# You should have received a copy of the GNU Affero General Public License
# along with this program. If not, see <https://www.gnu.org/licenses/>.
-import inspect
import logging
-
-import bs4
import validators
-from fba import csrf
-from fba import fba
+import bs4
-from fba.helpers import blacklist
from fba.helpers import config
+from fba.helpers import domain as domain_helper
from fba.helpers import tidyup
+from fba.http import federation
from fba.http import network
from fba.models import blocks
}
def fetch_blocks_from_about(domain: str) -> dict:
- logger.debug("domain(%d)='%s' - CALLED!", len(domain), domain)
- if not isinstance(domain, str):
- raise ValueError(f"Parameter domain[]='{type(domain)}' is not 'str'")
- elif domain == "":
- raise ValueError("Parameter 'domain' is empty")
- elif domain.lower() != domain:
- raise ValueError(f"Parameter domain='{domain}' must be all lower-case")
- elif not validators.domain(domain.split("/")[0]):
- raise ValueError(f"domain='{domain}' is not a valid domain")
- elif domain.endswith(".arpa"):
- raise ValueError(f"domain='{domain}' is a domain for reversed IP addresses, please don't crawl them!")
- elif domain.endswith(".tld"):
- raise ValueError(f"domain='{domain}' is a fake domain, please don't crawl them!")
+ logger.debug("domain='%s' - CALLED!", domain)
+ domain_helper.raise_on(domain)
- logger.debug("Fetching mastodon blocks from domain:", domain)
+ if not instances.is_registered(domain):
+ raise Exception(f"domain='{domain}' is not registered but function is invoked.")
+
+ logger.debug("Fetching mastodon blocks from domain='%s'", domain)
doc = None
for path in ["/about/more", "/about"]:
try:
- logger.debug(f"Fetching path='{path}' from domain='{domain}' ...")
+ logger.debug("Fetching path='%s' from domain='%s' ...", path, domain)
doc = bs4.BeautifulSoup(
network.fetch_response(
domain,
)
if len(doc.find_all("h3")) > 0:
- logger.debug(f"path='{path}' had some headlines - BREAK!")
+ logger.debug("path='%s' had some headlines - BREAK!", path)
break
except network.exceptions as exception:
- logger.warning(f"Cannot fetch from domain='{domain}',exception='{type(exception)}'")
+ logger.warning("Cannot fetch from domain='%s',exception='%s'", domain, type(exception))
instances.set_last_error(domain, exception)
break
"Silenced servers" : [],
}
- logger.debug("doc[]='%'", type(doc))
+ logger.debug("doc[]='%s'", type(doc))
if doc is None:
- logger.warning(f"Cannot fetch any /about pages for domain='{domain}' - EXIT!")
- return blocklist
+ logger.warning("Cannot fetch any /about pages for domain='%s' - EXIT!", domain)
+ return list()
for header in doc.find_all("h3"):
header_text = tidyup.reason(header.text)
logger.debug("header_text='%s'", header_text)
if header_text in language_mapping:
- logger.debug("header_text='%s'", header_text)
+ logger.debug("Translating header_text='%s' ...", header_text)
header_text = language_mapping[header_text]
else:
- logger.warning(f"header_text='{header_text}' not found in language mapping table")
+ logger.warning("header_text='%s' not found in language mapping table", header_text)
if header_text in blocklist or header_text.lower() in blocklist:
# replaced find_next_siblings with find_all_next to account for instances that e.g. hide lists in dropdown menu
for line in header.find_all_next("table")[0].find_all("tr")[1:]:
+ domain = line.find("span").text
+ hash = line.find("span")["title"][9:]
+ reason = line.find_all("td")[1].text
+
+ logger.debug("domain='%s',reason='%s' - BEFORE!", domain, reason)
+ domain = tidyup.domain(domain) if domain != "" else None
+ reason = tidyup.reason(reason) if reason != "" else None
+
+ logger.debug("domain='%s',reason='%s' - AFTER!", domain, reason)
+ if domain is None or domain == "":
+ logger.warning("domain='%s' is empty,line='%s' - SKIPPED!", domain, line)
+ continue
+
+ logger.debug("Appending domain='%s',hash='%s',reason='%s' to blocklist header_text='%s' ...", domain, hash, reason, blocklist)
blocklist[header_text].append({
- "domain": tidyup.domain(line.find("span").text),
- "hash" : tidyup.domain(line.find("span")["title"][9:]),
- "reason": tidyup.reason(line.find_all("td")[1].text),
+ "domain": domain,
+ "hash" : hash,
+ "reason": reason,
})
else:
- logger.warning(f"header_text='{header_text}' not found in blocklist()={len(blocklist)}")
+ logger.warning("header_text='%s' not found in blocklist()=%d", header_text, len(blocklist))
- logger.debug("Returning blocklist for domain:", domain)
+ logger.debug("Returning blocklist for domain='%s' - EXIT!", domain)
return {
"reject" : blocklist["Suspended servers"],
"media_removal" : blocklist["Filtered media"],
"followers_only": blocklist["Limited servers"] + blocklist["Silenced servers"],
}
-def fetch_blocks(domain: str, origin: str, nodeinfo_url: str):
- logger.debug(f"domain='{domain}',origin='{origin}',nodeinfo_url='{nodeinfo_url}' - CALLED!")
- if not isinstance(domain, str):
- raise ValueError(f"Parameter domain[]='{type(domain)}' is not 'str'")
- elif domain == "":
- raise ValueError("Parameter 'domain' is empty")
- elif domain.lower() != domain:
- raise ValueError(f"Parameter domain='{domain}' must be all lower-case")
- elif not validators.domain(domain.split("/")[0]):
- raise ValueError(f"domain='{domain}' is not a valid domain")
- elif domain.endswith(".arpa"):
- raise ValueError(f"domain='{domain}' is a domain for reversed IP addresses, please don't crawl them!")
- elif domain.endswith(".tld"):
- raise ValueError(f"domain='{domain}' is a fake domain, please don't crawl them!")
- elif not isinstance(origin, str) and origin is not None:
- raise ValueError(f"Parameter origin[]='{type(origin)}' is not 'str'")
- elif origin == "":
- raise ValueError("Parameter 'origin' is empty")
- elif not isinstance(nodeinfo_url, str):
- raise ValueError(f"Parameter nodeinfo_url[]='{type(nodeinfo_url)}' is not 'str'")
- elif nodeinfo_url == "":
- raise ValueError("Parameter 'nodeinfo_url' is empty")
-
- # No CSRF by default, you don't have to add network.api_headers by yourself here
- headers = tuple()
+def fetch_blocks(domain: str) -> list:
+ logger.debug("domain='%s' - CALLED!", domain)
+ domain_helper.raise_on(domain)
- try:
- logger.debug(f"Checking CSRF for domain='{domain}'")
- headers = csrf.determine(domain, dict())
- except network.exceptions as exception:
- logger.warning(f"Exception '{type(exception)}' during checking CSRF (fetch_blocks,{__name__}) - EXIT!")
- instances.set_last_error(domain, exception)
- return
+ if not instances.is_registered(domain):
+ raise Exception(f"domain='{domain}' is not registered but function is invoked.")
- try:
- # json endpoint for newer mastodongs
- found_blocks = list()
- blocklist = list()
+ blocklist = list()
- rows = {
- "reject" : [],
- "media_removal" : [],
- "followers_only": [],
- "report_removal": [],
- }
+ logger.debug("Invoking federation.fetch_blocks(%s) ...", domain)
+ rows = federation.fetch_blocks(domain)
- logger.debug("Querying API domain_blocks:", domain)
- data = network.get_json_api(
- domain,
- "/api/v1/instance/domain_blocks",
- headers,
- (config.get("connection_timeout"), config.get("read_timeout"))
- )
-
- logger.debug("data[]='%s'", type(data))
- if "error_message" in data:
- logger.debug(f"Was not able to fetch domain_blocks from domain='{domain}': status_code='{data['status_code']}',error_message='{data['error_message']}'")
- instances.set_last_error(domain, data)
- return
- elif "json" in data and "error" in data["json"]:
- logger.warning(f"JSON API returned error message: '{data['json']['error']}'")
- instances.set_last_error(domain, data)
- return
- else:
- # Getting blocklist
- blocklist = data["json"]
+ logger.debug("rows[%s]()=%d", type(rows), len(rows))
+ if len(rows) == 0:
+ logger.debug("domain='%s' has returned zero rows, trying /about/more page ...", domain)
+ rows = fetch_blocks_from_about(domain)
- if len(blocklist) > 0:
- logger.info("Checking %d entries from domain='%s' ...", len(blocklist), domain)
- for block in blocklist:
- # Check type
- logger.debug(f"block[]='{type(block)}'")
- if not isinstance(block, dict):
- logger.debug(f"block[]='{type(block)}' is of type 'dict' - SKIPPED!")
- continue
-
- # Map block -> entry
- logger.debug(f"block[{type(block)}]='{block}'")
- entry = {
- "domain": block["domain"],
- "hash" : block["digest"],
- "reason": block["comment"] if "comment" in block else None
- }
-
- logger.debug("severity,domain,hash,comment:", block['severity'], block['domain'], block['digest'], block['comment'])
- if block['severity'] == 'suspend':
- logger.debug(f"Adding entry='{entry}' with severity='{block['severity']}' ...")
- rows['reject'].append(entry)
- elif block['severity'] == 'silence':
- logger.debug(f"Adding entry='{entry}' with severity='{block['severity']}' ...")
- rows['followers_only'].append(entry)
- elif block['severity'] == 'reject_media':
- logger.debug(f"Adding entry='{entry}' with severity='{block['severity']}' ...")
- rows['media_removal'].append(entry)
- elif block['severity'] == 'reject_reports':
- logger.debug(f"Adding entry='{entry}' with severity='{block['severity']}' ...")
- rows['report_removal'].append(entry)
- else:
- logger.warning(f"Unknown severity='{block['severity']}', domain='{block['domain']}'")
- else:
- logger.debug(f"domain='{domain}' has returned zero rows, trying /about/more page ...")
- rows = fetch_blocks_from_about(domain)
-
- logger.info("Checking %d entries from domain='%s' ...", len(rows.items()), domain)
- for block_level, blocklist in rows.items():
- logger.debug("domain,block_level,blocklist():", domain, block_level, len(blocklist))
- block_level = tidyup.domain(block_level)
-
- logger.debug("AFTER-block_level:", block_level)
- if block_level == "":
- logger.warning("block_level is empty, domain:", domain)
+ logger.debug("rows[%s]()=%d", type(rows), len(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 block_level == "accept":
- logger.debug(f"domain='{domain}' skipping block_level='accept'")
+ elif "domain" not in block:
+ logger.debug("block='%s'", block)
+ logger.warning("block()=%d does not contain element 'domain' - SKIPPED!", len(block))
+ continue
+ elif not domain_helper.is_wanted(block["domain"]):
+ logger.debug("block[domain]='%s' is not wanted - SKIPPED!", block["domain"])
+ 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
- logger.debug(f"Checking {len(blocklist)} entries from domain='{domain}',block_level='{block_level}' ...")
- for block in blocklist:
- logger.debug(f"block[]='{type(block)}'")
- blocked, blocked_hash, reason = block.values()
- logger.debug(f"blocked='{blocked}',blocked_hash='{blocked_hash}',reason='{reason}':")
- blocked = tidyup.domain(blocked)
- reason = tidyup.reason(reason) if reason is not None and reason != "" else None
- logger.debug(f"blocked='{blocked}',reason='{reason}' - AFTER!")
-
- if blocked == "":
- logger.warning("blocked is empty, domain='%s'", domain)
- continue
- elif blacklist.is_blacklisted(blocked):
- logger.debug("blocked='%s' is blacklisted - SKIPPED!", blocked)
- continue
- elif blocked.count("*") > 0:
- # Doing the hash search for instance names as well to tidy up DB
- row = instances.deobscure("*", blocked, blocked_hash)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning(f"Cannot deobsfucate blocked='{blocked}',blocked_hash='{blocked_hash}' - SKIPPED!")
- continue
-
- logger.debug("Updating domain: ", row[0])
- blocked = row[0]
- origin = row[1]
- nodeinfo_url = row[2]
- elif blocked.count("?") > 0:
- # Doing the hash search for instance names as well to tidy up DB
- row = instances.deobscure("?", blocked, blocked_hash)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning(f"Cannot deobsfucate blocked='{blocked}',blocked_hash='{blocked_hash}' - SKIPPED!")
- continue
-
- logger.debug("Updating domain: ", row[0])
- blocked = row[0]
- origin = row[1]
- nodeinfo_url = row[2]
-
- logger.debug("Looking up instance by domain:", blocked)
- if not validators.domain(blocked):
- logger.warning(f"blocked='{blocked}' is not a valid domain name - SKIPPED!")
- continue
- elif blocked.endswith(".arpa"):
- logger.warning(f"blocked='{blocked}' is a reversed .arpa domain and should not be used generally.")
- continue
- elif blocked.endswith(".tld"):
- logger.warning(f"blocked='{blocked}' is a fake domain, please don't crawl them!")
- continue
- elif blacklist.is_blacklisted(blocked):
- logger.debug("blocked='%s' is blacklisted - SKIPPED!", blocked)
- continue
- elif not instances.is_registered(blocked):
- logger.debug(f"Domain blocked='{blocked}' wasn't found, adding ..., domain='{domain}',origin='{origin}',nodeinfo_url='{nodeinfo_url}'")
- instances.add(blocked, domain, inspect.currentframe().f_code.co_name, nodeinfo_url)
-
- logger.debug("Looking up instance by domain:", blocked)
- if not validators.domain(blocked):
- logger.warning(f"blocked='{blocked}' is not a valid domain name - SKIPPED!")
- continue
- elif blocked.endswith(".arpa"):
- logger.warning(f"blocked='{blocked}' is a reversed .arpa domain and should not be used generally.")
- continue
- elif blocked.endswith(".tld"):
- logger.warning(f"blocked='{blocked}' is a fake domain, please don't crawl them!")
- continue
- elif blacklist.is_blacklisted(blocked):
- logger.debug("blocked='%s' is blacklisted - SKIPPED!", blocked)
- continue
- elif not instances.is_registered(blocked):
- logger.debug("Hash wasn't found, adding:", blocked, domain)
- instances.add(blocked, domain, inspect.currentframe().f_code.co_name, nodeinfo_url)
-
- if not blocks.is_instance_blocked(domain, blocked, block_level):
- logger.debug("Blocking:", domain, blocked, block_level)
- blocks.add_instance(domain, blocked, reason, block_level)
-
- if block_level == "reject":
- found_blocks.append({
- "blocked": blocked,
- "reason" : reason
- })
- else:
- logger.debug(f"Updating block last seen and reason for domain='{domain}',blocked='{blocked}' ...")
- blocks.update_last_seen(domain, blocked, block_level)
- blocks.update_reason(reason, domain, blocked, block_level)
-
- logger.debug("Committing changes ...")
- fba.connection.commit()
- except network.exceptions as exception:
- logger.warning(f"domain='{domain}',exception[{type(exception)}]:'{str(exception)}'")
- instances.set_last_error(domain, exception)
-
- logger.debug("EXIT!")
+ 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"],
+ "hash" : 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)
+
+ logger.debug("blocklist()=%d - EXIT!", len(blocklist))
+ return blocklist
projects / fba.git / blobdiff