cdransf/ai.robots.txt
1
1
Fork 0
mirror of https://github.com/ai-robots-txt/ai.robots.txt.git synced 2025-04-03 02:43:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Add tests for Nginx config generator

Browse source
This commit is contained in:
Thomas Leister 2025-03-27 16:12:18 +01:00
parent 4f3f4cd0dd
commit 7c3b5a2cb2
No known key found for this signature in database
GPG key ID: B7CFA61477D48DCB
2 changed files with 14 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
code/test_files/nginx-block-ai-bots.conf Normal file
View file

@ -0,0 +1,3 @@
if ($http_user_agent ~* "(AI2Bot|Ai2Bot\-Dolma|Amazonbot|anthropic\-ai|Applebot|Applebot\-Extended|Bytespider|CCBot|ChatGPT\-User|Claude\-Web|ClaudeBot|cohere\-ai|Diffbot|FacebookBot|facebookexternalhit|FriendlyCrawler|Google\-Extended|GoogleOther|GoogleOther\-Image|GoogleOther\-Video|GPTBot|iaskspider/2\.0|ICC\-Crawler|ImagesiftBot|img2dataset|ISSCyberRiskCrawler|Kangaroo\ Bot|Meta\-ExternalAgent|Meta\-ExternalFetcher|OAI\-SearchBot|omgili|omgilibot|PerplexityBot|PetalBot|Scrapy|Sidetrade\ indexer\ bot|Timpibot|VelenPublicWebCrawler|Webzio\-Extended|YouBot|crawler\.with\.dots|star\*\*\*crawler|Is\ this\ a\ crawler\?|a\[mazing\]\{42\}\(robot\)|2\^32\$|curl\|sudo\ bash)") {
return 403;
}

12
code/tests.py
View file

@ -4,7 +4,7 @@
import json
import unittest
from robots import json_to_txt, json_to_table, json_to_htaccess
from robots import json_to_txt, json_to_table, json_to_htaccess, json_to_nginx
class RobotsUnittestExtensions:
def loadJson(self, pathname):
@ -50,6 +50,16 @@ class TestHtaccessGeneration(unittest.TestCase, RobotsUnittestExtensions):
robots_htaccess = json_to_htaccess(self.robots_dict)
self.assertEqualsFile("test_files/.htaccess", robots_htaccess)
class TestNginxConfigGeneration(unittest.TestCase, RobotsUnittestExtensions):
maxDiff = 8192
def setUp(self):
self.robots_dict = self.loadJson("test_files/robots.json")
def test_nginx_generation(self):
robots_nginx = json_to_nginx(self.robots_dict)
self.assertEqualsFile("test_files/nginx-block-ai-bots.conf", robots_nginx)
if __name__ == "__main__":
import os