X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=fba%2Ffba.py;h=e52ddc36eeaaad55f4621d5bcc437359c993a89c;hb=d9577fe6ef2961b8fa1b256d3535ecf23748d96a;hp=d4df7c8037334ae26e593279a43b679b66eed26a;hpb=2c4499be32b730eae02187a6af5446bf49e26f3f;p=fba.git diff --git a/fba/fba.py b/fba/fba.py index d4df7c8..e52ddc3 100644 --- a/fba/fba.py +++ b/fba/fba.py @@ -23,30 +23,18 @@ import json import sqlite3 import sys import time -import urllib import validators +from urllib.parse import urlparse + +from fba import blacklist from fba import cache from fba import config from fba import instances -# 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 = { @@ -75,34 +63,6 @@ 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() @@ -122,35 +82,35 @@ patterns = [ ##### Other functions ##### def is_primitive(var: any) -> bool: - # NOISY-DEBUG: # DEBUG: print(f"DEBUG: var[]='{type(var)}' - CALLED!") + # DEBUG: print(f"DEBUG: var[]='{type(var)}' - CALLED!") return type(var) in {int, str, float, bool} or var == None def fetch_instances(domain: str, origin: str, software: str, script: str, path: str = None): - # DEBUG: print(f"DEBUG: domain={domain},origin={origin},software={software},path={path} - CALLED!") + # DEBUG: print(f"DEBUG: domain='{domain}',origin='{origin}',software='{software}',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") + raise ValueError(f"Parameter 'domain' is empty") elif type(origin) != str and origin != None: raise ValueError(f"Parameter origin[]={type(origin)} is not 'str'") elif type(script) != str: raise ValueError(f"Parameter script[]={type(script)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError(f"Parameter 'domain' is empty") - 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: @@ -168,15 +128,15 @@ 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]) + instances.add(instance, domain, script) except BaseException as e: print(f"ERROR: instance='{instance}',exception[{type(e)}]:'{str(e)}'") continue @@ -186,16 +146,16 @@ 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] != 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 @@ -321,24 +281,11 @@ 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: raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError(f"Parameter 'domain' is empty") try: # Prevent updating any pending errors, nodeinfo was found @@ -353,31 +300,16 @@ def get_hash(domain: str) -> str: 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(f"Parameter 'domain' is empty") return hashlib.sha256(domain.encode("utf-8")).hexdigest() -def update_last_blocked(domain: str): - 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_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, response: requests.models.Response): # DEBUG: print("DEBUG: domain,response[]:", domain, type(response)) 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(f"Parameter 'domain' is empty") try: # DEBUG: print("DEBUG: BEFORE response[]:", type(response)) @@ -408,249 +340,42 @@ def log_error(domain: str, response: requests.models.Response): # DEBUG: print("DEBUG: EXIT!") -def update_last_error(domain: str, response: requests.models.Response): - # DEBUG: print("DEBUG: domain,response[]:", domain, type(response)) - 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 response[]:", type(response)) - if isinstance(response, BaseException) or isinstance(response, json.decoder.JSONDecodeError): - response = f"{type}:str(response)" - - # DEBUG: print("DEBUG: AFTER response[]:", type(response)) - if type(response) is str: - # DEBUG: print(f"DEBUG: Setting last_error_details='{response}'"); - instances.set("last_status_code" , domain, 999) - instances.set("last_error_details", domain, response) - else: - # DEBUG: print(f"DEBUG: Setting last_error_details='{response.reason}'"); - instances.set("last_status_code" , domain, response.status_code) - instances.set("last_error_details", domain, response.reason) - - # Running pending updated - # DEBUG: print(f"DEBUG: Invoking instances.update_instance_data({domain}) ...") - instances.update_instance_data(domain) - - log_error(domain, response) - - # 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: raise ValueError(f"Parameter domain[]={type(domain)} is not 'str'") elif domain == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError(f"Parameter 'domain' is empty") elif type(software) != str and software != 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: - response = get_response(domain, "/api/v3/site", api_headers, (config.get("connection_timeout"), config.get("read_timeout"))) - - 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, 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: - # 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, response) - - 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: - response = get_response(domain, "/api/v1/server/{mode}?start={start}&count=100", headers, (config.get("connection_timeout"), config.get("read_timeout"))) - - 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, 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: - response = get_response(domain, get_peers_url, api_headers, (config.get("connection_timeout"), config.get("read_timeout"))) + response = get_response(domain, "/api/v1/instance/peers", api_headers, (config.get("connection_timeout"), config.get("read_timeout"))) 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: - # DEBUG: print(f"DEBUG: Was not able to fetch '{get_peers_url}', trying alternative ...") + # DEBUG: print(f"DEBUG: Was not able to fetch peers, trying alternative ...") response = get_response(domain, "/api/v3/site", api_headers, (config.get("connection_timeout"), config.get("read_timeout"))) 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, response) + 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) @@ -660,29 +385,30 @@ 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, response) + 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) + instances.update_last_error(domain, 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) + 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: + # DEBUG: print(f"DEBUG: domain='{domain}',path='{path}',parameter='{parameter}',extra_headers()={len(extra_headers)} - 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") + raise ValueError(f"Parameter 'domain' is empty") elif type(path) != str: raise ValueError(f"path[]={type(path)} is not 'str'") elif path == "": @@ -704,7 +430,7 @@ def post_json_api(domain: str, path: str, parameter: str, extra_headers: dict = # 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(f"WARNING: Cannot query JSON API: domain='{domain}',path='{path}',parameter()={len(parameter)},response.status_code='{response.status_code}',data[]='{type(data)}'") - update_last_error(domain, response) + instances.update_last_error(domain, response) except BaseException as e: print(f"WARNING: Some error during post(): domain='{domain}',path='{path}',parameter()={len(parameter)},exception[{type(e)}]:'{str(e)}'") @@ -717,7 +443,7 @@ def fetch_nodeinfo(domain: str, path: str = None) -> list: 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(f"Parameter 'domain' is empty") elif type(path) != str and path != None: raise ValueError(f"Parameter path[]={type(path)} is not 'str'") @@ -760,23 +486,23 @@ def fetch_nodeinfo(domain: str, path: str = None) -> list: sys.exit(255) elif not response.ok or response.status_code >= 400: print("WARNING: Failed fetching nodeinfo from domain:", domain) - update_last_error(domain, response) + instances.update_last_error(domain, response) continue except BaseException as e: # DEBUG: print("DEBUG: Cannot fetch API request:", request) - update_last_error(domain, e) + instances.update_last_error(domain, e) 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!") + # 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") + raise ValueError(f"Parameter 'domain' is empty") # DEBUG: print("DEBUG: Fetching .well-known info for domain:", domain) data = {} @@ -795,7 +521,7 @@ 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"]) - response = get_url(link["href"]) + response = get_url(link["href"], api_headers, (config.get("connection_timeout"), config.get("read_timeout"))) data = json_from_response(response) # DEBUG: print("DEBUG: href,response.ok,response.status_code:", link["href"], response.ok, response.status_code) @@ -811,7 +537,7 @@ def fetch_wellknown_nodeinfo(domain: str) -> list: except BaseException as e: print("WARNING: Failed fetching .well-known info:", domain) - update_last_error(domain, e) + instances.update_last_error(domain, e) pass # DEBUG: print("DEBUG: Returning data[]:", type(data)) @@ -822,11 +548,11 @@ def fetch_generator_from_path(domain: str, path: str = "/") -> str: 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(f"Parameter 'domain' is empty") elif type(path) != str: raise ValueError(f"path[]={type(path)} is not 'str'") elif path == "": - raise ValueError(f"Parameter 'domain' cannot be empty") + raise ValueError(f"Parameter 'domain' is empty") # DEBUG: print(f"DEBUG: domain='{domain}',path='{path}' - CALLED!") software = None @@ -860,7 +586,7 @@ def fetch_generator_from_path(domain: str, path: str = "/") -> str: except BaseException as e: # DEBUG: print(f"DEBUG: Cannot fetch / from '{domain}':", e) - update_last_error(domain, e) + instances.update_last_error(domain, e) pass # DEBUG: print(f"DEBUG: software[]={type(software)}") @@ -893,7 +619,7 @@ def determine_software(domain: str, path: str = None) -> str: 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(f"Parameter 'domain' is empty") elif type(path) != str and path != None: raise ValueError(f"Parameter path[]={type(path)} is not 'str'") @@ -911,11 +637,11 @@ 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 / ...") @@ -938,7 +664,7 @@ 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]); @@ -973,231 +699,25 @@ def determine_software(domain: str, path: str = None) -> str: # 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") - - if reason != None: - # Maybe needs cleaning - reason = tidyup_reason(reason) - - 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!") +def send_bot_post(instance: str, blocklist: dict): + # DEBUG: print(f"DEBUG: instance={instance},blocklist()={len(blocklist)} - 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(f"DEBUG: domain='{domain}' - CALLED!") - if not cache.key_exists("is_registered"): - # 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) - - # 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'") + 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() - ), - ) + raise ValueError("Parameter 'domain' is empty") + elif type(blocklist) != dict: + raise ValueError(f"Parameter blocklist[]='{type(blocklist)}' is not 'dict'") - 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: + if len(blocklist) > 20: truncated = True - blocks = blocks[0 : 19] + blocklist = blocklist[0 : 19] - for block in blocks: + # DEBUG: print(f"DEBUG: blocklist()={len(blocklist)}") + for block in blocklist: + # DEBUG: print(f"DEBUG: block['{type(block)}']={block}") if block["reason"] == None or block["reason"] == '': message = message + block["blocked"] + " with unspecified reason\n" else: @@ -1224,65 +744,15 @@ def send_bot_post(instance: str, blocks: dict): 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( - get_response(domain, "/about", headers, (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!") +def fetch_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") + raise ValueError(f"Parameter 'domain' is empty") # DEBUG: print("DEBUG: Fetching friendica blocks from domain:", domain) - blocks = [] + blocked = list() try: doc = bs4.BeautifulSoup( @@ -1291,7 +761,7 @@ def get_friendica_blocks(domain: str) -> dict: ) except BaseException as e: print("WARNING: Failed to fetch /friendica from domain:", domain, e) - update_last_error(domain, e) + instances.update_last_error(domain, e) return {} blocklist = doc.find(id="about_blocklist") @@ -1301,27 +771,37 @@ def get_friendica_blocks(domain: str) -> dict: # DEBUG: print("DEBUG: Instance has no block list:", domain) return {} - for line in blocklist.find("table").find_all("tr")[1:]: + table = blocklist.find("table") + + # DEBUG: print(f"DEBUG: table[]='{type(table)}'") + if table.find("tbody"): + rows = table.find("tbody").find_all("tr") + else: + rows = table.find_all("tr") + + # DEBUG: print(f"DEBUG: Found rows()={len(rows)}") + for line in rows: # DEBUG: print(f"DEBUG: line='{line}'") - blocks.append({ + blocked.append({ "domain": tidyup_domain(line.find_all("td")[0].text), - "reason": tidyup_domain(line.find_all("td")[1].text) + "reason": tidyup_reason(line.find_all("td")[1].text) }) + # DEBUG: print("DEBUG: Next!") - # DEBUG: print("DEBUG: Returning blocks() for domain:", domain, len(blocks)) + # DEBUG: print("DEBUG: Returning blocklist() for domain:", domain, len(blocklist)) return { - "reject": blocks + "reject": blocked } -def get_misskey_blocks(domain: str) -> dict: - # DEBUG: print(f"DEBUG: domain={domain} - CALLED!") +def fetch_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") + raise ValueError(f"Parameter 'domain' is empty") # DEBUG: print("DEBUG: Fetching misskey blocks from domain:", domain) - blocks = { + blocklist = { "suspended": [], "blocked" : [] } @@ -1367,10 +847,12 @@ def get_misskey_blocks(domain: str) -> dict: # DEBUG: print("DEBUG: Raising offset by step:", step) offset = offset + step + count = 0 for instance in fetched: - # just in case - if instance["isSuspended"] and not has_element(blocks["suspended"], "domain", instance): - blocks["suspended"].append( + # Is it there? + if instance["isSuspended"] and not has_key(blocklist["suspended"], "domain", instance): + count = count + 1 + blocklist["suspended"].append( { "domain": tidyup_domain(instance["host"]), # no reason field, nothing @@ -1378,9 +860,14 @@ def get_misskey_blocks(domain: str) -> dict: } ) + # DEBUG: print(f"DEBUG: count={count}") + if count == 0: + # DEBUG: print(f"DEBUG: API is no more returning new instances, aborting loop!") + break + except BaseException as e: print("WARNING: Caught error, exiting loop:", domain, e) - update_last_error(domain, e) + instances.update_last_error(domain, e) offset = 0 break @@ -1420,26 +907,34 @@ def get_misskey_blocks(domain: str) -> dict: # DEBUG: print("DEBUG: Raising offset by step:", step) offset = offset + step + count = 0 for instance in fetched: - if instance["isBlocked"] and not has_element(blocks["blocked"], "domain", instance): - blocks["blocked"].append({ + # Is it there? + if instance["isBlocked"] and not has_key(blocklist["blocked"], "domain", instance): + count = count + 1 + blocklist["blocked"].append({ "domain": tidyup_domain(instance["host"]), "reason": None }) + # DEBUG: print(f"DEBUG: count={count}") + if count == 0: + # DEBUG: print(f"DEBUG: API is no more returning new instances, aborting loop!") + break + except BaseException as e: print("ERROR: Exception during POST:", domain, e) - update_last_error(domain, e) + instances.update_last_error(domain, e) offset = 0 break # 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 for domain,blocked(),suspended():", domain, len(blocks["blocked"]), len(blocks["suspended"])) + # DEBUG: print("DEBUG: Returning for domain,blocked(),suspended():", domain, len(blocklist["blocked"]), len(blocklist["suspended"])) return { - "reject" : blocks["blocked"], - "followers_only": blocks["suspended"] + "reject" : blocklist["blocked"], + "followers_only": blocklist["suspended"] } def tidyup_reason(reason: str) -> str: @@ -1453,7 +948,7 @@ def tidyup_reason(reason: str) -> str: # Replace â with " reason = re.sub("â", "\"", reason) - ## DEBUG: print(f"DEBUG: reason='{reason}' - EXIT!") + # DEBUG: print(f"DEBUG: reason='{reason}' - EXIT!") return reason def tidyup_domain(domain: str) -> str: @@ -1478,6 +973,10 @@ 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 @@ -1503,11 +1002,11 @@ def get_response(domain: str, path: str, headers: dict, timeout: list) -> reques if type(domain) != str: raise ValueError(f"Parameter domain[]='{type(domain)}' is not 'str'") elif domain == "": - raise ValueError("Parameter 'domain' cannot be empty") + raise ValueError("Parameter 'domain' is empty") elif type(path) != str: raise ValueError(f"Parameter path[]='{type(path)}' is not 'str'") elif path == "": - raise ValueError("Parameter 'path' cannot be empty") + raise ValueError("Parameter 'path' is empty") try: # DEBUG: print(f"DEBUG: Sending request to '{domain}{path}' ...") @@ -1518,28 +1017,30 @@ def get_response(domain: str, path: str, headers: dict, timeout: list) -> reques ); except requests.exceptions.ConnectionError as e: # DEBUG: print(f"DEBUG: Fetching '{path}' from '{domain}' failed. exception[{type(e)}]='{str(e)}'") - update_last_error(domain, e) + instances.update_last_error(domain, e) raise e # DEBUG: print(f"DEBUG: response[]='{type(response)}' - EXXIT!") return response -def has_element(elements: list, key: str, value: any) -> bool: - # DEBUG: print(f"DEBUG: element()={len(element)},key='{key}',value[]='{type(value)}' - CALLED!") - if type(key) != str: - raise ValueError(f"Parameter key[]='{type(key)}' is not 'str'") - elif key == "": - raise ValueError("Parameter 'key' cannot be empty") +def has_key(keys: list, search: str, value: any) -> bool: + # DEBUG: print(f"DEBUG: keys()={len(keys)},search='{search}',value[]='{type(value)}' - CALLED!") + if type(keys) != list: + raise ValueError(f"Parameter keys[]='{type(keys)}' is not 'list'") + elif type(search) != str: + raise ValueError(f"Parameter search[]='{type(search)}' is not 'str'") + elif search == "": + raise ValueError("Parameter 'search' is empty") has = False - # DEBUG: print(f"DEBUG: Checking elements()={len(elements)} ...") - for element in elements: - # DEBUG: print(f"DEBUG: element[]='{type(element)}'") - if type(element) != dict: - raise ValueError(f"element[]='{type(element)}' is not 'dict'") - elif not key in element: - raise KeyError(f"Cannot find key='{key}'") - elif element[key] == value: + # DEBUG: print(f"DEBUG: Checking keys()={len(keys)} ...") + for key in keys: + # DEBUG: print(f"DEBUG: key['{type(key)}']={key}") + if type(key) != dict: + raise ValueError(f"key[]='{type(key)}' is not 'dict'") + elif not search in key: + raise KeyError(f"Cannot find search='{search}'") + elif key[search] == value: has = True break @@ -1567,7 +1068,7 @@ def find_domains(tag: bs4.element.Tag) -> list: # DEBUG: print(f"DEBUG: domain='{domain}',reason='{reason}'") - if is_blacklisted(domain): + if blacklist.is_blacklisted(domain): print(f"WARNING: domain='{domain}' is blacklisted - skipped!") continue elif domain == "gab.com/.ai, develop.gab.com": @@ -1598,21 +1099,22 @@ def find_domains(tag: bs4.element.Tag) -> list: # DEBUG: print(f"DEBUG: domains()={len(domains)} - EXIT!") return domains -def get_url(url: str) -> requests.models.Response: - # DEBUG: print(f"DEBUG: url='{url}' - CALLED!") +def get_url(url: str, headers: dict, timeout: list) -> requests.models.Response: + # DEBUG: print(f"DEBUG: url='{url}',headers()={len(headers)},timeout={timeout} - CALLED!") if type(url) != str: raise ValueError(f"Parameter url[]='{type(url)}' is not 'str'") elif url == "": - raise ValueError("Parameter 'url' cannot be empty") + raise ValueError("Parameter 'url' is empty") # DEBUG: print(f"DEBUG: Parsing url='{url}'") - components = urllib.parse(url) + components = urlparse(url) # Invoke other function, avoid trailing ? + # DEBUG: print(f"DEBUG: components[{type(components)}]={components}") if components.query != "": - response = get_response(components.hostname, f"{components.path}?{components.query}") + response = get_response(components.hostname, f"{components.path}?{components.query}", headers, timeout) else: - response = get_response(components.hostname, f"{components.path}") + response = get_response(components.hostname, f"{components.path}", headers, timeout) # DEBUG: print(f"DEBUG: response[]='{type(response)}' - EXXIT!") return response