在 Django 应用程序中，如果不明确为每个 View 编写单元测试，我如何测试是否没有损坏的链接？
测试应该“浏览”它在站点上找到的所有链接并测试“200 OK”响应。
最佳答案
下面的一段代码被证明对我非常有用。 它立即显示了一些损坏的链接，所以我想在这里分享。
它递归遍历所有链接，从 URL 列表开始，并检查 200 OK 响应。
还可以给出要避免的 URL 列表。
注意事项:
- 使用 django 1.8.16 和 python 2.7 测试
- 需要 beatifulsoup4(pip 安装 beautifulsoup4)
开始吧:
from __future__ import print_function
from django.test import TestCase
from bs4 import BeautifulSoup
import re
from django.contrib.auth.models import User
VERBOSE = True
class TraverseLinksTest(TestCase):
def setUp(self):
# By default, login as superuser
self.superuser = User.objects.create_superuser('superuser1', 'superuser1@example.com', 'pwd')
if self.client.login(username="superuser1@example.com", password="pwd"):
if VERBOSE: print('\nLogin as superuser OK')
else:
raise BaseException('Login failed')
@classmethod
def setUpTestData(cls):
# Initialise your database here as needed
pass
def test_traverse_urls(self):
# Fill these lists as needed with your site specific URLs to check and to avoid
to_traverse_list = ['/mysite', '/mysite/sub-page']
to_avoid_list = ['^/$', '^$', 'javascript:history\.back()', 'javascript:history\.go\(-1\)', '^mailto:.*', '.*github\.io.*']
done_list = []
error_list = []
source_of_link = dict()
for link in to_traverse_list:
source_of_link[link] = 'initial'
(to_traverse_list, to_avoid_list, done_list, error_list, source_of_link) = \
self.recurse_into_path(to_traverse_list, to_avoid_list, done_list, error_list, source_of_link)
print('END REACHED\nStats:')
if VERBOSE: print('\nto_traverse_list = ' + str(to_traverse_list))
if VERBOSE: print('\nto_avoid_list = ' + str(to_avoid_list))
if VERBOSE: print('\nsource_of_link = ' + str(source_of_link))
if VERBOSE: print('\ndone_list = ' + str(done_list))
print('Followed ' + str(len(done_list)) + ' links successfully')
print('Avoided ' + str(len(to_avoid_list)) + ' links')
if error_list:
print('!! ' + str(len(error_list)) + ' error(s) : ')
for error in error_list:
print(str(error) + ' found in page ' + source_of_link[error[0]])
print('Errors found traversing links')
assert False
else:
print('No errors')
def recurse_into_path(self, to_traverse_list, to_avoid_list, done_list, error_list, source_of_link):
""" Dives into first item of to_traverse_list
Returns: (to_traverse_list, to_avoid_list, done_list, source_of_link)
"""
if to_traverse_list:
url = to_traverse_list.pop()
if not match_any(url, to_avoid_list):
print('Surfing to ' + str(url) + ', discovered in ' + str(source_of_link[url]))
response = self.client.get(url, follow=True)
if response.status_code == 200:
soup = BeautifulSoup(response.content, 'html.parser')
text = soup.get_text()
for link in soup.find_all('a'):
new_link = link.get('href')
if VERBOSE: print(' Found link: ' + str(new_link))
if match_any(new_link, to_avoid_list):
if VERBOSE: print(' Avoiding it')
elif new_link in done_list:
if VERBOSE: print(' Already done, ignoring')
elif new_link in to_traverse_list:
if VERBOSE: print(' Already in to traverse list, ignoring')
else:
if VERBOSE: print(' New, unknown link: Storing it to traverse later')
source_of_link[new_link] = url
to_traverse_list.append(new_link)
done_list.append(url)
if VERBOSE: print('Done')
else:
error_list.append((url, response.status_code))
to_avoid_list.append(url)
if VERBOSE: print('Diving into next level')
return self.recurse_into_path(to_traverse_list, to_avoid_list, done_list, error_list, source_of_link)
else:
# Nothing to traverse
if VERBOSE: print('Returning to upper level')
return to_traverse_list, to_avoid_list, done_list, error_list, source_of_link
def match_any(my_string, regexp_list):
if my_string:
combined = "(" + ")|(".join(regexp_list) + ")"
return re.match(combined, my_string)
else:
# 'None' as string always matches
return True
关于python - 在 Django 中，我如何从我网站上的已知 URL 列表开始，对每个 View 递归地测试所有链接(检查 200 OK)，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/43264291/