# 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
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.http import nodeinfo
from fba.models import blocks
from fba.models import instances
# Language mapping X -> English
language_mapping = {
# English -> English
- "Reject": "Suspended servers",
+ "filtered media" : "filtered_media",
+ "limited servers" : "followers_only",
+ "followers-only" : "followers_only",
+ "media removal" : "media_removal",
+ "media_removal" : "media_removal",
+ "media force-set as sensitive": "media_nsfw",
+ "nsfw" : "media_nsfw",
+ "reject" : "reject",
+ "suspended servers": "reject",
+ "silenced servers" : "silenced",
+ "removal from \"the whole known network\" timeline": "federated_timeline_removal",
}
-def fetch_blocks(domain: str, nodeinfo_url: str) -> list:
- logger.debug("domain='%s',nodeinfo_url='%s' - CALLED!", domain, nodeinfo_url)
+def fetch_blocks(domain: str) -> list:
+ logger.debug("domain='%s' - CALLED!", domain)
domain_helper.raise_on(domain)
- if 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")
+ if blacklist.is_blacklisted(domain):
+ raise Exception(f"domain='{domain}' is blacklisted but function is invoked.")
+ elif not instances.is_registered(domain):
+ raise Exception(f"domain='{domain}' is not registered but function is invoked.")
blockdict = list()
rows = None
+
try:
- logger.debug(f"Fetching nodeinfo: domain='{domain}',nodeinfo_url='{nodeinfo_url}'")
- rows = federation.fetch_nodeinfo(domain, nodeinfo_url)
+ logger.debug("Fetching nodeinfo: domain='%s'", domain)
+ rows = nodeinfo.fetch(domain, update_mode=False)
+
+ if "error_message" in rows:
+ logger.warning("Error message '%s' during fetching nodeinfo for domain='%s'", rows["error_message"], domain)
+ instances.set_last_error(domain, rows)
+ instances.update(domain)
+
+ logger.debug("Returning empty list ... - EXIT!")
+ return list()
+ elif "exception" in rows:
+ logger.warning("Exception '%s' during fetching nodeinfo for domain='%s' - EXIT!", type(rows["exception"]), domain)
+ return list()
+ elif "json" in rows:
+ logger.debug("rows[json] found for domain='%s'", domain)
+ rows = rows["json"]
+
except network.exceptions as exception:
logger.warning("Exception '%s' during fetching nodeinfo from domain='%s'", type(exception), domain)
instances.set_last_error(domain, exception)
if rows is None:
- logger.warning("Could not fetch nodeinfo from domain='%s'", domain)
+ logger.warning("Could not fetch nodeinfo from domain='%s' - EXIT!", domain)
return list()
elif "metadata" not in rows:
- logger.warning("rows()=%d does not have key 'metadata', domain='%s'", len(rows), domain)
+ logger.warning("rows()=%d does not have key 'metadata', domain='%s' - EXIT!", len(rows), domain)
return list()
elif "federation" not in rows["metadata"]:
- logger.warning("rows()=%d does not have key 'federation', domain='%s'", len(rows['metadata']), domain)
+ logger.warning("rows()=%d does not have key 'federation', domain='%s' - EXIT!", len(rows["metadata"]), domain)
return list()
data = rows["metadata"]["federation"]
}
).items():
logger.debug("block_level='%s', blocklist()=%d", block_level, len(blocklist))
- block_level = tidyup.domain(block_level)
+ block_level = tidyup.domain(block_level) if block_level != "" else None
logger.debug("block_level='%s' - AFTER!", block_level)
if block_level == "":
logger.debug("domain='%s' skipping block_level='accept'", domain)
continue
+ block_level = blocks.alias_block_level(block_level)
+
logger.debug("Checking %d entries from domain='%s',block_level='%s' ...", len(blocklist), domain, block_level)
if len(blocklist) > 0:
for blocked in blocklist:
logger.debug("blocked='%s' - BEFORE!", blocked)
- blocked = tidyup.domain(blocked)
+ blocked = tidyup.domain(blocked) if blocked != "" else None
logger.debug("blocked='%s' - AFTER!", blocked)
- if blocked == "":
- logger.warning("blocked is empty after tidyup.domain(): domain='%s',block_level='%s'", domain, block_level)
- continue
- elif blocked.endswith(".arpa"):
- logger.debug("blocked='%s' is a reverse IP address - SKIPPED!", blocked)
- continue
- elif blocked.endswith(".tld"):
- logger.debug("blocked='%s' is a fake domain - SKIPPED!", blocked)
+ if blocked is None or blocked == "":
+ logger.warning("blocked='%s' is empty after tidyup.domain(): domain='%s',block_level='%s' - SKIPPED!", blocked, domain, block_level)
continue
- elif blocked.count("*") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("*", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
- elif blocked.count("?") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("?", blocked)
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
+ logger.debug("Invoking utils.deobfuscate(%s, %s) ...", blocked, domain)
+ blocked = utils.deobfuscate(blocked, domain)
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
-
- logger.debug("blocked='%s'", blocked)
- if not utils.is_domain_wanted(blocked):
+ logger.debug("blocked='%s' - DEOBFUSCATED!", blocked)
+ if blocked is None or blocked == "":
+ logger.warning("instance[host]='%s' is None or empty after tidyup.domain() - SKIPPED!", instance["host"])
+ continue
+ elif not domain_helper.is_wanted(blocked):
logger.debug("blocked='%s' is not wanted - SKIPPED!", blocked)
continue
for blocked in data["quarantined_instances"]:
logger.debug("blocked='%s' - BEFORE!", blocked)
- blocked = tidyup.domain(blocked)
- logger.debug("blocked='%s' - AFTER!", blocked)
+ blocked = tidyup.domain(blocked) if blocked != "" else None
- if blocked == "":
+ logger.debug("blocked='%s' - AFTER!", blocked)
+ if blocked is None or blocked == "":
logger.warning("blocked is empty after tidyup.domain(): domain='%s',block_level='%s'", domain, block_level)
continue
- elif blocked.endswith(".arpa"):
- logger.debug("blocked='%s' is a reverse IP address - SKIPPED!", blocked)
- continue
- elif blocked.endswith(".tld"):
- logger.debug("blocked='%s' is a fake domain - SKIPPED!", blocked)
- continue
- elif blocked.count("*") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("*", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
- elif blocked.count("?") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("?", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
-
- logger.debug("blocked='%s' - DEobfuscatED!", blocked)
- if not utils.is_domain_wanted(blocked):
+ elif not domain_helper.is_wanted(blocked):
logger.debug("blocked='%s' is not wanted - SKIPPED!", blocked)
continue
}
).items():
logger.debug("block_level='%s', info.items()=%d", block_level, len(info.items()))
- block_level = tidyup.domain(block_level)
+ block_level = tidyup.domain(block_level) if block_level != "" else None
logger.debug("block_level='%s' - AFTER!", block_level)
- if block_level == "":
- logger.warning("block_level is now empty!")
+ if block_level is None or block_level == "":
+ logger.warning("block_level='%s' is now empty!", block_level)
continue
elif block_level == "accept":
- logger.debug("domain='%s' skipping block_level='accept'", domain)
+ logger.debug("domain='%s': Skipping block_level='%s' ...", domain, block_level)
continue
+ block_level = blocks.alias_block_level(block_level)
+
logger.debug("Checking %d entries from domain='%s',block_level='%s' ...", len(info.items()), domain, block_level)
for blocked, reason in info.items():
logger.debug("blocked='%s',reason[%s]='%s' - BEFORE!", blocked, type(reason), reason)
- blocked = tidyup.domain(blocked)
+ blocked = tidyup.domain(blocked) if blocked != "" else None
logger.debug("blocked='%s' - AFTER!", blocked)
if isinstance(reason, str):
reason = tidyup.reason(reason)
elif isinstance(reason, dict) and "reason" in reason:
logger.debug("reason[] is a dict")
- reason = tidyup.reason(reason["reason"])
+ reason = tidyup.reason(reason["reason"]) if isinstance(reason["reason"], str) else None
elif reason is not None:
raise ValueError(f"Cannot handle reason[]='{type(reason)}'")
if blocked == "":
logger.warning("blocked is empty after tidyup.domain(): domain='%s',block_level='%s'", domain, block_level)
continue
- elif blocked.count("*") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("*", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
- elif blocked.count("?") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("?", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
-
- logger.debug("blocked='%s' - DEobfuscatED!", blocked)
- if not utils.is_domain_wanted(blocked):
- logger.debug("blocked='%s' is not wanted - SKIPPED!", blocked)
- continue
logger.debug("Checking %d blockdict records ...", len(blockdict))
for block in blockdict:
- logger.debug("block[blocked]='%s',blocked='%s'", block['blocked'], blocked)
- if block['blocked'] == blocked:
- logger.debug("Updating reason='%s' for blocker='%s'", reason, block['blocked'])
- block['reason'] = reason
+ logger.debug("block[blocked]='%s',blocked='%s'", block["blocked"], blocked)
+ if block["blocked"] == blocked:
+ logger.debug("Updating reason='%s' for blocker='%s'", reason, block["blocked"])
+ block["reason"] = reason
elif "quarantined_instances_info" in data and "quarantined_instances" in data["quarantined_instances_info"]:
logger.debug("Found 'quarantined_instances_info' in JSON response: domain='%s'", domain)
rows = data["quarantined_instances_info"]["quarantined_instances"]
for blocked in rows:
logger.debug("blocked='%s' - BEFORE!", blocked)
- blocked = tidyup.domain(blocked)
- logger.debug("blocked='%s' - AFTER!", blocked)
+ reason = tidyup.reason(rows[blocked]["reason"])
+ blocked = tidyup.domain(blocked) if blocked != "" else None
+ logger.debug("blocked='%s',reason='%s' - AFTER!", blocked, reason)
if blocked not in rows or "reason" not in rows[blocked]:
logger.warning("Cannot find blocked='%s' in rows()=%d,domain='%s' - BREAK!", blocked, len(rows), domain)
break
-
- reason = rows[blocked]["reason"]
- logger.debug("reason='%s'", reason)
-
- if blocked == "":
+ elif blocked == "":
logger.warning("blocked is empty after tidyup.domain(): domain='%s',block_level='%s'", domain, block_level)
continue
- elif blocked.count("*") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("*", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
- elif blocked.count("?") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("?", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
-
- logger.debug("blocked='%s'", blocked)
- if not utils.is_domain_wanted(blocked):
- logger.debug("blocked='%s' is not wanted - SKIPPED!", blocked)
- continue
logger.debug("Checking %d blockdict records ...", len(blockdict))
for block in blockdict:
- logger.debug("block[blocked]='%s',blocked='%s'", block['blocked'], blocked)
- if block['blocked'] == blocked:
- logger.debug("Updating reason='%s' for blocker='%s'", reason, block['blocked'])
- block['reason'] = reason
+ logger.debug("block[blocked]='%s',blocked='%s'", block["blocked"], blocked)
+ if block["blocked"] == blocked:
+ logger.debug("Updating reason='%s' for blocker='%s'", reason, block["blocked"])
+ block["reason"] = reason
else:
logger.warning("Cannot find 'mrf_simple_info' or 'quarantined_instances_info' in JSON reply: domain='%s'", domain)
logger.debug("blocklist()=%d", len(blocklist))
if len(blocklist) > 0:
- logger.info("Checking %d record(s) ...", len(blocklist))
+ logger.info("Checking %d different blocklists ...", len(blocklist))
for block_level in blocklist:
logger.debug("block_level='%s'", block_level)
-
rows = blocklist[block_level]
- logger.debug("rows[%s]()=%d'", type(rows), len(rows))
- for record in rows:
- logger.debug("record[]='%s'", type(record))
- blocked = tidyup.domain(record["blocked"])
- reason = tidyup.reason(record["reason"])
- logger.debug("blocked='%s',reason='%s' - AFTER!", blocked, reason)
-
- if blocked == "":
- logger.warning("blocked is empty after tidyup.domain(): domain='%s',block_level='%s'", domain, block_level)
- continue
- elif blocked.count("*") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("*", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
- elif blocked.count("?") > 0:
- logger.debug("domain='%s' uses obfuscated domains, marking ...", domain)
- instances.set_has_obfuscation(domain, True)
-
- # Obscured domain name with no hash
- row = instances.deobfuscate("?", blocked)
-
- logger.debug("row[]='%s'", type(row))
- if row is None:
- logger.warning("Cannot deobfuscate blocked='%s',domain='%s' - SKIPPED!", blocked, domain)
- continue
-
- logger.debug("blocked='%s' de-obscured to '%s'", blocked, row[0])
- blocked = row[0]
-
- logger.debug("blocked='%s' - DEobfuscatED!", blocked)
- if not utils.is_domain_wanted(blocked):
- logger.warning("blocked='%s' is not wanted - SKIPPED!", blocked)
- continue
- logger.debug("Appending blocker='%s',blocked='%s',reason='%s',block_level='%s' ...",domain, blocked, reason, block_level)
+ logger.debug("rows[%s]()=%d'", type(rows), len(rows))
+ for block in rows:
+ logger.debug("Appending blocker='%s',block[blocked]='%s',block[reason]='%s',block_level='%s' ...",domain, block["blocked"], block["reason"], block_level)
blockdict.append({
"blocker" : domain,
- "blocked" : blocked,
- "reason" : reason,
+ "blocked" : block["blocked"],
+ "reason" : block["reason"],
"block_level": block_level,
})
return blockdict
def fetch_blocks_from_about(domain: str) -> dict:
- logger.debug("domain(%d)='%s' - CALLED!", len(domain), domain)
+ 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 is invoked.")
+ elif 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 ["/instance/about/index.html"]:
(config.get("connection_timeout"), config.get("read_timeout"))
)
- logger.debug("response.ok='%s',response.status_code='%d',response.text()=%d", response.ok, response.status_code, len(response.text))
+ logger.debug("response.ok='%s',response.status_code=%d,response.text()=%d", response.ok, response.status_code, len(response.text))
if not response.ok or response.text.strip() == "":
logger.warning("path='%s' does not exist on domain='%s' - SKIPPED!", path, domain)
continue
break
blocklist = {
- "Suspended servers": [],
- "Filtered media" : [],
- "Limited servers" : [],
- "Silenced servers" : [],
+ "reject" : [],
+ "filtered_media": [],
+ "followers_only": [],
+ "silenced" : [],
+ "media_nsfw" : [],
+ "media_removal" : [],
+ "federated_timeline_removal": [],
}
logger.debug("doc[]='%s'", type(doc))
logger.warning("Cannot fetch any /about pages for domain='%s' - EXIT!", domain)
return list()
- for header in doc.find_all("h2"):
- header_text = tidyup.reason(header.text)
+ headers = doc.find_all("h2")
+
+ logger.debug("headers[]='%s'", type(headers))
+ if headers is None:
+ logger.warning("Cannot fetch any /about pages for domain='%s' - EXIT!", domain)
+ return list()
+
+ logger.info("Checking %d headers ...", len(headers))
+ for header in headers:
+ logger.debug("header[%s]='%s'", type(header), header)
+ block_level = tidyup.reason(header.text).lower()
- logger.debug("header_text='%s' - BEFORE!", header_text)
- if header_text in language_mapping:
- logger.debug("header_text='%s' - FOUND!", header_text)
- header_text = language_mapping[header_text]
+ logger.debug("block_level='%s' - BEFORE!", block_level)
+ if block_level in language_mapping:
+ logger.debug("block_level='%s' - FOUND!", block_level)
+ block_level = language_mapping[block_level].lower()
else:
- logger.warning("header_text='%s' not found in language mapping table", header_text)
+ logger.warning("block_level='%s' not found in language mapping table", block_level)
- logger.debug("header_text='%s - AFTER!'", header_text)
- if header_text in blocklist or header_text.lower() in blocklist:
+ logger.debug("block_level='%s - AFTER!'", block_level)
+ if block_level in blocklist:
# replaced find_next_siblings with find_all_next to account for instances that e.g. hide lists in dropdown menu
- logger.debug("Found header_text='%s', importing domain blocks ...", header_text)
+ logger.debug("Found block_level='%s', importing domain blocks ...", block_level)
for line in header.find_next("table").find_all("tr")[1:]:
logger.debug("line[]='%s'", type(line))
- blocklist[header_text].append({
- "blocked": tidyup.domain(line.find_all("td")[0].text),
- "reason" : tidyup.reason(line.find_all("td")[1].text),
+ blocked = line.find_all("td")[0].text
+ logger.debug("blocked='%s'", blocked)
+
+ blocked = tidyup.domain(blocked) if blocked != "" else None
+ reason = tidyup.reason(line.find_all("td")[1].text)
+ logger.debug("blocked='%s',reason='%s' - AFTER!", blocked, reason)
+
+ if blocked is None or blocked == "":
+ logger.debug("domain='%s',block_level='%s': blocked is empty - SKIPPED!", domain, block_level)
+ continue
+ elif not domain_helper.is_wanted(blocked):
+ logger.debug("blocked='%s' is not wanted - SKIPPED!", blocked)
+ continue
+
+ logger.debug("Appending block_level='%s',blocked='%s',reason='%s' ...", block_level, blocked, reason)
+ blocklist[block_level].append({
+ "blocked": blocked,
+ "reason" : reason,
})
else:
- logger.warning("header_text='%s' not found in blocklist()=%d", header_text, len(blocklist))
+ logger.warning("block_level='%s' not found in blocklist()=%d", block_level, len(blocklist))
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"],
- }
+ return blocklist
projects / fba.git / blobdiff