1 # Fedi API Block - An aggregator for fetching blocking data from fediverse nodes
2 # Copyright (C) 2023 Free Software Foundation
4 # This program is free software: you can redistribute it and/or modify
5 # it under the terms of the GNU Affero General Public License as published
6 # by the Free Software Foundation, either version 3 of the License, or
7 # (at your option) any later version.
9 # This program is distributed in the hope that it will be useful,
10 # but WITHOUT ANY WARRANTY; without even the implied warranty of
11 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 # GNU Affero General Public License for more details.
14 # You should have received a copy of the GNU Affero General Public License
15 # along with this program. If not, see <https://www.gnu.org/licenses/>.
19 from urllib.parse import urlparse
23 from fba.helpers import blacklist
25 from fba.models import instances
27 logging.basicConfig(level=logging.INFO)
28 logger = logging.getLogger(__name__)
30 def raise_on(domain: str):
31 logger.debug("domain='%s' - CALLED!", domain)
33 if not isinstance(domain, str):
34 raise ValueError(f"Parameter domain[]='{type(domain)}' is not of type 'str'")
36 raise ValueError("Parameter 'domain' is empty")
37 elif domain.lower() != domain:
38 raise ValueError(f"Parameter domain='{domain}' must be all lower-case")
39 elif not validators.domain(domain.split("/")[0]):
40 raise ValueError(f"domain='{domain}' is not a valid domain")
41 elif domain.endswith(".arpa"):
42 raise ValueError(f"domain='{domain}' is a domain for reversed IP addresses, please don't crawl them!")
43 elif domain.endswith(".onion"):
44 raise ValueError(f"domain='{domain}' is a TOR, please don't crawl them!")
45 elif domain.endswith(".tld"):
46 raise ValueError(f"domain='{domain}' is a fake domain, please don't crawl them!")
50 def is_in_url(domain: str, url: str) -> bool:
51 logger.debug("domain='%s',url='%s' - CALLED!", domain, url)
54 if not isinstance(url, str):
55 raise ValueError(f"Parameter url[]='{type(url)}' is not of type 'str'")
57 raise ValueError("Parameter 'url' is empty")
59 punycode = domain.encode("idna").decode("utf-8")
61 components = urlparse(url)
62 logger.debug("components[]='%s',punycode='%s'", type(components), punycode)
64 is_found = (punycode in [components.netloc, components.hostname])
66 logger.debug("is_found='%s' - EXIT!", is_found)
69 def is_wanted(domain: str) -> bool:
70 logger.debug("domain='%s' - CALLED!", domain)
72 if not isinstance(domain, str):
73 raise ValueError(f"Parameter domain[]='{type(domain)}' is not of type 'str'")
75 raise ValueError("Parameter 'domain' is empty")
78 if domain.lower() != domain:
80 elif not validators.domain(domain.split("/")[0]):
81 logger.debug("domain='%s' is not a valid domain name - setting False ...", domain)
83 elif domain.endswith(".arpa"):
84 logger.debug("domain='%s' is a domain for reversed IP addresses - setting False ...", domain)
86 elif domain.endswith(".onion"):
87 logger.debug("domain='%s' is a TOR .onion domain - setting False ...", domain)
89 elif domain.endswith(".tld"):
90 logger.debug("domain='%s' is a fake domain - setting False ...", domain)
92 elif blacklist.is_blacklisted(domain):
93 logger.debug("domain='%s' is blacklisted - setting False ...", domain)
95 elif domain.find("/profile/") > 0 or domain.find("/users/") > 0 or (instances.is_registered(domain.split("/")[0]) and domain.find("/c/") > 0):
96 logger.debug("domain='%s' is a single user", domain)
98 elif domain.find("/tag/") > 0:
99 logger.debug("domain='%s' is a tag", domain)
102 logger.debug("wanted='%s' - EXIT!", wanted)