from urllib.parse import urlparse
import bs4
+import requests
import validators
-from fba import csrf
-from fba import utils
-
+from fba.helpers import blacklist
from fba.helpers import config
from fba.helpers import cookies
from fba.helpers import domain as domain_helper
from fba.helpers import tidyup
from fba.helpers import version
+from fba.http import csrf
from fba.http import network
+from fba.http import nodeinfo
+from fba.models import blocks
from fba.models import instances
from fba.networks import lemmy
from fba.networks import misskey
from fba.networks import peertube
+# Depth counter, being raised and lowered
+_DEPTH = 0
+
+# API paths
+_api_paths = [
+ "/api/v1/instance/peers",
+ "/api/v3/site",
+]
+
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
def fetch_instances(domain: str, origin: str, software: str, command: str, path: str = None):
- logger.debug("domain='%s',origin='%s',software='%s',command='%s',path='%s' - CALLED!", domain, origin, software, command, path)
+ global _DEPTH
+ logger.debug("domain='%s',origin='%s',software='%s',command='%s',path='%s',_DEPTH=%d - CALLED!", domain, origin, software, command, path, _DEPTH)
domain_helper.raise_on(domain)
- if not isinstance(origin, str) and origin is not None:
- raise ValueError(f"Parameter origin[]='{type(origin)}' is not 'str'")
+ if blacklist.is_blacklisted(domain):
+ raise Exception(f"domain='{domain}' is blacklisted but function was invoked")
+ elif not isinstance(origin, str) and origin is not None:
+ raise ValueError(f"Parameter origin[]='{type(origin)}' is not of type 'str'")
elif not isinstance(command, str):
- raise ValueError(f"Parameter command[]='{type(command)}' is not 'str'")
+ raise ValueError(f"Parameter command[]='{type(command)}' is not of type 'str'")
elif command == "":
raise ValueError("Parameter 'command' is empty")
- elif software is None:
+ elif command in ["fetch_blocks", "fetch_cs", "fetch_bkali", "fetch_relays", "fetch_fedipact", "fetch_joinmobilizon", "fetch_joinmisskey", "fetch_joinfediverse", "fetch_relaylist"] and origin is None:
+ raise ValueError(f"Parameter command='{command}' but origin is None, please fix invoking this function.")
+ elif not isinstance(path, str) and path is not None:
+ raise ValueError(f"Parameter path[]='{type(path)}' is not of type 'str'")
+ elif path is not None and not path.startswith("/"):
+ raise ValueError(f"path='{path}' does not start with a slash")
+ elif _DEPTH > 0 and instances.is_recent(domain, "last_instance_fetch"):
+ raise ValueError(f"domain='{domain}' has recently been fetched but function was invoked")
+ elif software is None and not instances.is_recent(domain, "last_nodeinfo"):
try:
- logger.debug("Software for domain='%s' is not set, determining ...", domain)
+ logger.debug("Software for domain='%s',path='%s' is not set, determining ...", domain, path)
software = determine_software(domain, path)
except network.exceptions as exception:
logger.warning("Exception '%s' during determining software type", type(exception))
instances.set_last_error(domain, exception)
logger.debug("Determined software='%s' for domain='%s'", software, domain)
+ elif software is None:
+ logger.debug("domain='%s' has unknown software or nodeinfo has recently being fetched", domain)
elif not isinstance(software, str):
- raise ValueError(f"Parameter software[]='{type(software)}' is not 'str'")
+ raise ValueError(f"Parameter software[]='{type(software)}' is not of type 'str'")
+
+ # Increase depth
+ _DEPTH = _DEPTH + 1
+ logger.debug("Checking if domain='%s' is registered ...", domain)
if not instances.is_registered(domain):
logger.debug("Adding new domain='%s',origin='%s',command='%s',path='%s',software='%s'", domain, origin, command, path, software)
instances.add(domain, origin, command, path, software)
+ logger.debug("software='%s'", software)
+ if software is not None and software_helper.is_relay(software):
+ logger.debug("software='%s' is a relay software - EXIT!", software)
+ _DEPTH = _DEPTH - 1
+ return
+
logger.debug("Updating last_instance_fetch for domain='%s' ...", domain)
instances.set_last_instance_fetch(domain)
peerlist = list()
- try:
- logger.debug("Fetching instances for domain='%s',software='%s',origin='%s'", domain, software, origin)
- peerlist = fetch_peers(domain, software, origin)
- except network.exceptions as exception:
- logger.warning("Cannot fetch peers from domain='%s': '%s'", domain, type(exception))
+ logger.debug("software='%s'", software)
+ if software is not None:
+ try:
+ logger.debug("Fetching instances for domain='%s',software='%s',origin='%s'", domain, software, origin)
+ peerlist = fetch_peers(domain, software, origin)
+ except network.exceptions as exception:
+ _DEPTH = _DEPTH - 1
+ raise exception
logger.debug("peerlist[]='%s'", type(peerlist))
if isinstance(peerlist, list):
logger.debug("Invoking instances.set_total_peerlist(%s,%d) ...", domain, len(peerlist))
instances.set_total_peers(domain, peerlist)
- logger.debug("peerlist[]='%s'", type(peerlist))
- if peerlist is None or len(peerlist) == 0:
- logger.warning("Cannot fetch peers: domain='%s'", domain)
+ logger.debug("Invoking cookies.clear(%s) ...", domain)
+ cookies.clear(domain)
- logger.debug("Invoking cookies.clear(%s) ...", domain)
- cookies.clear(domain)
+ logger.debug("peerlist[]='%s'", type(peerlist))
+ if peerlist is None:
+ logger.warning("Cannot fetch peers: domain='%s',software='%s'", domain, software)
+ if instances.has_pending(domain):
+ logger.debug("Flushing updates for domain='%s' ...", domain)
+ instances.update(domain)
+ _DEPTH = _DEPTH - 1
logger.debug("EXIT!")
return
+ elif len(peerlist) == 0:
+ logger.info("domain='%s' returned an empty peer list.", domain)
+ if instances.has_pending(domain):
+ logger.debug("Flushing updates for domain='%s' ...", domain)
+ instances.update(domain)
+
+ _DEPTH = _DEPTH - 1
+ logger.debug("domain='%s',software='%s' has an empty peer list returned - EXIT!", domain, software)
+ return
- logger.info("Checking %d instance(s) from domain='%s',software='%s' ...", len(peerlist), domain, software)
+ logger.info("Checking %d instance(s) from domain='%s',software='%s',depth=%d ...", len(peerlist), domain, software, _DEPTH)
for instance in peerlist:
- logger.debug("instance='%s'", instance)
- if instance is None:
- # Skip "None" types as tidup.domain() cannot parse them
+ logger.debug("instance[%s]='%s'", type(instance), instance)
+ if instance in [None, ""]:
+ logger.debug("instance[%s]='%s' is either None or empty - SKIPPED!", type(instance), instance)
continue
logger.debug("instance='%s' - BEFORE!", instance)
- instance = tidyup.domain(instance)
+ instance = tidyup.domain(instance) if isinstance(instance, str) and instance != "" else None
logger.debug("instance='%s' - AFTER!", instance)
- if instance == "":
- logger.warning("Empty instance after tidyup.domain(), domain='%s'", domain)
+ if instance in [None, ""]:
+ logger.warning("instance='%s' is empty after tidyup.domain(), domain='%s'", instance, domain)
continue
- elif not utils.is_domain_wanted(instance):
+ elif ".." in instance:
+ logger.warning("instance='%s' contains double-dot, removing ...", instance)
+ instance = instance.replace("..", ".")
+
+ logger.debug("instance='%s' - BEFORE!", instance)
+ instance = instance.encode("idna").decode("utf-8")
+ logger.debug("instance='%s' - AFTER!", instance)
+
+ if not domain_helper.is_wanted(instance):
logger.debug("instance='%s' is not wanted - SKIPPED!", instance)
continue
elif instance.find("/profile/") > 0 or instance.find("/users/") > 0 or (instances.is_registered(instance.split("/")[0]) and instance.find("/c/") > 0):
logger.debug("instance='%s' is a link to a tag - SKIPPED!", instance)
continue
elif not instances.is_registered(instance):
- logger.debug("Adding new instance='%s',domain='%s',command='%s'", instance, domain, command)
- instances.add(instance, domain, command)
-
- logger.debug("Invoking cookies.clear(%s) ...", domain)
- cookies.clear(domain)
+ logger.debug("Checking if domain='%s' has pending updates ...", domain)
+ if instances.has_pending(domain):
+ logger.debug("Flushing updates for domain='%s' ...", domain)
+ instances.update(domain)
+
+ logger.debug("instance='%s',origin='%s',_DEPTH=%d reached!", instance, origin, _DEPTH)
+ if _DEPTH <= config.get("max_crawl_depth") and len(peerlist) >= config.get("min_peers_length"):
+ logger.debug("Fetching instance='%s',origin='%s',command='%s',path='%s',_DEPTH=%d ...", instance, domain, command, path, _DEPTH)
+ fetch_instances(instance, domain, None, command, path)
+ else:
+ logger.debug("Adding instance='%s',domain='%s',command='%s',_DEPTH=%d ...", instance, domain, command, _DEPTH)
+ instances.add(instance, domain, command)
logger.debug("Checking if domain='%s' has pending updates ...", domain)
if instances.has_pending(domain):
logger.debug("Flushing updates for domain='%s' ...", domain)
- instances.update_data(domain)
+ instances.update(domain)
+ _DEPTH = _DEPTH - 1
logger.debug("EXIT!")
def fetch_peers(domain: str, software: str, origin: str) -> list:
logger.debug("domain='%s',software='%s',origin='%s' - CALLED!", domain, software, origin)
domain_helper.raise_on(domain)
- if not isinstance(software, str) and software is not None:
- raise ValueError(f"software[]='{type(software)}' is not 'str'")
+ if blacklist.is_blacklisted(domain):
+ raise Exception(f"domain='{domain}' is blacklisted but function was invoked")
+ elif not isinstance(software, str) and software is not None:
+ raise ValueError(f"Parameter software[]='{type(software)}' is not of type 'str'")
+ elif isinstance(software, str) and software == "":
+ raise ValueError("Parameter 'software' is empty")
+ elif software is not None and software_helper.is_relay(software):
+ raise ValueError(f"domain='{domain}' is of software='{software}' and isn't supported here.")
+ elif not isinstance(origin, str) and origin is not None:
+ raise ValueError(f"Parameter origin[]='{type(origin)}' is not of type 'str'")
+ elif isinstance(origin, str) and origin == "":
+ raise ValueError("Parameter 'origin' is empty")
if software == "misskey":
logger.debug("Invoking misskey.fetch_peers(%s) ...", domain)
logger.debug("Checking CSRF for domain='%s'", domain)
headers = csrf.determine(domain, dict())
except network.exceptions as exception:
- logger.warning("Exception '%s' during checking CSRF (fetch_peers,%s) - EXIT!", type(exception), __name__)
+ logger.warning("Exception '%s' during checking CSRF (fetch_peers,%s)", type(exception), __name__)
instances.set_last_error(domain, exception)
- return list()
- paths = [
- "/api/v1/instance/peers",
- "/api/v3/site",
- ]
+ logger.debug("Returning empty list ... - EXIT!")
+ return list()
# Init peers variable
peers = list()
- logger.debug("Checking %d paths ...", len(paths))
- for path in paths:
+ logger.debug("Checking %d API paths ...", len(_api_paths))
+ for path in _api_paths:
logger.debug("Fetching path='%s' from domain='%s',software='%s' ...", path, domain, software)
data = network.get_json_api(
domain,
path,
- headers,
- (config.get("connection_timeout"), config.get("read_timeout"))
+ headers=headers,
+ timeout=(config.get("connection_timeout"), config.get("read_timeout"))
)
- logger.debug("data[]='%s'", type(data))
+ logger.debug("data(%d)[]='%s'", len(data), type(data))
if "error_message" in data:
logger.debug("Was not able to fetch peers from path='%s',domain='%s' ...", path, domain)
instances.set_last_error(domain, data)
break
if not isinstance(peers, list):
- logger.warning("peers[]='%s' is not 'list', maybe bad API response?", type(peers))
+ logger.warning("peers[]='%s' is not of type 'list', maybe bad API response?", type(peers))
peers = list()
logger.debug("Invoking instances.set_total_peers(%s,%d) ...", domain, len(peers))
logger.debug("peers()=%d - EXIT!", len(peers))
return peers
-def fetch_nodeinfo(domain: str, path: str = None) -> dict:
- logger.debug("domain='%s',path='%s' - CALLED!", domain, path)
- domain_helper.raise_on(domain)
-
- if not isinstance(path, str) and path is not None:
- raise ValueError(f"Parameter path[]='{type(path)}' is not 'str'")
-
- logger.debug("Fetching nodeinfo from domain='%s' ...", domain)
- nodeinfo = fetch_wellknown_nodeinfo(domain)
-
- logger.debug("nodeinfo[%s](%d='%s'", type(nodeinfo), len(nodeinfo), nodeinfo)
- if "error_message" not in nodeinfo and "json" in nodeinfo and len(nodeinfo["json"]) > 0:
- logger.debug("Found nodeinfo[json]()=%d - EXIT!", len(nodeinfo['json']))
- return nodeinfo["json"]
-
- # No CSRF by default, you don't have to add network.api_headers by yourself here
- headers = tuple()
- data = dict()
-
- try:
- logger.debug("Checking CSRF for domain='%s'", domain)
- headers = csrf.determine(domain, dict())
- except network.exceptions as exception:
- logger.warning("Exception '%s' during checking CSRF (nodeinfo,%s) - EXIT!", type(exception), __name__)
- instances.set_last_error(domain, exception)
- return {
- "status_code" : 500,
- "error_message": f"exception[{type(exception)}]='{str(exception)}'",
- "exception" : exception,
- }
-
- request_paths = [
- "/nodeinfo/2.1.json",
- "/nodeinfo/2.1",
- "/nodeinfo/2.0.json",
- "/nodeinfo/2.0",
- "/nodeinfo/1.0.json",
- "/nodeinfo/1.0",
- "/api/v1/instance",
- ]
-
- for request in request_paths:
- logger.debug("request='%s'", request)
- http_url = f"http://{domain}{path}"
- https_url = f"https://{domain}{path}"
-
- logger.debug("path[%s]='%s',request='%s',http_url='%s',https_url='%s'", type(path), path, request, http_url, https_url)
- if path is None or path in [request, http_url, https_url]:
- logger.debug("Fetching request='%s' from domain='%s' ...", request, domain)
- if path in [http_url, https_url]:
- logger.debug("domain='%s',path='%s' has protocol in path, splitting ...", domain, path)
- components = urlparse(path)
- path = components.path
-
- data = network.get_json_api(
- domain,
- request,
- headers,
- (config.get("nodeinfo_connection_timeout"), config.get("nodeinfo_read_timeout"))
- )
-
- logger.debug("data[]='%s'", type(data))
- if "error_message" not in data and "json" in data:
- logger.debug("Success: request='%s'", request)
- instances.set_detection_mode(domain, "STATIC_CHECK")
- instances.set_nodeinfo_url(domain, request)
- break
-
- logger.warning("Failed fetching nodeinfo from domain='%s',status_code='%s',error_message='%s'", domain, data['status_code'], data['error_message'])
-
- logger.debug("data()=%d - EXIT!", len(data))
- return data
-
-def fetch_wellknown_nodeinfo(domain: str) -> dict:
- logger.debug("domain='%s' - CALLED!", domain)
- domain_helper.raise_on(domain)
-
- # "rel" identifiers (no real URLs)
- nodeinfo_identifier = [
- "https://nodeinfo.diaspora.software/ns/schema/2.1",
- "http://nodeinfo.diaspora.software/ns/schema/2.1",
- "https://nodeinfo.diaspora.software/ns/schema/2.0",
- "http://nodeinfo.diaspora.software/ns/schema/2.0",
- "https://nodeinfo.diaspora.software/ns/schema/1.1",
- "http://nodeinfo.diaspora.software/ns/schema/1.1",
- "https://nodeinfo.diaspora.software/ns/schema/1.0",
- "http://nodeinfo.diaspora.software/ns/schema/1.0",
- ]
-
- # No CSRF by default, you don't have to add network.api_headers by yourself here
- headers = tuple()
-
- try:
- logger.debug("Checking CSRF for domain='%s'", domain)
- headers = csrf.determine(domain, dict())
- except network.exceptions as exception:
- logger.warning("Exception '%s' during checking CSRF (fetch_wellknown_nodeinfo,%s) - EXIT!", type(exception), __name__)
- instances.set_last_error(domain, exception)
- return {
- "status_code" : 500,
- "error_message": type(exception),
- "exception" : exception,
- }
-
- logger.debug("Fetching .well-known info for domain='%s'", domain)
- data = network.get_json_api(
- domain,
- "/.well-known/nodeinfo",
- headers,
- (config.get("nodeinfo_connection_timeout"), config.get("nodeinfo_read_timeout"))
- )
-
- logger.debug("data[]='%s'", type(data))
- if "error_message" not in data:
- nodeinfo = data["json"]
-
- logger.debug("Marking domain='%s' as successfully handled ...", domain)
- instances.set_success(domain)
-
- logger.debug("Found entries: nodeinfo()=%d,domain='%s'", len(nodeinfo), domain)
- if "links" in nodeinfo:
- logger.debug("Found nodeinfo[links]()=%d record(s),", len(nodeinfo["links"]))
- for niid in nodeinfo_identifier:
- data = dict()
-
- logger.debug("Checking niid='%s' ...", niid)
- for link in nodeinfo["links"]:
- logger.debug("link[%s]='%s'", type(link), link)
- if not isinstance(link, dict) or not "rel" in link:
- logger.warning("link[]='%s' is not 'dict' or no element 'rel' found", type(link))
- elif link["rel"] == niid:
- # Default is that 'href' has a complete URL, but some hosts don't send that
- logger.debug("link[href]='%s' matches niid='%s'", link["href"], niid)
- url = link["href"]
- components = urlparse(link["href"])
-
- logger.debug("components[%s]='%s'", type(components), components)
- if components.scheme == "" and components.netloc == "":
- logger.debug("link[href]='%s' has no scheme and host name in it, prepending from domain='%s'", link['href'], domain)
- url = f"https://{domain}{url}"
- components = urlparse(url)
-
- if not utils.is_domain_wanted(components.netloc):
- logger.debug("components.netloc='%s' is not wanted - SKIPPED!", components.netloc)
- continue
-
- logger.debug("Fetching nodeinfo from url='%s' ...", url)
- data = network.fetch_api_url(
- url,
- (config.get("connection_timeout"), config.get("read_timeout"))
- )
-
- logger.debug("link[href]='%s',data[]='%s'", link["href"], type(data))
- if "error_message" not in data and "json" in data:
- logger.debug("Found JSON data()=%d,link[href]='%s'", len(data), link["href"])
- instances.set_detection_mode(domain, "AUTO_DISCOVERY")
- instances.set_nodeinfo_url(domain, link["href"])
-
- logger.debug("Marking domain='%s' as successfully handled ...", domain)
- instances.set_success(domain)
- break
- else:
- logger.debug("Setting last error for domain='%s',data[]='%s'", domain, type(data))
- instances.set_last_error(domain, data)
-
- logger.debug("data()=%d", len(data))
- if "error_message" not in data and "json" in data:
- logger.debug("Auto-discovery successful: domain='%s'", domain)
- break
- else:
- logger.warning("nodeinfo does not contain 'links': domain='%s'", domain)
-
- logger.debug("Returning data[]='%s' - EXIT!", type(data))
- return data
-
def fetch_generator_from_path(domain: str, path: str = "/") -> str:
- logger.debug("domain(%d)='%s',path='%s' - CALLED!", len(domain), domain, path)
+ logger.debug("domain='%s',path='%s' - CALLED!", domain, path)
domain_helper.raise_on(domain)
- if not isinstance(path, str):
- raise ValueError(f"path[]='{type(path)}' is not 'str'")
+ if blacklist.is_blacklisted(domain):
+ raise Exception(f"domain='{domain}' is blacklisted but function was invoked")
+ elif not isinstance(path, str):
+ raise ValueError(f"path[]='{type(path)}' is not of type 'str'")
elif path == "":
raise ValueError("Parameter 'path' is empty")
+ elif not path.startswith("/"):
+ raise ValueError(f"path='{path}' does not start with / but should")
- logger.debug("domain='%s',path='%s' - CALLED!", domain, path)
software = None
logger.debug("Fetching path='%s' from domain='%s' ...", path, domain)
- response = network.fetch_response(domain, path, network.web_headers, (config.get("connection_timeout"), config.get("read_timeout")))
+ response = network.fetch_response(
+ domain,
+ path,
+ headers=network.web_headers,
+ timeout=(config.get("connection_timeout"), config.get("read_timeout")),
+ allow_redirects=True
+ )
logger.debug("response.ok='%s',response.status_code=%d,response.text()=%d", response.ok, response.status_code, len(response.text))
- if response.ok and response.status_code < 300 and response.text.find("<html") > 0:
+ if ((response.ok and response.status_code == 200) or response.status_code == 410) and response.text.find("<html") > 0 and domain_helper.is_in_url(domain, response.url.split("#")[0]):
logger.debug("Parsing response.text()=%d Bytes ...", len(response.text))
doc = bs4.BeautifulSoup(response.text, "html.parser")
logger.debug("doc[]='%s'", type(doc))
+ platform = doc.find("meta", {"property": "og:platform"})
generator = doc.find("meta", {"name" : "generator"})
site_name = doc.find("meta", {"property": "og:site_name"})
+ app_name = doc.find("meta", {"name" : "application-name"})
+
+ logger.debug("generator[]='%s',site_name[]='%s',platform[]='%s',app_name[]='%s'", type(generator), type(site_name), type(platform), type(app_name))
+ if isinstance(platform, bs4.element.Tag) and isinstance(platform.get("content"), str) and platform.get("content") != "":
+ logger.debug("Found property=og:platform, domain='%s'", domain)
+ software = tidyup.domain(platform.get("content"))
+ logger.debug("software[%s]='%s' after tidyup.domain() ...", type(software), software)
- logger.debug("generator[]='%s',site_name[]='%s'", type(generator), type(site_name))
- if isinstance(generator, bs4.element.Tag) and isinstance(generator.get("content"), str):
+ if software is not None and software != "":
+ logger.debug("domain='%s' has og:platform='%s' - Setting detection_mode=PLATFORM ...", domain, software)
+ instances.set_detection_mode(domain, "PLATFORM")
+ elif isinstance(generator, bs4.element.Tag) and isinstance(generator.get("content"), str) and generator.get("content") != "":
logger.debug("Found generator meta tag: domain='%s'", domain)
software = tidyup.domain(generator.get("content"))
logger.debug("software[%s]='%s'", type(software), software)
if software is not None and software != "":
- logger.info("domain='%s' is generated by software='%s'", domain, software)
+ logger.info("domain='%s' is generated by software='%s' - Setting detection_mode=GENERATOR ...", domain, software)
instances.set_detection_mode(domain, "GENERATOR")
- elif isinstance(site_name, bs4.element.Tag) and isinstance(site_name.get("content"), str):
+ elif isinstance(app_name, bs4.element.Tag) and isinstance(app_name.get("content"), str) and app_name.get("content") != "":
+ logger.debug("Found property=og:app_name, domain='%s'", domain)
+ software = tidyup.domain(app_name.get("content"))
+
+ logger.debug("software[%s]='%s'", type(software), software)
+ if software is not None and software != "":
+ logger.debug("domain='%s' has application-name='%s' - Setting detection_mode=app_name ...", domain, software)
+ instances.set_detection_mode(domain, "APP_NAME")
+ elif isinstance(site_name, bs4.element.Tag) and isinstance(site_name.get("content"), str) and site_name.get("content") != "":
logger.debug("Found property=og:site_name, domain='%s'", domain)
software = tidyup.domain(site_name.get("content"))
logger.debug("software[%s]='%s'", type(software), software)
if software is not None and software != "":
- logger.info("domain='%s' has og:site_name='%s'", domain, software)
+ logger.debug("domain='%s' has og:site_name='%s' - Setting detection_mode=SITE_NAME ...", domain, software)
instances.set_detection_mode(domain, "SITE_NAME")
+ elif not domain_helper.is_in_url(domain, response.url.split("#")[0]):
+ logger.warning("domain='%s' doesn't match response.url='%s', maybe redirect to other domain?", domain, response.url)
+
+ components = urlparse(response.url)
+ domain2 = components.netloc.lower().split(":")[0]
+
+ logger.debug("domain2='%s'", domain2)
+ if not domain_helper.is_wanted(domain2):
+ logger.debug("domain2='%s' is not wanted - EXIT!", domain2)
+ return None
+ elif not instances.is_registered(domain2):
+ logger.info("components.netloc='%s' is not registered, adding ...", components.netloc)
+ instances.add(domain2, domain, "redirect_target")
+
+ message = f"Redirect from domain='{domain}' to response.url='{response.url}'"
+ instances.set_last_error(domain, message)
+ instances.set_software(domain, None)
+ instances.set_detection_mode(domain, None)
+ instances.set_nodeinfo_url(domain, None)
+
+ raise requests.exceptions.TooManyRedirects(message)
logger.debug("software[]='%s'", type(software))
if isinstance(software, str) and software == "":
logger.debug("software='%s' may contain a version number, domain='%s', removing it ...", software, domain)
software = version.remove(software)
- logger.debug("software[]='%s'", type(software))
+ logger.debug("software[%s]='%s'", type(software), software)
if isinstance(software, str) and "powered by " in software:
logger.debug("software='%s' has 'powered by' in it", software)
- software = version.remove(version.strip_powered_by(software))
+ software = version.remove(software_helper.strip_powered_by(software))
elif isinstance(software, str) and " hosted on " in software:
logger.debug("software='%s' has 'hosted on' in it", software)
- software = version.remove(version.strip_hosted_on(software))
+ software = version.remove(software_helper.strip_hosted_on(software))
elif isinstance(software, str) and " by " in software:
logger.debug("software='%s' has ' by ' in it", software)
- software = version.strip_until(software, " by ")
+ software = software_helper.strip_until(software, " by ")
elif isinstance(software, str) and " see " in software:
logger.debug("software='%s' has ' see ' in it", software)
- software = version.strip_until(software, " see ")
+ software = software_helper.strip_until(software, " see ")
- logger.debug("software='%s' - EXIT!", software)
+ logger.debug("software[%s]='%s' - EXIT!", type(software), software)
return software
def determine_software(domain: str, path: str = None) -> str:
- logger.debug("domain(%d)='%s',path='%s' - CALLED!", len(domain), domain, path)
+ logger.debug("domain='%s',path='%s' - CALLED!", domain, path)
domain_helper.raise_on(domain)
- if not isinstance(path, str) and path is not None:
- raise ValueError(f"Parameter path[]='{type(path)}' is not 'str'")
+ if blacklist.is_blacklisted(domain):
+ raise Exception(f"domain='{domain}' is blacklisted but function was invoked")
+ elif not isinstance(path, str) and path is not None:
+ raise ValueError(f"Parameter path[]='{type(path)}' is not of type 'str'")
+ elif path is not None and not path.startswith("/"):
+ raise ValueError(f"path='{path}' does not start with a slash")
- logger.debug("Determining software for domain='%s',path='%s'", domain, path)
+ logger.debug("Fetching nodeinfo from domain='%s',path='%s' ...", domain, path)
+ data = nodeinfo.fetch(domain, path)
software = None
- logger.debug("Fetching nodeinfo from domain='%s' ...", domain)
- data = fetch_nodeinfo(domain, path)
-
- logger.debug("data[]='%s'", type(data))
+ logger.debug("data[%s]='%s'", type(data), data)
if "exception" in data:
# Continue raising it
logger.debug("data()=%d contains exception='%s' - raising ...", len(data), type(data["exception"]))
logger.debug("domain='%s',path='%s',data[json] found ...", domain, path)
data = data["json"]
else:
- logger.debug("JSON response from domain='%s' does not include [software][name], fetching / ...", domain)
+ logger.debug("Auto-detection for domain='%s' was failing, fetching / ...", domain)
software = fetch_generator_from_path(domain)
logger.debug("Generator for domain='%s' is: '%s'", domain, software)
if "status" in data and data["status"] == "error" and "message" in data:
- logger.warning("JSON response is an error: '%s'", data["message"])
+ logger.warning("JSON response is an error: '%s' - Resetting detection_mode,nodeinfo_url ...", data["message"])
instances.set_last_error(domain, data["message"])
instances.set_detection_mode(domain, None)
instances.set_nodeinfo_url(domain, None)
software = data["software"]["name"]
logger.debug("software[%s]='%s' - FOUND!", type(software), software)
elif "message" in data:
- logger.warning("JSON response contains only a message: '%s'", data["message"])
+ logger.warning("JSON response contains only a message: '%s' - Resetting detection_mode,nodeinfo_url ...", data["message"])
instances.set_last_error(domain, data["message"])
instances.set_detection_mode(domain, None)
instances.set_nodeinfo_url(domain, None)
+
+ logger.debug("Invoking fetch_generator_from_path(%s) ...", domain)
software = fetch_generator_from_path(domain)
logger.debug("Generator for domain='%s' is: '%s'", domain, software)
+ elif "server" in data and "software" in data["server"]:
+ logger.debug("Found data[server][software]='%s' for domain='%s'", data["server"]["software"].lower(), domain)
+ software = data["server"]["software"].lower()
+ logger.debug("Detected software for domain='%s' is: '%s'", domain, software)
elif "software" not in data or "name" not in data["software"]:
- logger.debug("JSON response from domain='%s' does not include [software][name], fetching / ...", domain)
+ logger.debug("JSON response from domain='%s' does not include [software][name] - Resetting detection_mode,nodeinfo_url ...", domain)
instances.set_detection_mode(domain, None)
instances.set_nodeinfo_url(domain, None)
+
+ logger.debug("Invoking fetch_generator_from_path(%s) ...", domain)
software = fetch_generator_from_path(domain)
logger.debug("Generator for domain='%s' is: '%s'", domain, software)
logger.debug("software[%s]='%s'", type(software), software)
- if software is None:
+ if software in [None, ""]:
logger.debug("Returning None - EXIT!")
return None
+ logger.debug("Setting original software='%s' for domain='%s' ...", software, domain)
+ instances.set_original_software(domain, software)
+
logger.debug("software='%s'- BEFORE!", software)
software = software_helper.alias(software)
logger.debug("software['%s']='%s' - AFTER!", type(software), software)
- if str(software) == "":
+ if software in [None, ""]:
logger.debug("software for domain='%s' was not detected, trying generator ...", domain)
software = fetch_generator_from_path(domain)
elif len(str(software)) > 0 and ("." in software or " " in software):
logger.debug("software[]='%s'", type(software))
if isinstance(software, str) and "powered by" in software:
logger.debug("software='%s' has 'powered by' in it", software)
- software = version.remove(version.strip_powered_by(software))
+ software = version.remove(software_helper.strip_powered_by(software))
+
+ software = software.strip()
- logger.debug("software='%s' - EXIT!", software)
+ logger.debug("software[%s]='%s' - EXIT!", type(software), software)
return software
def find_domains(tag: bs4.element.Tag) -> list:
logger.debug("tag[]='%s' - CALLED!", type(tag))
+
if not isinstance(tag, bs4.element.Tag):
raise ValueError(f"Parameter tag[]='{type(tag)}' is not type of bs4.element.Tag")
elif len(tag.select("tr")) == 0:
logger.debug("domain='%s',reason='%s'", domain, reason)
- if not utils.is_domain_wanted(domain):
+ if not domain_helper.is_wanted(domain):
logger.debug("domain='%s' is blacklisted - SKIPPED!", domain)
continue
elif domain == "gab.com/.ai, develop.gab.com":
- logger.debug("Multiple domains detected in one row")
+ logger.debug("Multiple gab.com domains detected in one row")
domains.append({
"domain": "gab.com",
"reason": reason,
def add_peers(rows: dict) -> list:
logger.debug("rows[]='%s' - CALLED!", type(rows))
+
if not isinstance(rows, dict):
- raise ValueError(f"Parameter rows[]='{type(rows)}' is not 'dict'")
+ raise ValueError(f"Parameter rows[]='{type(rows)}' is not of type 'dict'")
+ elif len(rows) == 0:
+ raise ValueError("Parameter 'rows' is empty")
peers = list()
for key in ["linked", "allowed", "blocked"]:
- logger.debug("Checking key='%s'", key)
+ logger.debug("key='%s'", key)
if key not in rows or rows[key] is None:
logger.debug("Cannot find key='%s' or it is NoneType - SKIPPED!", key)
continue
logger.debug("Adding %d peer(s) to peers list ...", len(rows[key]))
for peer in rows[key]:
logger.debug("peer[%s]='%s' - BEFORE!", type(peer), peer)
- if peer is None or peer == "":
+ if peer in [None, ""]:
logger.debug("peer is empty - SKIPPED")
continue
elif isinstance(peer, dict) and "domain" in peer:
raise ValueError(f"peer[]='{type(peer)}' is not supported,key='{key}'")
logger.debug("peer[%s]='%s' - AFTER!", type(peer), peer)
- if not utils.is_domain_wanted(peer):
+ if not domain_helper.is_wanted(peer):
logger.debug("peer='%s' is not wanted - SKIPPED!", peer)
continue
logger.debug("peers()=%d - EXIT!", len(peers))
return peers
+
+def fetch_blocks(domain: str) -> list:
+ logger.debug("domain='%s' - CALLED!", domain)
+ domain_helper.raise_on(domain)
+
+ if blacklist.is_blacklisted(domain):
+ raise Exception(f"domain='{domain}' is blacklisted but function was invoked")
+ elif not instances.is_registered(domain):
+ raise Exception(f"domain='{domain}' is not registered but function is invoked.")
+
+ # Init block list
+ blocklist = list()
+
+ # No CSRF by default, you don't have to add network.api_headers by yourself here
+ headers = tuple()
+
+ try:
+ logger.debug("Checking CSRF for domain='%s'", domain)
+ headers = csrf.determine(domain, dict())
+ except network.exceptions as exception:
+ logger.warning("Exception '%s' during checking CSRF (fetch_blocks,%s)", type(exception), __name__)
+ instances.set_last_error(domain, exception)
+
+ logger.debug("Returning empty list ... - EXIT!")
+ return list()
+
+ try:
+ # json endpoint for newer mastodongs
+ logger.info("Fetching domain_blocks from domain='%s' ...", domain)
+ data = network.get_json_api(
+ domain,
+ "/api/v1/instance/domain_blocks",
+ headers=headers,
+ timeout=(config.get("connection_timeout"), config.get("read_timeout"))
+ )
+ rows = list()
+
+ logger.debug("data(%d)[]='%s'", len(data), type(data))
+ if "error_message" in data:
+ logger.debug("Was not able to fetch domain_blocks from domain='%s': status_code=%d,error_message='%s'", domain, data['status_code'], data['error_message'])
+ instances.set_last_error(domain, data)
+
+ logger.debug("blocklist()=%d - EXIT!", len(blocklist))
+ return blocklist
+ elif "json" in data and "error" in data["json"]:
+ logger.warning("JSON API returned error message: '%s'", data["json"]["error"])
+ instances.set_last_error(domain, data)
+
+ logger.debug("blocklist()=%d - EXIT!", len(blocklist))
+ return blocklist
+ else:
+ # Getting blocklist
+ rows = data["json"]
+
+ logger.debug("Marking domain='%s' as successfully handled ...", domain)
+ instances.set_success(domain)
+
+ logger.debug("rows(%d)[]='%s'", len(rows), type(rows))
+ if len(rows) > 0:
+ logger.debug("Checking %d entries from domain='%s' ...", len(rows), domain)
+ for block in rows:
+ # Check type
+ logger.debug("block[]='%s'", type(block))
+ if not isinstance(block, dict):
+ logger.debug("block[]='%s' is of type 'dict' - SKIPPED!", type(block))
+ continue
+ elif "domain" not in block:
+ logger.warning("block()=%d does not contain element 'domain' - SKIPPED!", len(block))
+ continue
+ elif "severity" not in block:
+ logger.warning("block()=%d does not contain element 'severity' - SKIPPED!", len(block))
+ continue
+ elif block["severity"] in ["accept", "accepted"]:
+ logger.debug("block[domain]='%s' has unwanted severity level '%s' - SKIPPED!", block["domain"], block["severity"])
+ continue
+ elif "digest" in block and not validators.hashes.sha256(block["digest"]):
+ logger.warning("block[domain]='%s' has invalid block[digest]='%s' - SKIPPED!", block["domain"], block["digest"])
+ continue
+
+ reason = tidyup.reason(block["comment"]) if "comment" in block and block["comment"] is not None and block["comment"] != "" else None
+
+ logger.debug("Appending blocker='%s',blocked='%s',reason='%s',block_level='%s' ...", domain, block["domain"], reason, block["severity"])
+ blocklist.append({
+ "blocker" : domain,
+ "blocked" : block["domain"],
+ "digest" : block["digest"] if "digest" in block else None,
+ "reason" : reason,
+ "block_level": blocks.alias_block_level(block["severity"]),
+ })
+ else:
+ logger.debug("domain='%s' has no block list", domain)
+
+ except network.exceptions as exception:
+ logger.warning("domain='%s',exception[%s]='%s'", domain, type(exception), str(exception))
+ instances.set_last_error(domain, exception)
+
+ logger.debug("blocklist()=%d - EXIT!", len(blocklist))
+ return blocklist
projects / fba.git / blobdiff