X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=fba%2Ffba.py;h=49b1044a091e2a76c78e57fec3b1f22c94f42bd8;hb=eb07351489e809640a7116f9303e8497b312654a;hp=a94558e0dde8d96b7a5f165f7452dbdaf596d5bd;hpb=c6aebe91d210763f8206d20a73d6b1f12c8b43c9;p=fba.git diff --git a/fba/fba.py b/fba/fba.py index a94558e..49b1044 100644 --- a/fba/fba.py +++ b/fba/fba.py @@ -1,4 +1,3 @@ -# Fedi API Block - An aggregator for fetching blocking data from fediverse nodes # Copyright (C) 2023 Free Software Foundation # # This program is free software: you can redistribute it and/or modify @@ -17,7 +16,6 @@ import bs4 import hashlib import re -import reqto import requests import json import sqlite3 @@ -25,27 +23,17 @@ import sys import time import validators +from urllib.parse import urlparse + +from fba import blacklist from fba import cache from fba import config from fba import instances +from fba import network -# Don't check these, known trolls/flooders/testing/developing -blacklist = [ - # Floods network with fake nodes as "research" project - "activitypub-troll.cf", - # Similar troll - "gab.best", - # Similar troll - "4chan.icu", - # Flooder (?) - "social.shrimpcam.pw", - # Flooder (?) - "mastotroll.netz.org", - # Testing/developing installations - "ngrok.io", - "ngrok-free.app", - "misskeytest.chn.moe", -] +from fba.federation import lemmy +from fba.federation import misskey +from fba.federation import peertube # Array with pending errors needed to be written to database pending_errors = { @@ -74,34 +62,6 @@ api_headers = { "Content-Type": "application/json", } -language_mapping = { - # English -> English - "Silenced instances" : "Silenced servers", - "Suspended instances" : "Suspended servers", - "Limited instances" : "Limited servers", - # Mappuing German -> English - "Gesperrte Server" : "Suspended servers", - "Gefilterte Medien" : "Filtered media", - "Stummgeschaltete Server" : "Silenced servers", - # Japanese -> English - "停止済みのサーバー" : "Suspended servers", - "制限中のサーバー" : "Limited servers", - "メディアを拒否しているサーバー": "Filtered media", - "サイレンス済みのサーバー" : "Silenced servers", - # ??? -> English - "שרתים מושעים" : "Suspended servers", - "מדיה מסוננת" : "Filtered media", - "שרתים מוגבלים" : "Silenced servers", - # French -> English - "Serveurs suspendus" : "Suspended servers", - "Médias filtrés" : "Filtered media", - "Serveurs limités" : "Limited servers", - "Serveurs modérés" : "Limited servers", -} - -# URL for fetching peers -get_peers_url = "/api/v1/instance/peers" - # Connect to database connection = sqlite3.connect("blocks.db") cursor = connection.cursor() @@ -121,45 +81,51 @@ patterns = [ ##### Other functions ##### def is_primitive(var: any) -> bool: - # NOISY-DEBUG: print(f"DEBUG: var[]='{type(var)}' - CALLED!") - return type(var) in {int, str, float, bool} or var == None + # DEBUG: print(f"DEBUG: var[]='{type(var)}' - CALLED!") + return type(var) in {int, str, float, bool} or var is None def fetch_instances(domain: str, origin: str, software: str, script: str, path: str = None): - if type(domain) != str: + # DEBUG: print(f"DEBUG: domain='{domain}',origin='{origin}',software='{software}',path='{path}' - CALLED!") + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - elif type(origin) != str and origin != None: + raise ValueError("Parameter 'domain' is empty") + elif not isinstance(origin, str) and origin is not None: raise ValueError(f"Parameter origin[]={type(origin)} is not 'str'") - elif type(script) != str: + elif software is None: + print(f"DEBUG: software for domain='{domain}' is not set, determining ...") + software = determine_software(domain, path) + print(f"DEBUG: Determined software='{software}' for domain='{domain}'") + elif not isinstance(software, str): + raise ValueError(f"Parameter software[]={type(software)} is not 'str'") + elif not isinstance(script, str): raise ValueError(f"Parameter script[]={type(script)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError("Parameter 'domain' is empty") - # DEBUG: print("DEBUG: domain,origin,software,path:", domain, origin, software, path) - if not is_instance_registered(domain): + if not instances.is_registered(domain): # DEBUG: print("DEBUG: Adding new domain:", domain, origin) - add_instance(domain, origin, script, path) + instances.add(domain, origin, script, path) # DEBUG: print("DEBUG: Fetching instances for domain:", domain, software) - peerlist = get_peers(domain, software) + peerlist = fetch_peers(domain, software) if (peerlist is None): print("ERROR: Cannot fetch peers:", domain) return elif instances.has_pending_instance_data(domain): # DEBUG: print(f"DEBUG: domain='{domain}' has pending nodeinfo data, flushing ...") - instances.update_instance_data(domain) + instances.update_data(domain) print(f"INFO: Checking {len(peerlist)} instances from {domain} ...") for instance in peerlist: - if instance == None: + if instance is None: # Skip "None" types as tidup() cannot parse them continue - # DEBUG: print(f"DEBUG: instance[{type(instance}]={instance} - BEFORE") + # DEBUG: print(f"DEBUG: instance='{instance}' - BEFORE") instance = tidyup_domain(instance) - # DEBUG: print(f"DEBUG: instance[{type(instance}]={instance} - AFTER") + # DEBUG: print(f"DEBUG: instance='{instance}' - AFTER") if instance == "": print("WARNING: Empty instance after tidyup_domain(), domain:", domain) @@ -167,17 +133,17 @@ def fetch_instances(domain: str, origin: str, software: str, script: str, path: elif not validators.domain(instance.split("/")[0]): print(f"WARNING: Bad instance='{instance}' from domain='{domain}',origin='{origin}',software='{software}'") continue - elif is_blacklisted(instance): + elif blacklist.is_blacklisted(instance): # DEBUG: print("DEBUG: instance is blacklisted:", instance) continue # DEBUG: print("DEBUG: Handling instance:", instance) try: - if not is_instance_registered(instance): + if not instances.is_registered(instance): # DEBUG: print("DEBUG: Adding new instance:", instance, domain) - add_instance(instance, domain, sys.argv[0]) - except BaseException as e: - print(f"ERROR: instance='{instance}',exception[{type(e)}]:'{str(e)}'") + instances.add(instance, domain, script) + except BaseException as exception: + print(f"ERROR: instance='{instance}',exception[{type(exception)}]:'{str(exception)}'") continue # DEBUG: print("DEBUG: EXIT!") @@ -185,16 +151,16 @@ def fetch_instances(domain: str, origin: str, software: str, script: str, path: def add_peers(rows: dict) -> list: # DEBUG: print(f"DEBUG: rows()={len(rows)} - CALLED!") peers = list() - for element in ["linked", "allowed", "blocked"]: - # DEBUG: print(f"DEBUG: Checking element='{element}'") - if element in rows and rows[element] != None: - # DEBUG: print(f"DEBUG: Adding {len(rows[element])} peer(s) to peers list ...") - for peer in rows[element]: + for key in ["linked", "allowed", "blocked"]: + # DEBUG: print(f"DEBUG: Checking key='{key}'") + if key in rows and rows[key] is not None: + # DEBUG: print(f"DEBUG: Adding {len(rows[key])} peer(s) to peers list ...") + for peer in rows[key]: # DEBUG: print(f"DEBUG: peer='{peer}' - BEFORE!") peer = tidyup_domain(peer) # DEBUG: print(f"DEBUG: peer='{peer}' - AFTER!") - if is_blacklisted(peer): + if blacklist.is_blacklisted(peer): # DEBUG: print(f"DEBUG: peer='{peer}' is blacklisted, skipped!") continue @@ -230,7 +196,6 @@ def remove_version(software: str) -> str: # DEBUG: print(f"DEBUG: Was not able to find common seperator, returning untouched software='{software}'") return software - matches = None match = None # DEBUG: print(f"DEBUG: Checking {len(patterns)} patterns ...") for pattern in patterns: @@ -238,11 +203,12 @@ def remove_version(software: str) -> str: match = pattern.match(version) # DEBUG: print(f"DEBUG: match[]={type(match)}") - if type(match) is re.Match: + if isinstance(match, re.Match): + # DEBUG: print(f"DEBUG: version='{version}' is matching pattern='{pattern}'") break # DEBUG: print(f"DEBUG: version[{type(version)}]='{version}',match='{match}'") - if type(match) is not re.Match: + if not isinstance(match, re.Match): print(f"WARNING: version='{version}' does not match regex, leaving software='{software}' untouched.") return software @@ -260,11 +226,12 @@ def remove_version(software: str) -> str: def strip_powered_by(software: str) -> str: # DEBUG: print(f"DEBUG: software='{software}' - CALLED!") - if software == "": - print(f"ERROR: Bad method call, 'software' is empty") - raise Exception("Parameter 'software' is empty") + if not isinstance(software, str): + raise ValueError(f"Parameter software[]='{type(software)}' is not 'str'") + elif software == "": + raise ValueError("Parameter 'software' is empty") elif not "powered by" in software: - print(f"WARNING: Cannot find 'powered by' in '{software}'!") + print(f"WARNING: Cannot find 'powered by' in software='{software}'!") return software start = software.find("powered by ") @@ -280,9 +247,10 @@ def strip_powered_by(software: str) -> str: def strip_hosted_on(software: str) -> str: # DEBUG: print(f"DEBUG: software='{software}' - CALLED!") - if software == "": - print(f"ERROR: Bad method call, 'software' is empty") - raise Exception("Parameter 'software' is empty") + if not isinstance(software, str): + raise ValueError(f"Parameter software[]='{type(software)}' is not 'str'") + elif software == "": + raise ValueError("Parameter 'software' is empty") elif not "hosted on" in software: print(f"WARNING: Cannot find 'hosted on' in '{software}'!") return software @@ -290,7 +258,7 @@ def strip_hosted_on(software: str) -> str: end = software.find("hosted on ") # DEBUG: print(f"DEBUG: end[{type(end)}]='{end}'") - software = software[0, start].strip() + software = software[0, end].strip() # DEBUG: print(f"DEBUG: software='{software}'") software = strip_until(software, " - ") @@ -300,12 +268,14 @@ def strip_hosted_on(software: str) -> str: def strip_until(software: str, until: str) -> str: # DEBUG: print(f"DEBUG: software='{software}',until='{until}' - CALLED!") - if software == "": - print(f"ERROR: Bad method call, 'software' is empty") - raise Exception("Parameter 'software' is empty") + if not isinstance(software, str): + raise ValueError(f"Parameter software[]='{type(software)}' is not 'str'") + elif software == "": + raise ValueError("Parameter 'software' is empty") + elif not isinstance(until, str): + raise ValueError(f"Parameter until[]='{type(until)}' is not 'str'") elif until == "": - print(f"ERROR: Bad method call, 'until' is empty") - raise Exception("Parameter 'until' is empty") + raise ValueError("Parameter 'until' is empty") elif not until in software: print(f"WARNING: Cannot find '{until}' in '{software}'!") return software @@ -320,24 +290,11 @@ def strip_until(software: str, until: str) -> str: # DEBUG: print(f"DEBUG: software='{software}' - EXIT!") return software -def is_blacklisted(domain: str) -> bool: - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - blacklisted = False - for peer in blacklist: - if peer in domain: - blacklisted = True - - return blacklisted - def remove_pending_error(domain: str): - if type(domain) != str: + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError("Parameter 'domain' is empty") try: # Prevent updating any pending errors, nodeinfo was found @@ -349,308 +306,86 @@ def remove_pending_error(domain: str): # DEBUG: print("DEBUG: EXIT!") def get_hash(domain: str) -> str: - if type(domain) != str: + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError("Parameter 'domain' is empty") return hashlib.sha256(domain.encode("utf-8")).hexdigest() -def update_last_blocked(domain: str): - if type(domain) != str: +def log_error(domain: str, response: requests.models.Response): + # DEBUG: print("DEBUG: domain,response[]:", domain, type(response)) + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - # DEBUG: print("DEBUG: Updating last_blocked for domain", domain) - instances.set("last_blocked", domain, time.time()) - - # Running pending updated - # DEBUG: print(f"DEBUG: Invoking instances.update_instance_data({domain}) ...") - instances.update_instance_data(domain) - - # DEBUG: print("DEBUG: EXIT!") - -def log_error(domain: str, res: any): - # DEBUG: print("DEBUG: domain,res[]:", domain, type(res)) - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError("Parameter 'domain' is empty") try: - # DEBUG: print("DEBUG: BEFORE res[]:", type(res)) - if isinstance(res, BaseException) or isinstance(res, json.decoder.JSONDecodeError): - res = str(res) + # DEBUG: print("DEBUG: BEFORE response[]:", type(response)) + if isinstance(response, BaseException) or isinstance(response, json.decoder.JSONDecodeError): + response = str(response) - # DEBUG: print("DEBUG: AFTER res[]:", type(res)) - if type(res) is str: + # DEBUG: print("DEBUG: AFTER response[]:", type(response)) + if isinstance(response, str): cursor.execute("INSERT INTO error_log (domain, error_code, error_message, created) VALUES (?, 999, ?, ?)",[ domain, - res, + response, time.time() ]) else: cursor.execute("INSERT INTO error_log (domain, error_code, error_message, created) VALUES (?, ?, ?, ?)",[ domain, - res.status_code, - res.reason, + response.status_code, + response.reason, time.time() ]) # Cleanup old entries # DEBUG: print(f"DEBUG: Purging old records (distance: {config.get('error_log_cleanup')})") cursor.execute("DELETE FROM error_log WHERE created < ?", [time.time() - config.get("error_log_cleanup")]) - except BaseException as e: - print(f"ERROR: failed SQL query: domain='{domain}',exception[{type(e)}]:'{str(e)}'") + except BaseException as exception: + print(f"ERROR: failed SQL query: domain='{domain}',exception[{type(exception)}]:'{str(exception)}'") sys.exit(255) # DEBUG: print("DEBUG: EXIT!") -def update_last_error(domain: str, res: any): - # DEBUG: print("DEBUG: domain,res[]:", domain, type(res)) - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - # DEBUG: print("DEBUG: BEFORE res[]:", type(res)) - if isinstance(res, BaseException) or isinstance(res, json.decoder.JSONDecodeError): - res = f"{type}:str(res)" - - # DEBUG: print("DEBUG: AFTER res[]:", type(res)) - if type(res) is str: - # DEBUG: print(f"DEBUG: Setting last_error_details='{res}'"); - instances.set("last_status_code" , domain, 999) - instances.set("last_error_details", domain, res) - else: - # DEBUG: print(f"DEBUG: Setting last_error_details='{res.reason}'"); - instances.set("last_status_code" , domain, res.status_code) - instances.set("last_error_details", domain, res.reason) - - # Running pending updated - # DEBUG: print(f"DEBUG: Invoking instances.update_instance_data({domain}) ...") - instances.update_instance_data(domain) - - log_error(domain, res) - - # DEBUG: print("DEBUG: EXIT!") - -def update_last_instance_fetch(domain: str): - # DEBUG: print(f"DEBUG: domain={domain} - CALLED!") - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - # DEBUG: print("DEBUG: Updating last_instance_fetch for domain:", domain) - instances.set("last_instance_fetch", domain, time.time()) - - # Running pending updated - # DEBUG: print(f"DEBUG: Invoking instances.update_instance_data({domain}) ...") - instances.update_instance_data(domain) - - # DEBUG: print("DEBUG: EXIT!") - -def update_last_nodeinfo(domain: str): - # DEBUG: print(f"DEBUG: domain={domain} - CALLED!") - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - # DEBUG: print("DEBUG: Updating last_nodeinfo for domain:", domain) - instances.set("last_nodeinfo", domain, time.time()) - instances.set("last_updated" , domain, time.time()) - - # Running pending updated - # DEBUG: print(f"DEBUG: Invoking instances.update_instance_data({domain}) ...") - instances.update_instance_data(domain) - - # DEBUG: print("DEBUG: EXIT!") - -def get_peers(domain: str, software: str) -> list: +def fetch_peers(domain: str, software: str) -> list: # DEBUG: print(f"DEBUG: domain({len(domain)})={domain},software={software} - CALLED!") - if type(domain) != str: + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - elif type(software) != str and software != None: + raise ValueError("Parameter 'domain' is empty") + elif not isinstance(software, str) and software is not None: raise ValueError(f"software[]={type(software)} is not 'str'") - peers = list() - if software == "misskey": - # DEBUG: print(f"DEBUG: domain='{domain}' is misskey, sending API POST request ...") - offset = 0 - step = config.get("misskey_limit") - - # iterating through all "suspended" (follow-only in its terminology) - # instances page-by-page, since that troonware doesn't support - # sending them all at once - while True: - # DEBUG: print(f"DEBUG: Fetching offset='{offset}' from '{domain}' ...") - if offset == 0: - fetched = post_json_api(domain, "/api/federation/instances", json.dumps({ - "sort" : "+pubAt", - "host" : None, - "limit": step - }), { - "Origin": domain - }) - else: - fetched = post_json_api(domain, "/api/federation/instances", json.dumps({ - "sort" : "+pubAt", - "host" : None, - "limit" : step, - "offset": offset - 1 - }), { - "Origin": domain - }) - - # DEBUG: print(f"DEBUG: fetched()={len(fetched))}") - if len(fetched) == 0: - # DEBUG: print("DEBUG: Returned zero bytes, exiting loop:", domain) - break - elif len(fetched) != config.get("misskey_limit"): - # DEBUG: print(f"DEBUG: Fetched '{len(fetched)}' row(s) but expected: '{config.get('misskey_limit')}'") - offset = offset + (config.get("misskey_limit") - len(fetched)) - else: - # DEBUG: print("DEBUG: Raising offset by step:", step) - offset = offset + step - - # Check records - # DEBUG: print(f"DEBUG: fetched({len(fetched)})[]={type(fetched)}") - if isinstance(fetched, dict) and "error" in fetched and "message" in fetched["error"]: - print(f"WARNING: post_json_api() returned error: {fetched['error']['message']}") - update_last_error(domain, fetched["error"]["message"]) - break - - already = 0 - for row in fetched: - # DEBUG: print(f"DEBUG: row():{len(row)}") - if not "host" in row: - print(f"WARNING: row()={len(row)} does not contain element 'host': {row},domain='{domain}'") - continue - elif type(row["host"]) != str: - print(f"WARNING: row[host][]={type(row['host'])} is not 'str'") - continue - elif is_blacklisted(row["host"]): - # DEBUG: print(f"DEBUG: row[host]='{row['host']}' is blacklisted. domain='{domain}'") - continue - elif row["host"] in peers: - # DEBUG: print(f"DEBUG: Not adding row[host]='{row['host']}', already found.") - already = already + 1 - continue - - # DEBUG: print(f"DEBUG: Adding peer: '{row['host']}'") - peers.append(row["host"]) - - if already == len(fetched): - print(f"WARNING: Host returned same set of '{already}' instances, aborting loop!") - break - - # DEBUG: print(f"DEBUG: Adding '{len(peers)}' for domain='{domain}'") - instances.set("total_peers", domain, len(peers)) - - # DEBUG: print(f"DEBUG: Updating last_instance_fetch for domain='{domain}' ...") - update_last_instance_fetch(domain) - - # DEBUG: print("DEBUG: Returning peers[]:", type(peers)) - return peers + # DEBUG: print(f"DEBUG: Invoking misskey.fetch_peers({domain}) ...") + return misskey.fetch_peers(domain) elif software == "lemmy": - # DEBUG: print(f"DEBUG: domain='{domain}' is Lemmy, fetching JSON ...") - try: - res = reqto.get(f"https://{domain}/api/v3/site", headers=api_headers, timeout=(config.get("connection_timeout"), config.get("read_timeout"))) - - data = json_from_response(res) - - # DEBUG: print(f"DEBUG: res.ok={res.ok},res.status_code='{res.status_code}',data[]='{type(data)}'") - if not res.ok or res.status_code >= 400: - print("WARNING: Could not reach any JSON API:", domain) - update_last_error(domain, res) - elif res.ok and isinstance(data, list): - # DEBUG: print(f"DEBUG: domain='{domain}' returned a list: '{data}'") - sys.exit(255) - elif "federated_instances" in data: - # DEBUG: print(f"DEBUG: Found federated_instances for domain='{domain}'") - peers = peers + add_peers(data["federated_instances"]) - # DEBUG: print("DEBUG: Added instance(s) to peers") - else: - print("WARNING: JSON response does not contain 'federated_instances':", domain) - update_last_error(domain, res) - - except BaseException as e: - print(f"WARNING: Exception during fetching JSON: domain='{domain}',exception[{type(e)}]:'{str(e)}'") - - # DEBUG: print(f"DEBUG: Adding '{len(peers)}' for domain='{domain}'") - instances.set("total_peers", domain, len(peers)) - - # DEBUG: print(f"DEBUG: Updating last_instance_fetch for domain='{domain}' ...") - update_last_instance_fetch(domain) - - # DEBUG: print("DEBUG: Returning peers[]:", type(peers)) - return peers + # DEBUG: print(f"DEBUG: Invoking lemmy.fetch_peers({domain}) ...") + return lemmy.fetch_peers(domain) elif software == "peertube": - # DEBUG: print(f"DEBUG: domain='{domain}' is a PeerTube, fetching JSON ...") - - start = 0 - for mode in ["followers", "following"]: - # DEBUG: print(f"DEBUG: domain='{domain}',mode='{mode}'") - while True: - try: - res = reqto.get(f"https://{domain}/api/v1/server/{mode}?start={start}&count=100", headers=headers, timeout=(config.get("connection_timeout"), config.get("read_timeout"))) - - data = json_from_response(res) - # DEBUG: print(f"DEBUG: res.ok={res.ok},res.status_code='{res.status_code}',data[]='{type(data)}'") - if res.ok and isinstance(data, dict): - # DEBUG: print("DEBUG: Success, data:", len(data)) - if "data" in data: - # DEBUG: print(f"DEBUG: Found {len(data['data'])} record(s).") - for record in data["data"]: - # DEBUG: print(f"DEBUG: record()={len(record)}") - if mode in record and "host" in record[mode]: - # DEBUG: print(f"DEBUG: Found host={record[mode]['host']}, adding ...") - peers.append(record[mode]["host"]) - else: - print(f"WARNING: record from '{domain}' has no '{mode}' or 'host' record: {record}") - - if len(data["data"]) < 100: - # DEBUG: print("DEBUG: Reached end of JSON response:", domain) - break - - # Continue with next row - start = start + 100 - - except BaseException as e: - print(f"WARNING: Exception during fetching JSON: domain='{domain}',exception[{type(e)}]:'{str(e)}'") - - # DEBUG: print(f"DEBUG: Adding '{len(peers)}' for domain='{domain}'") - instances.set("total_peers", domain, len(peers)) - - # DEBUG: print(f"DEBUG: Updating last_instance_fetch for domain='{domain}' ...") - update_last_instance_fetch(domain) - - # DEBUG: print("DEBUG: Returning peers[]:", type(peers)) - return peers - - # DEBUG: print(f"DEBUG: Fetching get_peers_url='{get_peers_url}' from '{domain}' ...") + # DEBUG: print(f"DEBUG: Invoking peertube.fetch_peers({domain}) ...") + return peertube.fetch_peers(domain) + + # DEBUG: print(f"DEBUG: Fetching peers from '{domain}',software='{software}' ...") + peers = list() try: - res = reqto.get(f"https://{domain}{get_peers_url}", headers=api_headers, timeout=(config.get("connection_timeout"), config.get("read_timeout"))) + response = network.fetch_response(domain, "/api/v1/instance/peers", api_headers, (config.get("connection_timeout"), config.get("read_timeout"))) - data = json_from_response(res) + data = json_from_response(response) - # DEBUG: print(f"DEBUG: res.ok={res.ok},res.status_code={res.status_code},data[]='{type(data)}'") - if not res.ok or res.status_code >= 400: - # DEBUG: print(f"DEBUG: Was not able to fetch '{get_peers_url}', trying alternative ...") - res = reqto.get(f"https://{domain}/api/v3/site", headers=api_headers, timeout=(config.get("connection_timeout"), config.get("read_timeout"))) + # DEBUG: print(f"DEBUG: response.ok={response.ok},response.status_code={response.status_code},data[]='{type(data)}'") + if not response.ok or response.status_code >= 400: + # DEBUG: print(f"DEBUG: Was not able to fetch peers, trying alternative ...") + response = network.fetch_response(domain, "/api/v3/site", api_headers, (config.get("connection_timeout"), config.get("read_timeout"))) - data = json_from_response(res) - # DEBUG: print(f"DEBUG: res.ok={res.ok},res.status_code={res.status_code},data[]='{type(data)}'") - if not res.ok or res.status_code >= 400: + data = json_from_response(response) + # DEBUG: print(f"DEBUG: response.ok={response.ok},response.status_code={response.status_code},data[]='{type(data)}'") + if not response.ok or response.status_code >= 400: print("WARNING: Could not reach any JSON API:", domain) - update_last_error(domain, res) - elif res.ok and isinstance(data, list): + instances.update_last_error(domain, response) + elif response.ok and isinstance(data, list): # DEBUG: print(f"DEBUG: domain='{domain}' returned a list: '{data}'") sys.exit(255) elif "federated_instances" in data: @@ -659,63 +394,34 @@ def get_peers(domain: str, software: str) -> list: # DEBUG: print("DEBUG: Added instance(s) to peers") else: print("WARNING: JSON response does not contain 'federated_instances':", domain) - update_last_error(domain, res) + instances.update_last_error(domain, response) else: # DEBUG: print("DEBUG: Querying API was successful:", domain, len(data)) peers = data - except BaseException as e: - print("WARNING: Some error during get():", domain, e) - update_last_error(domain, e) + except BaseException as exception: + print("WARNING: Some error during get():", domain, exception) + instances.update_last_error(domain, exception) # DEBUG: print(f"DEBUG: Adding '{len(peers)}' for domain='{domain}'") instances.set("total_peers", domain, len(peers)) # DEBUG: print(f"DEBUG: Updating last_instance_fetch for domain='{domain}' ...") - update_last_instance_fetch(domain) + instances.update_last_instance_fetch(domain) # DEBUG: print("DEBUG: Returning peers[]:", type(peers)) return peers -def post_json_api(domain: str, path: str, parameter: str, extra_headers: dict = {}) -> dict: - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - elif type(path) != str: - raise ValueError(f"path[]={type(path)} is not 'str'") - elif path == "": - raise ValueError("Parameter 'path' cannot be empty") - elif type(parameter) != str: - raise ValueError(f"parameter[]={type(parameter)} is not 'str'") - - # DEBUG: print("DEBUG: Sending POST to domain,path,parameter:", domain, path, parameter, extra_headers) - data = {} - try: - res = reqto.post(f"https://{domain}{path}", data=parameter, headers={**api_headers, **extra_headers}, timeout=(config.get("connection_timeout"), config.get("read_timeout"))) - - data = json_from_response(res) - # DEBUG: print(f"DEBUG: res.ok={res.ok},res.status_code={res.status_code},data[]='{type(data)}'") - if not res.ok or res.status_code >= 400: - print(f"WARNING: Cannot query JSON API: domain='{domain}',path='{path}',parameter()={len(parameter)},res.status_code='{res.status_code}',data[]='{type(data)}'") - update_last_error(domain, res) - - except BaseException as e: - print(f"WARNING: Some error during post(): domain='{domain}',path='{path}',parameter()={len(parameter)},exception[{type(e)}]:'{str(e)}'") - - # DEBUG: print(f"DEBUG: Returning data({len(data)})=[]:{type(data)}") - return data - def fetch_nodeinfo(domain: str, path: str = None) -> list: # DEBUG: print(f"DEBUG: domain='{domain}',path={path} - CALLED!") - if type(domain) != str: + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - elif type(path) != str and path != None: + raise ValueError("Parameter 'domain' is empty") + elif not isinstance(path, str) and path is not None: raise ValueError(f"Parameter path[]={type(path)} is not 'str'") - # DEBUG: print("DEBUG: Fetching nodeinfo from domain,path:", domain, path) + # DEBUG: print(f"DEBUG: Fetching nodeinfo from domain='{domain}' ...") nodeinfo = fetch_wellknown_nodeinfo(domain) # DEBUG: print(f"DEBUG: nodeinfo({len(nodeinfo)})={nodeinfo}") @@ -723,64 +429,64 @@ def fetch_nodeinfo(domain: str, path: str = None) -> list: # DEBUG: print("DEBUG: nodeinfo()={len(nodeinfo))} - EXIT!") return nodeinfo - requests = [ - f"https://{domain}/nodeinfo/2.1.json", - f"https://{domain}/nodeinfo/2.1", - f"https://{domain}/nodeinfo/2.0.json", - f"https://{domain}/nodeinfo/2.0", - f"https://{domain}/nodeinfo/1.0", - f"https://{domain}/api/v1/instance" + request_paths = [ + "/nodeinfo/2.1.json", + "/nodeinfo/2.1", + "/nodeinfo/2.0.json", + "/nodeinfo/2.0", + "/nodeinfo/1.0", + "/api/v1/instance" ] data = {} - for request in requests: - if path != None and path != "" and request != path: + for request in request_paths: + if path is not None and path != "" and path != f"https://{domain}{path}": # DEBUG: print(f"DEBUG: path='{path}' does not match request='{request}' - SKIPPED!") continue try: - # DEBUG: print("DEBUG: Fetching request:", request) - res = reqto.get(request, headers=api_headers, timeout=(config.get("nodeinfo_connection_timeout"), config.get("nodeinfo_read_timeout"))) + # DEBUG: print(f"DEBUG: Fetching request='{request}' from domain='{domain}' ...") + response = network.fetch_response(domain, request, api_headers, (config.get("nodeinfo_connection_timeout"), config.get("nodeinfo_read_timeout"))) - data = json_from_response(res) - # DEBUG: print(f"DEBUG: res.ok={res.ok},res.status_code={res.status_code},data[]='{type(data)}'") - if res.ok and isinstance(data, dict): + data = json_from_response(response) + # DEBUG: print(f"DEBUG: response.ok={response.ok},response.status_code={response.status_code},data[]='{type(data)}'") + if response.ok and isinstance(data, dict): # DEBUG: print("DEBUG: Success:", request) instances.set("detection_mode", domain, "STATIC_CHECK") instances.set("nodeinfo_url" , domain, request) break - elif res.ok and isinstance(data, list): - # DEBUG: print(f"DEBUG: domain='{domain}' returned a list: '{data}'") + elif response.ok and isinstance(data, list): + print(f"UNSUPPORTED: domain='{domain}' returned a list: '{data}'") sys.exit(255) - elif not res.ok or res.status_code >= 400: + elif not response.ok or response.status_code >= 400: print("WARNING: Failed fetching nodeinfo from domain:", domain) - update_last_error(domain, res) + instances.update_last_error(domain, response) continue - except BaseException as e: + except BaseException as exception: # DEBUG: print("DEBUG: Cannot fetch API request:", request) - update_last_error(domain, e) + instances.update_last_error(domain, exception) pass # DEBUG: print(f"DEBUG: data()={len(data)} - EXIT!") return data def fetch_wellknown_nodeinfo(domain: str) -> list: - # DEBUG: print(f"DEBUG: domain={domain} - CALLED!") - if type(domain) != str: + # DEBUG: print(f"DEBUG: domain='{domain}' - CALLED!") + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError("Parameter 'domain' is empty") # DEBUG: print("DEBUG: Fetching .well-known info for domain:", domain) data = {} try: - res = reqto.get(f"https://{domain}/.well-known/nodeinfo", headers=api_headers, timeout=(config.get("nodeinfo_connection_timeout"), config.get("nodeinfo_read_timeout"))) + response = network.fetch_response(domain, "/.well-known/nodeinfo", api_headers, (config.get("nodeinfo_connection_timeout"), config.get("nodeinfo_read_timeout"))) - data = json_from_response(res) - # DEBUG: print("DEBUG: domain,res.ok,data[]:", domain, res.ok, type(data)) - if res.ok and isinstance(data, dict): + data = json_from_response(response) + # DEBUG: print("DEBUG: domain,response.ok,data[]:", domain, response.ok, type(data)) + if response.ok and isinstance(data, dict): nodeinfo = data # DEBUG: print("DEBUG: Found entries:", len(nodeinfo), domain) if "links" in nodeinfo: @@ -789,11 +495,11 @@ def fetch_wellknown_nodeinfo(domain: str) -> list: # DEBUG: print("DEBUG: rel,href:", link["rel"], link["href"]) if link["rel"] in nodeinfo_identifier: # DEBUG: print("DEBUG: Fetching nodeinfo from:", link["href"]) - res = reqto.get(link["href"]) + response = fetch_url(link["href"], api_headers, (config.get("connection_timeout"), config.get("read_timeout"))) - data = json_from_response(res) - # DEBUG: print("DEBUG: href,res.ok,res.status_code:", link["href"], res.ok, res.status_code) - if res.ok and isinstance(data, dict): + data = json_from_response(response) + # DEBUG: print("DEBUG: href,response.ok,response.status_code:", link["href"], response.ok, response.status_code) + if response.ok and isinstance(data, dict): # DEBUG: print("DEBUG: Found JSON nodeinfo():", len(data)) instances.set("detection_mode", domain, "AUTO_DISCOVERY") instances.set("nodeinfo_url" , domain, link["href"]) @@ -803,9 +509,9 @@ def fetch_wellknown_nodeinfo(domain: str) -> list: else: print("WARNING: nodeinfo does not contain 'links':", domain) - except BaseException as e: + except BaseException as exception: print("WARNING: Failed fetching .well-known info:", domain) - update_last_error(domain, e) + instances.update_last_error(domain, exception) pass # DEBUG: print("DEBUG: Returning data[]:", type(data)) @@ -813,26 +519,26 @@ def fetch_wellknown_nodeinfo(domain: str) -> list: def fetch_generator_from_path(domain: str, path: str = "/") -> str: # DEBUG: print(f"DEBUG: domain({len(domain)})={domain},path={path} - CALLED!") - if type(domain) != str: + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - elif type(path) != str: + raise ValueError("Parameter 'domain' is empty") + elif not isinstance(path, str): raise ValueError(f"path[]={type(path)} is not 'str'") elif path == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError("Parameter 'path' is empty") # DEBUG: print(f"DEBUG: domain='{domain}',path='{path}' - CALLED!") software = None try: # DEBUG: print(f"DEBUG: Fetching path='{path}' from '{domain}' ...") - res = reqto.get(f"https://{domain}{path}", headers=headers, timeout=(config.get("connection_timeout"), config.get("read_timeout"))) + response = network.fetch_response(domain, path, headers, (config.get("connection_timeout"), config.get("read_timeout"))) - # DEBUG: print("DEBUG: domain,res.ok,res.status_code,res.text[]:", domain, res.ok, res.status_code, type(res.text)) - if res.ok and res.status_code < 300 and len(res.text) > 0: + # DEBUG: print("DEBUG: domain,response.ok,response.status_code,response.text[]:", domain, response.ok, response.status_code, type(response.text)) + if response.ok and response.status_code < 300 and len(response.text) > 0: # DEBUG: print("DEBUG: Search for :", domain) - doc = bs4.BeautifulSoup(res.text, "html.parser") + doc = bs4.BeautifulSoup(response.text, "html.parser") # DEBUG: print("DEBUG: doc[]:", type(doc)) generator = doc.find("meta", {"name": "generator"}) @@ -852,30 +558,30 @@ def fetch_generator_from_path(domain: str, path: str = "/") -> str: instances.set("detection_mode", domain, "SITE_NAME") remove_pending_error(domain) - except BaseException as e: - # DEBUG: print(f"DEBUG: Cannot fetch / from '{domain}':", e) - update_last_error(domain, e) + except BaseException as exception: + # DEBUG: print(f"DEBUG: Cannot fetch / from '{domain}':", exception) + instances.update_last_error(domain, exception) pass # DEBUG: print(f"DEBUG: software[]={type(software)}") - if type(software) is str and software == "": + if isinstance(software, str) and software == "": # DEBUG: print(f"DEBUG: Corrected empty string to None for software of domain='{domain}'") software = None - elif type(software) is str and ("." in software or " " in software): + elif isinstance(software, str) and ("." in software or " " in software): # DEBUG: print(f"DEBUG: software='{software}' may contain a version number, domain='{domain}', removing it ...") software = remove_version(software) # DEBUG: print(f"DEBUG: software[]={type(software)}") - if type(software) is str and " powered by " in software: + if isinstance(software, str) and " powered by " in software: # DEBUG: print(f"DEBUG: software='{software}' has 'powered by' in it") software = remove_version(strip_powered_by(software)) - elif type(software) is str and " hosted on " in software: + elif isinstance(software, str) and " hosted on " in software: # DEBUG: print(f"DEBUG: software='{software}' has 'hosted on' in it") software = remove_version(strip_hosted_on(software)) - elif type(software) is str and " by " in software: + elif isinstance(software, str) and " by " in software: # DEBUG: print(f"DEBUG: software='{software}' has ' by ' in it") software = strip_until(software, " by ") - elif type(software) is str and " see " in software: + elif isinstance(software, str) and " see " in software: # DEBUG: print(f"DEBUG: software='{software}' has ' see ' in it") software = strip_until(software, " see ") @@ -884,11 +590,11 @@ def fetch_generator_from_path(domain: str, path: str = "/") -> str: def determine_software(domain: str, path: str = None) -> str: # DEBUG: print(f"DEBUG: domain({len(domain)})={domain},path={path} - CALLED!") - if type(domain) != str: + if not isinstance(domain, str): raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - elif type(path) != str and path != None: + raise ValueError("Parameter 'domain' is empty") + elif not isinstance(path, str) and path is not None: raise ValueError(f"Parameter path[]={type(path)} is not 'str'") # DEBUG: print("DEBUG: Determining software for domain,path:", domain, path) @@ -905,11 +611,11 @@ def determine_software(domain: str, path: str = None) -> str: # DEBUG: print("DEBUG: data():", len(data), data) if "status" in data and data["status"] == "error" and "message" in data: print("WARNING: JSON response is an error:", data["message"]) - update_last_error(domain, data["message"]) + instances.update_last_error(domain, data["message"]) return fetch_generator_from_path(domain) elif "message" in data: print("WARNING: JSON response contains only a message:", data["message"]) - update_last_error(domain, data["message"]) + instances.update_last_error(domain, data["message"]) return fetch_generator_from_path(domain) elif "software" not in data or "name" not in data["software"]: # DEBUG: print(f"DEBUG: JSON response from domain='{domain}' does not include [software][name], fetching / ...") @@ -932,17 +638,17 @@ def determine_software(domain: str, path: str = None) -> str: software = "misskey" elif software.find("/") > 0: print("WARNING: Spliting of slash:", software) - software = software.split("/")[-1]; + software = tidup_domain(software.split("/")[-1]); elif software.find("|") > 0: print("WARNING: Spliting of pipe:", software) software = tidyup_domain(software.split("|")[0]); elif "powered by" in software: # DEBUG: print(f"DEBUG: software='{software}' has 'powered by' in it") software = strip_powered_by(software) - elif type(software) is str and " by " in software: + elif isinstance(software, str) and " by " in software: # DEBUG: print(f"DEBUG: software='{software}' has ' by ' in it") software = strip_until(software, " by ") - elif type(software) is str and " see " in software: + elif isinstance(software, str) and " see " in software: # DEBUG: print(f"DEBUG: software='{software}' has ' see ' in it") software = strip_until(software, " see ") @@ -960,482 +666,31 @@ def determine_software(domain: str, path: str = None) -> str: software = remove_version(software) # DEBUG: print(f"DEBUG: software[]={type(software)}") - if type(software) is str and "powered by" in software: + if isinstance(software, str) and "powered by" in software: # DEBUG: print(f"DEBUG: software='{software}' has 'powered by' in it") software = remove_version(strip_powered_by(software)) # DEBUG: print("DEBUG: Returning domain,software:", domain, software) return software -def update_block_reason(reason: str, blocker: str, blocked: str, block_level: str): - # DEBUG: print(f"DEBUG: reason='{reason}',blocker={blocker},blocked={blocked},block_level={block_level} - CALLED!") - if type(reason) != str and reason != None: - raise ValueError(f"Parameter reason[]='{type(reason)}' is not 'str'") - elif type(blocker) != str: - raise ValueError(f"Parameter blocker[]='{type(blocker)}' is not 'str'") - elif type(blocked) != str: - raise ValueError(f"Parameter blocked[]='{type(blocked)}' is not 'str'") - elif type(block_level) != str: - raise ValueError(f"Parameter block_level[]='{type(block_level)}' is not 'str'") - - # DEBUG: print("DEBUG: Updating block reason:", reason, blocker, blocked, block_level) - try: - cursor.execute( - "UPDATE blocks SET reason = ?, last_seen = ? WHERE blocker = ? AND blocked = ? AND block_level = ? AND reason IN ('','unknown') LIMIT 1", - ( - reason, - time.time(), - blocker, - blocked, - block_level - ), - ) - - # DEBUG: print(f"DEBUG: cursor.rowcount={cursor.rowcount}") - if cursor.rowcount == 0: - # DEBUG: print(f"DEBUG: Did not update any rows: blocker='{blocker}',blocked='{blocked}',block_level='{block_level}',reason='{reason}' - EXIT!") - return - - except BaseException as e: - print(f"ERROR: failed SQL query: reason='{reason}',blocker='{blocker}',blocked='{blocked}',block_level='{block_level}',exception[{type(e)}]:'{str(e)}'") - sys.exit(255) - - # DEBUG: print("DEBUG: EXIT!") - -def update_last_seen(blocker: str, blocked: str, block_level: str): - # DEBUG: print("DEBUG: Updating last_seen for:", blocker, blocked, block_level) - try: - cursor.execute( - "UPDATE blocks SET last_seen = ? WHERE blocker = ? AND blocked = ? AND block_level = ? LIMIT 1", - ( - time.time(), - blocker, - blocked, - block_level - ) - ) - - # DEBUG: print(f"DEBUG: cursor.rowcount={cursor.rowcount}") - if cursor.rowcount == 0: - # DEBUG: print(f"DEBUG: Did not update any rows: blocker='{blocker}',blocked='{blocked}',block_level='{block_level}' - EXIT!") - return - - except BaseException as e: - print(f"ERROR: failed SQL query: blocker='{blocker}',blocked='{blocked}',block_level='{block_level}',exception[{type(e)}]:'{str(e)}'") - sys.exit(255) - - # DEBUG: print("DEBUG: EXIT!") - -def is_instance_blocked(blocker: str, blocked: str, block_level: str) -> bool: - # DEBUG: print(f"DEBUG: blocker={blocker},blocked={blocked},block_level={block_level} - CALLED!") - if type(blocker) != str: - raise ValueError(f"Parameter blocker[]={type(blocker)} is not of type 'str'") - elif blocker == "": - raise ValueError("Parameter 'blocker' cannot be empty") - elif type(blocked) != str: - raise ValueError(f"Parameter blocked[]={type(blocked)} is not of type 'str'") - elif blocked == "": - raise ValueError("Parameter 'blocked' cannot be empty") - elif type(block_level) != str: - raise ValueError(f"Parameter block_level[]={type(block_level)} is not of type 'str'") - elif block_level == "": - raise ValueError("Parameter 'block_level' cannot be empty") - - cursor.execute( - "SELECT * FROM blocks WHERE blocker = ? AND blocked = ? AND block_level = ? LIMIT 1", - ( - blocker, - blocked, - block_level - ), - ) - - is_blocked = cursor.fetchone() != None - - # DEBUG: print(f"DEBUG: is_blocked='{is_blocked}' - EXIT!") - return is_blocked - -def block_instance(blocker: str, blocked: str, reason: str, block_level: str): - # DEBUG: print("DEBUG: blocker,blocked,reason,block_level:", blocker, blocked, reason, block_level) - if type(blocker) != str: - raise ValueError(f"Parameter blocker[]={type(blocker)} is not 'str'") - elif blocker == "": - raise ValueError(f"Parameter 'blocker' cannot be empty") - elif not validators.domain(blocker.split("/")[0]): - raise ValueError(f"Bad blocker='{blocker}'") - elif type(blocked) != str: - raise ValueError(f"Parameter blocked[]={type(blocked)} is not 'str'") - elif blocked == "": - raise ValueError(f"Parameter 'blocked' cannot be empty") - elif not validators.domain(blocked.split("/")[0]): - raise ValueError(f"Bad blocked='{blocked}'") - elif is_blacklisted(blocker): - raise Exception(f"blocker='{blocker}' is blacklisted but function invoked") - elif is_blacklisted(blocked): - raise Exception(f"blocked='{blocked}' is blacklisted but function invoked") - - print(f"INFO: New block: blocker='{blocker}',blocked='{blocked}', reason='{reason}', block_level='{block_level}'") - try: - cursor.execute( - "INSERT INTO blocks (blocker, blocked, reason, block_level, first_seen, last_seen) VALUES(?, ?, ?, ?, ?, ?)", - ( - blocker, - blocked, - reason, - block_level, - time.time(), - time.time() - ), - ) - except BaseException as e: - print(f"ERROR: failed SQL query: blocker='{blocker}',blocked='{blocked}',reason='{reason}',block_level='{block_level}',exception[{type(e)}]:'{str(e)}'") - sys.exit(255) - - # DEBUG: print("DEBUG: EXIT!") - -def is_instance_registered(domain: str) -> bool: - # DEBUG: print(f"DEBUG: domain={domain} - CALLED!") - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - # NOISY-DEBUG: # DEBUG: print(f"DEBUG: domain='{domain}' - CALLED!") - if not cache.key_exists("is_registered"): - # NOISY-DEBUG: # DEBUG: print(f"DEBUG: Cache for 'is_registered' not initialized, fetching all rows ...") - try: - cursor.execute("SELECT domain FROM instances") - - # Check Set all - cache.set_all("is_registered", cursor.fetchall(), True) - except BaseException as e: - print(f"ERROR: failed SQL query: domain='{domain}',exception[{type(e)}]:'{str(e)}'") - sys.exit(255) - - # Is cache found? - registered = cache.sub_key_exists("is_registered", domain) - - # NOISY-DEBUG: # DEBUG: print(f"DEBUG: registered='{registered}' - EXIT!") - return registered - -def add_instance(domain: str, origin: str, originator: str, path: str = None): - # DEBUG: print(f"DEBUG: domain={domain},origin={origin},originator={originator},path={path} - CALLED!") - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - elif type(origin) != str and origin != None: - raise ValueError(f"origin[]={type(origin)} is not 'str'") - elif type(originator) != str: - raise ValueError(f"originator[]={type(originator)} is not 'str'") - elif originator == "": - raise ValueError(f"originator cannot be empty") - elif not validators.domain(domain.split("/")[0]): - raise ValueError(f"Bad domain name='{domain}'") - elif origin is not None and not validators.domain(origin.split("/")[0]): - raise ValueError(f"Bad origin name='{origin}'") - elif is_blacklisted(domain): - raise Exception(f"domain='{domain}' is blacklisted, but method invoked") - - # DEBUG: print("DEBUG: domain,origin,originator,path:", domain, origin, originator, path) - software = determine_software(domain, path) - # DEBUG: print("DEBUG: Determined software:", software) - - print(f"INFO: Adding instance domain='{domain}' (origin='{origin}',software='{software}')") - try: - cursor.execute( - "INSERT INTO instances (domain, origin, originator, hash, software, first_seen) VALUES (?, ?, ?, ?, ?, ?)", - ( - domain, - origin, - originator, - get_hash(domain), - software, - time.time() - ), - ) - - cache.set_sub_key("is_registered", domain, True) - - if instances.has_pending_instance_data(domain): - # DEBUG: print(f"DEBUG: domain='{domain}' has pending nodeinfo being updated ...") - instances.set("last_status_code" , domain, None) - instances.set("last_error_details", domain, None) - instances.update_instance_data(domain) - remove_pending_error(domain) - - if domain in pending_errors: - # DEBUG: print("DEBUG: domain has pending error being updated:", domain) - update_last_error(domain, pending_errors[domain]) - remove_pending_error(domain) - - except BaseException as e: - print(f"ERROR: failed SQL query: domain='{domain}',exception[{type(e)}]:'{str(e)}'") - sys.exit(255) - else: - # DEBUG: print("DEBUG: Updating nodeinfo for domain:", domain) - update_last_nodeinfo(domain) - - # DEBUG: print("DEBUG: EXIT!") - -def send_bot_post(instance: str, blocks: dict): - # DEBUG: print(f"DEBUG: instance={instance},blocks()={len(blocks)} - CALLED!") - message = instance + " has blocked the following instances:\n\n" - truncated = False - - if len(blocks) > 20: - truncated = True - blocks = blocks[0 : 19] - - for block in blocks: - if block["reason"] == None or block["reason"] == '': - message = message + block["blocked"] + " with unspecified reason\n" - else: - if len(block["reason"]) > 420: - block["reason"] = block["reason"][0:419] + "[…]" +def tidyup_reason(reason: str) -> str: + # DEBUG: print(f"DEBUG: reason='{reason}' - CALLED!") + if not isinstance(reason, str): + raise ValueError(f"Parameter reason[]={type(reason)} is not 'str'") - message = message + block["blocked"] + ' for "' + block["reason"].replace("@", "@\u200b") + '"\n' + # Strip string + reason = reason.strip() - if truncated: - message = message + "(the list has been truncated to the first 20 entries)" + # Replace â with " + reason = re.sub("â", "\"", reason) - botheaders = {**api_headers, **{"Authorization": "Bearer " + config.get("bot_token")}} - - req = reqto.post( - f"{config.get('bot_instance')}/api/v1/statuses", - data={ - "status" : message, - "visibility" : config.get('bot_visibility'), - "content_type": "text/plain" - }, - headers=botheaders, - timeout=10 - ).json() - - return True - -def get_mastodon_blocks(domain: str) -> dict: - # DEBUG: print(f"DEBUG: domain={domain} - CALLED!") - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - # DEBUG: print("DEBUG: Fetching mastodon blocks from domain:", domain) - blocks = { - "Suspended servers": [], - "Filtered media" : [], - "Limited servers" : [], - "Silenced servers" : [], - } - - try: - doc = bs4.BeautifulSoup( - reqto.get(f"https://{domain}/about", headers=headers, timeout=(config.get("connection_timeout"), config.get("read_timeout"))).text, - "html.parser", - ) - except BaseException as e: - print("ERROR: Cannot fetch from domain:", domain, e) - update_last_error(domain, e) - return {} - - for header in doc.find_all("h3"): - header_text = tidyup_domain(header.text) - - if header_text in language_mapping: - # DEBUG: print(f"DEBUG: header_text='{header_text}'") - header_text = language_mapping[header_text] - - if header_text in blocks or header_text.lower() in blocks: - # 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:]: - blocks[header_text].append( - { - "domain": tidyup_domain(line.find("span").text), - "hash" : tidyup_domain(line.find("span")["title"][9:]), - "reason": tidyup_domain(line.find_all("td")[1].text), - } - ) - - # DEBUG: print("DEBUG: Returning blocks for domain:", domain) - return { - "reject" : blocks["Suspended servers"], - "media_removal" : blocks["Filtered media"], - "followers_only": blocks["Limited servers"] + blocks["Silenced servers"], - } - -def get_friendica_blocks(domain: str) -> dict: - # DEBUG: print(f"DEBUG: domain={domain} - CALLED!") - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - # DEBUG: print("DEBUG: Fetching friendica blocks from domain:", domain) - blocks = [] - - try: - doc = bs4.BeautifulSoup( - reqto.get(f"https://{domain}/friendica", headers=headers, timeout=(config.get("connection_timeout"), config.get("read_timeout"))).text, - "html.parser", - ) - except BaseException as e: - print("WARNING: Failed to fetch /friendica from domain:", domain, e) - update_last_error(domain, e) - return {} - - blocklist = doc.find(id="about_blocklist") - - # Prevents exceptions: - if blocklist is None: - # DEBUG: print("DEBUG: Instance has no block list:", domain) - return {} - - for line in blocklist.find("table").find_all("tr")[1:]: - # DEBUG: print(f"DEBUG: line='{line}'") - blocks.append({ - "domain": tidyup_domain(line.find_all("td")[0].text), - "reason": tidyup_domain(line.find_all("td")[1].text) - }) - - # DEBUG: print("DEBUG: Returning blocks() for domain:", domain, len(blocks)) - return { - "reject": blocks - } - -def get_misskey_blocks(domain: str) -> dict: - # DEBUG: print(f"DEBUG: domain={domain} - CALLED!") - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") - elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") - - # DEBUG: print("DEBUG: Fetching misskey blocks from domain:", domain) - blocks = { - "suspended": [], - "blocked" : [] - } - - offset = 0 - step = config.get("misskey_limit") - while True: - # iterating through all "suspended" (follow-only in its terminology) - # instances page-by-page, since that troonware doesn't support - # sending them all at once - try: - # DEBUG: print(f"DEBUG: Fetching offset='{offset}' from '{domain}' ...") - if offset == 0: - # DEBUG: print("DEBUG: Sending JSON API request to domain,step,offset:", domain, step, offset) - fetched = post_json_api(domain, "/api/federation/instances", json.dumps({ - "sort" : "+pubAt", - "host" : None, - "suspended": True, - "limit" : step - }), { - "Origin": domain - }) - else: - # DEBUG: print("DEBUG: Sending JSON API request to domain,step,offset:", domain, step, offset) - fetched = post_json_api(domain, "/api/federation/instances", json.dumps({ - "sort" : "+pubAt", - "host" : None, - "suspended": True, - "limit" : step, - "offset" : offset - 1 - }), { - "Origin": domain - }) - - # DEBUG: print("DEBUG: fetched():", len(fetched)) - if len(fetched) == 0: - # DEBUG: print("DEBUG: Returned zero bytes, exiting loop:", domain) - break - elif len(fetched) != config.get("misskey_limit"): - # DEBUG: print(f"DEBUG: Fetched '{len(fetched)}' row(s) but expected: '{config.get('misskey_limit')}'") - offset = offset + (config.get("misskey_limit") - len(fetched)) - else: - # DEBUG: print("DEBUG: Raising offset by step:", step) - offset = offset + step - - for instance in fetched: - # just in case - if instance["isSuspended"]: - blocks["suspended"].append( - { - "domain": tidyup_domain(instance["host"]), - # no reason field, nothing - "reason": None - } - ) - - except BaseException as e: - print("WARNING: Caught error, exiting loop:", domain, e) - update_last_error(domain, e) - offset = 0 - break - - while True: - # same shit, different asshole ("blocked" aka full suspend) - try: - if offset == 0: - # DEBUG: print("DEBUG: Sending JSON API request to domain,step,offset:", domain, step, offset) - fetched = post_json_api(domain,"/api/federation/instances", json.dumps({ - "sort" : "+pubAt", - "host" : None, - "blocked": True, - "limit" : step - }), { - "Origin": domain - }) - else: - # DEBUG: print("DEBUG: Sending JSON API request to domain,step,offset:", domain, step, offset) - fetched = post_json_api(domain,"/api/federation/instances", json.dumps({ - "sort" : "+pubAt", - "host" : None, - "blocked": True, - "limit" : step, - "offset" : offset - 1 - }), { - "Origin": domain - }) - - # DEBUG: print("DEBUG: fetched():", len(fetched)) - if len(fetched) == 0: - # DEBUG: print("DEBUG: Returned zero bytes, exiting loop:", domain) - break - elif len(fetched) != config.get("misskey_limit"): - # DEBUG: print(f"DEBUG: Fetched '{len(fetched)}' row(s) but expected: '{config.get('misskey_limit')}'") - offset = offset + (config.get("misskey_limit") - len(fetched)) - else: - # DEBUG: print("DEBUG: Raising offset by step:", step) - offset = offset + step - - for instance in fetched: - if instance["isBlocked"]: - blocks["blocked"].append({ - "domain": tidyup_domain(instance["host"]), - "reason": None - }) - - except BaseException as e: - print("ERROR: Exception during POST:", domain, e) - update_last_error(domain, e) - offset = 0 - break - - # DEBUG: print(f"DEBUG: Updating last_instance_fetch for domain='{domain}' ...") - update_last_instance_fetch(domain) - - # DEBUG: print("DEBUG: Returning for domain,blocked(),suspended():", domain, len(blocks["blocked"]), len(blocks["suspended"])) - return { - "reject" : blocks["blocked"], - "followers_only": blocks["suspended"] - } + # DEBUG: print(f"DEBUG: reason='{reason}' - EXIT!") + return reason def tidyup_domain(domain: str) -> str: # DEBUG: print(f"DEBUG: domain='{domain}' - CALLED!") - if type(domain) != str: - raise ValueError(f"Parameter domain[]={type(domain)} is not expected") + if not isinstance(domain, str): + raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") # All lower-case and strip spaces out + last dot domain = domain.lower().strip().rstrip(".") @@ -1454,6 +709,10 @@ def tidyup_domain(domain: str) -> str: # No individual users in block lists domain = re.sub("(.+)\@", "", domain) + if domain.find("/profile/"): + domain = domain.split("/profile/")[0] + elif domain.find("/users/"): + domain = domain.split("/users/")[0] # DEBUG: print(f"DEBUG: domain='{domain}' - EXIT!") return domain @@ -1473,3 +732,103 @@ def json_from_response(response: requests.models.Response) -> list: # DEBUG: print(f"DEBUG: data[]={type(data)} - EXIT!") return data + +def has_key(lists: list, key: str, value: any) -> bool: + # DEBUG: print(f"DEBUG: lists()={len(lists)},key='{key}',value[]='{type(value)}' - CALLED!") + if not isinstance(lists, list): + raise ValueError(f"Parameter lists[]='{type(lists)}' is not 'list'") + elif not isinstance(key, str): + raise ValueError(f"Parameter key[]='{type(key)}' is not 'str'") + elif key == "": + raise ValueError("Parameter 'key' is empty") + + has = False + # DEBUG: print(f"DEBUG: Checking lists()={len(lists)} ...") + for row in lists: + # DEBUG: print(f"DEBUG: row['{type(row)}']={row}") + if not isinstance(row, dict): + raise ValueError(f"row[]='{type(row)}' is not 'dict'") + elif not key in row: + raise KeyError(f"Cannot find key='{key}'") + elif row[key] == value: + has = True + break + + # DEBUG: print(f"DEBUG: has={has} - EXIT!") + return has + +def find_domains(tag: bs4.element.Tag) -> list: + # DEBUG: print(f"DEBUG: tag[]={type(tag)} - CALLED!") + if not isinstance(tag, bs4.element.Tag): + raise ValueError(f"Parameter tag[]={type(tag)} is not type of bs4.element.Tag") + elif not isinstance(tag, bs4.element.Tag): + raise KeyError("Cannot find table with instances!") + elif len(tag.select("tr")) == 0: + raise KeyError("No table rows found in table!") + + domains = list() + for element in tag.select("tr"): + # DEBUG: print(f"DEBUG: element[]={type(element)}") + if not element.find("td"): + # DEBUG: print("DEBUG: Skipping element, no found") + continue + + domain = tidyup_domain(element.find("td").text) + reason = tidyup_reason(element.findAll("td")[1].text) + + # DEBUG: print(f"DEBUG: domain='{domain}',reason='{reason}'") + + if blacklist.is_blacklisted(domain): + print(f"WARNING: domain='{domain}' is blacklisted - skipped!") + continue + elif domain == "gab.com/.ai, develop.gab.com": + # DEBUG: print(f"DEBUG: Multiple domains detected in one row") + domains.append({ + "domain": "gab.com", + "reason": reason, + }) + domains.append({ + "domain": "gab.ai", + "reason": reason, + }) + domains.append({ + "domain": "develop.gab.com", + "reason": reason, + }) + continue + elif not validators.domain(domain): + print(f"WARNING: domain='{domain}' is not a valid domain - skipped!") + continue + + # DEBUG: print(f"DEBUG: Adding domain='{domain}' ...") + domains.append({ + "domain": domain, + "reason": reason, + }) + + # DEBUG: print(f"DEBUG: domains()={len(domains)} - EXIT!") + return domains + +def fetch_url(url: str, headers: dict, timeout: list) -> requests.models.Response: + # DEBUG: print(f"DEBUG: url='{url}',headers()={len(headers)},timeout={timeout} - CALLED!") + if not isinstance(url, str): + raise ValueError(f"Parameter url[]='{type(url)}' is not 'str'") + elif url == "": + raise ValueError("Parameter 'url' is empty") + elif not isinstance(headers, dict): + raise ValueError(f"Parameter headers[]='{type(headers)}' is not 'dict'") + elif not isinstance(timeout, list): + raise ValueError(f"Parameter timeout[]='{type(timeout)}' is not 'list'") + + # DEBUG: print(f"DEBUG: Parsing url='{url}'") + components = urlparse(url) + + # Invoke other function, avoid trailing ? + # DEBUG: print(f"DEBUG: components[{type(components)}]={components}") + if components.query != "": + response = network.fetch_response(components.hostname, f"{components.path}?{components.query}", headers, timeout) + else: + response = network.fetch_response(components.hostname, f"{components.path}", headers, timeout) + + # DEBUG: print(f"DEBUG: response[]='{type(response)}' - EXXIT!") + return response