import requests
import validators
-from fba import csrf
-
+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):
global _DEPTH
logger.debug("domain='%s',origin='%s',software='%s',command='%s',path='%s',_DEPTH=%d - CALLED!", domain, origin, software, command, path, _DEPTH)
- _DEPTH = _DEPTH + 1
domain_helper.raise_on(domain)
- if not isinstance(origin, str) and origin is not None:
+ 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 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 of type 'str'")
- elif not isinstance(path, str) and path is not None:
- raise ValueError(f"Parameter path[]='{type(path)}' 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)
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',software='%s': '%s'", domain, software, type(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("Invoking cookies.clear(%s) ...", domain)
+ cookies.clear(domain)
+
logger.debug("peerlist[]='%s'", type(peerlist))
- if peerlist is None or len(peerlist) == 0:
+ 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_data(domain)
-
- logger.debug("Invoking cookies.clear(%s) ...", domain)
- cookies.clear(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 or instance == "":
+ 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 ".." 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("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)
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("Adding instance='%s',domain='%s',command='%s',_DEPTH=%d ...", instance, domain, command, _DEPTH)
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_data(domain)
+ instances.update(domain)
_DEPTH = _DEPTH - 1
logger.debug("EXIT!")
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:
+ 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 == "":
logger.debug("Returning empty list ... - EXIT!")
return list()
- paths = [
- "/api/v1/instance/peers",
- "/api/v3/site",
- ]
-
# 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)
logger.debug("domain='%s',path='%s' - CALLED!", domain, path)
domain_helper.raise_on(domain)
- if not isinstance(path, 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")),
+ 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 and domain_helper.is_in_url(domain, response.url):
+ 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"})
- platform = doc.find("meta", {"property": "og:platform"})
+ app_name = doc.find("meta", {"name" : "application-name"})
- logger.debug("generator[]='%s',site_name[]='%s',platform[]='%s'", type(generator), type(site_name), type(platform))
- if isinstance(generator, bs4.element.Tag) and isinstance(generator.get("content"), str):
+ 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)
+
+ 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"))
if software is not None and software != "":
logger.info("domain='%s' is generated by software='%s' - Setting detection_mode=GENERATOR ...", domain, software)
instances.set_detection_mode(domain, "GENERATOR")
- elif isinstance(platform, bs4.element.Tag) and isinstance(platform.get("content"), str):
- logger.debug("Found property=og:platform, domain='%s'", domain)
- software = tidyup.domain(platform.get("content"))
+ 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 og:platform='%s' - Setting detection_mode=PLATFORM ...", domain, software)
- instances.set_detection_mode(domain, "PLATFORM")
- elif isinstance(site_name, bs4.element.Tag) and isinstance(site_name.get("content"), str):
+ 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"))
if software is not None and 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):
+ 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("components[]='%s'", type(components))
- if not instances.is_registered(components.netloc):
+ 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)
- fetch_instances(components.netloc, domain, None, "fetch_generator")
+ instances.add(domain2, domain, "redirect_target")
message = f"Redirect from domain='{domain}' to response.url='{response.url}'"
instances.set_last_error(domain, message)
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(software_helper.strip_powered_by(software))
logger.debug("software='%s' has ' see ' in it", software)
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='%s',path='%s' - CALLED!", domain, path)
domain_helper.raise_on(domain)
- if not isinstance(path, str) and path is not None:
+ 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 = nodeinfo.fetch_nodeinfo(domain, path)
-
logger.debug("data[%s]='%s'", type(data), data)
if "exception" in data:
# Continue raising it
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' has 'powered by' in it", software)
software = version.remove(software_helper.strip_powered_by(software))
- logger.debug("software='%s' - EXIT!", software)
+ software = software.strip()
+
+ 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' 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 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:
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