked_by_docker_task_runner(self):
class FakeRepo:
def __init__(self):
self.closed = False
self.mark_success_calls = []
def load_due_rows(self, now_ts, limit=30):
return [
{
"id": 1,
"source_url": "https://yyzb1.tv/room/190118?scheduleId=undefined",
"play_url_flv": "",
"play_url_hd_flv": "",
"play_url_m3u8": "",
"play_url_hd_m3u8": "",
}
]
def mark_success(self, row_id, stream_urls, now_ts):
self.mark_success_calls.append((row_id, stream_urls, now_ts))
def mark_failed(self, row_id, error_message, now_ts, has_existing_stream):
raise AssertionError("docker task runner path should not fall into mark_failed for this test")
def close(self):
self.closed = True
class FakeLock:
def acquire(self):
return False
def release(self):
return None
repo = FakeRepo()
stream_urls = {
"play_url_flv": "http://stream/flv",
"play_url_hd_flv": "http://stream/hdflv",
"play_url_m3u8": "http://stream/m3u8",
"play_url_hd_m3u8": "http://stream/hdm3u8",
}
with mock.patch.dict("os.environ", {"DQD_RUNNING_UNDER_TASK_RUNNER": "1"}, clear=False):
with mock.patch("match_live_stream.MatchLiveStreamRepository", return_value=repo):
with mock.patch("match_live_stream.TaskFileLock", return_value=FakeLock()):
with mock.patch("match_live_stream._fetch_stream_payload", return_value=stream_urls):
result = run({"prefix": "la_"})
self.assertTrue(result["success"])
self.assertEqual(result["candidate_count"], 1)
self.assertEqual(result["saved_count"], 1)
self.assertEqual(len(repo.mark_success_calls), 1)
self.assertTrue(repo.closed)
if __name__ == "__main__":
unittest.main()