diff --git a/README.md b/README.md index d69fc98..cf935f9 100644 --- a/README.md +++ b/README.md @@ -14,6 +14,8 @@ Monitoring tool for Status App communities. - `STATUS_DISPLAY_NAME` - The Status display name that will be used to create an account. - `STATUS_PASSWORD` - The Status password that will be used to create an account. - `STATUS_MNEMONIC` - The mnemonic used to recover the account. If passed a `.bkp` file will be loaded as well. Use this when you want to login to a bot account via Status App, join a community / leave community and export the `.bkp` file. +- `STATUS_INFURA_TOKEN` - [Infura token](https://www.infura.io/) is required for **token gated communities** +- `STATUS_COINGECKO_API_KEY` - [Coingecko API Key](https://www.coingecko.com/) is required for **token gated communities** ## Docker deployement @@ -26,6 +28,10 @@ STATUS_DISPLAY_NAME = "bot-status" STATUS_PASSWORD = "ChangeThisPassword" STATUS_MNEMONIC= "test test test test test test test test test test test test" +# Necessary for communities that have tokens +STATUS_INFURA_TOKEN = "Token from https://www.infura.io/" +STATUS_COINGECKO_API_KEY = "Token from https://www.coingecko.com/" + # Database config POSTGRES_HOST=database POSTGRES_PORT=5432 diff --git a/bot/account.py b/bot/account.py index bd967d8..ebc3b1b 100644 --- a/bot/account.py +++ b/bot/account.py @@ -402,7 +402,7 @@ class Account: {"type": "channel", "id": chat["chat_id"], "name": f"{community['name']} #{chat['name']}"} for community in self.communities for chat in community["channels"] - if chat["permissions"]["posting"] + if chat["permissions"]["posting"] and community["is_member"] ] contacts = [ {"type": "contact", "id": contact["chat_id"], "name": contact["display_name"]} diff --git a/config.yaml b/config.yaml index 8b6dbaf..ca3eed5 100644 --- a/config.yaml +++ b/config.yaml @@ -5,7 +5,10 @@ postgres: community: "raw_community_info" members: "raw_community_activity" # Value must be in MINUTES -sleep: 1440 # 1 day +sleep: 10 +files: + # Get the latest community chat dates for next run + current_state: "dates.pkl" # Parameters based on Account class bot_params: # localhost - to run code locally with Status Backend Dockerfile diff --git a/monitor.py b/monitor.py index dcd53f5..19fb951 100644 --- a/monitor.py +++ b/monitor.py @@ -1,11 +1,24 @@ import datetime, os, pickle, yaml, time import pandas as pd +from typing import Any from pathlib import Path from dotenv import load_dotenv # Manual file imports from bot import Account, Logger from postgres import Postgres +def to_midnight(timestamp: datetime.datetime) -> datetime.datetime: + """ + Convert the given timestamp to midnight + + Parameters: + - `timestamp` - current timestap + + Output: + - `timestamp` at midnight + """ + return timestamp.replace(minute=0, second=0, hour=0, microsecond=0) + def load_config(file_path: str) -> dict: """ Load the config file and the `.env` variables @@ -61,7 +74,7 @@ def extract_community_members(account: Account, community_id: str) -> pd.DataFra return pd.DataFrame() -def extract_community_channels(account: Account, community: dict, start_timestamp: datetime.datetime, end_timestamp: datetime.datetime) -> pd.DataFrame: +def extract_community_channels(account: Account, community: dict, latest_dates: dict[str, pd.Timestamp]) -> pd.DataFrame: """ Extract the community channel messages. @@ -76,22 +89,51 @@ def extract_community_channels(account: Account, community: dict, start_timestam """ final = [] for channel in community["channels"]: - messages = account.get_messages(channel["chat_id"], start_timestamp, end_timestamp) + now = datetime.datetime.now() + start_timestamp = latest_dates.get(channel["chat_id"]) + if start_timestamp: + start_timestamp += datetime.timedelta(seconds=1) + else: + # Node will only return known / fetched messages for this channel. + # Without enabling community archives feature the node can only fetch last 30 days (from store nodes). + start_timestamp = to_midnight(now - datetime.timedelta(days=30)) + + account.logger.info(f"Starting message extraction for # {channel['name']} [{start_timestamp} - {now}]") + messages = account.get_messages(channel["chat_id"], start_timestamp, now) messages = pd.DataFrame(messages) if len(messages) == 0: - account.logger.info(f"No messages found for # {channel['name']}") + account.logger.info(f"No messages found") continue - account.logger.info(f"Extracted {len(messages)} message(s) from # {channel['name']}") + account.logger.info(f"Extracted {len(messages)} message(s)") messages = messages.assign( community_id = community["id"], - extracted_timestamp = datetime.datetime.now() + extracted_timestamp = now ) final.append(messages) return pd.concat(final, ignore_index=True) if final else pd.DataFrame() +def save_file(file_path: str, data: Any): + """ + Save data to a pickle file. Creates directories if they don't exist. + + Parameters: + - `file_path` - Full pikle path + - `data` - Python object to be saved + """ + folder = os.path.dirname(file_path) + if len(folder) > 0: + os.makedirs(folder, exist_ok=True) + + if isinstance(data, pd.DataFrame): + data.to_csv(file_path, index=False) + return + + with open(file_path, "wb") as f: + pickle.dump(data, f) + def download(folder: str, config: dict): """ Download Status App messages / info from communities and store them in pickle files. @@ -103,57 +145,56 @@ def download(folder: str, config: dict): account = Account(**config.get("bot_params", {})) available_accounts = [acc["display_name"] for acc in account.available_accounts] - messages_folder = os.path.join(folder, "messages") - os.makedirs(messages_folder, exist_ok=True) - - community_info_folder = os.path.join(folder, "community") - os.makedirs(community_info_folder, exist_ok=True) - - members_info_folder = os.path.join(folder, "members") - os.makedirs(members_info_folder, exist_ok=True) - prefix = "STATUS_" params = { key.replace(prefix, "").lower(): value - for key, value in os.environ.items() + for key, value in config["env_vars"].items() if key.startswith(prefix) } if params["display_name"] in available_accounts: params.pop("mnemonic") account.login(**params) + account.logger.info(f"Account Information:\nCompressed Key: {account.info['compressed_key']}\nPublic Key: {account.info['public_key']}\nURL: {account.info['url']}") + file_path = os.path.join(os.path.dirname(__file__), config["files"]["current_state"]) + latest_dates: dict[str, pd.Timestamp] = pd.read_pickle(file_path) if os.path.exists(file_path) else {} - now = datetime.datetime.now() - # Node will only return known / fetched messages for this channel. - # Without enabling community archives feature the node can only fetch last 30 days (from store nodes). - to_midnight = lambda date: date.replace(minute=0, second=0, hour=0, microsecond=0) - start_timestamp: datetime.datetime = to_midnight(now - datetime.timedelta(days=30)) - end_timestamp: datetime.datetime = to_midnight(now - datetime.timedelta(days=1)) + get_file_name = lambda: str(to_midnight(datetime.datetime.now()).timestamp()).replace(".", "") + communities = account.communities + if not communities: + account.logger.warning("No communities found...") - get_file_name = lambda: str(to_midnight(datetime.datetime.now()).timestamp()).replace(".", "") + ".pkl" + for community in communities: - for community in account.communities: - account.logger.info(f"Extracting data for {community['name']} from {start_timestamp} to {end_timestamp}") + if not community["is_member"]: + continue + + community_folder_name = community["name"].replace(" ", "-") + messages_folder = os.path.join(folder, "messages", community_folder_name) + community_info_folder = os.path.join(folder, "community", community_folder_name) + members_info_folder = os.path.join(folder, "members", community_folder_name) + + account.logger.info(f"Extracting data for {community['name']}") community["extracted_timestamp"] = datetime.datetime.now() - file_path = os.path.join(community_info_folder, get_file_name()) + + file_path = os.path.join(community_info_folder, get_file_name() + ".pkl") if not os.path.exists(file_path): - with open(file_path, "wb") as f: - pickle.dump(community, f) + save_file(file_path, community) account.logger.info(f"Created {file_path}") - file_path = os.path.join(members_info_folder, get_file_name()) + file_path = os.path.join(members_info_folder, get_file_name() + ".csv") if not os.path.exists(file_path): members = extract_community_members(account, community["id"]) if len(members) > 0: - members.to_pickle(file_path) + save_file(file_path, members) account.logger.info(f"Created {file_path}") - file_path = os.path.join(messages_folder, get_file_name()) + file_path = os.path.join(messages_folder, get_file_name() + ".csv") if not os.path.exists(file_path): - messages = extract_community_channels(account, community, start_timestamp, end_timestamp) + messages = extract_community_channels(account, community, latest_dates) if len(messages) > 0: - messages.to_pickle(file_path) + save_file(file_path, messages) account.logger.info(f"Created {file_path}") def store(folder: str, config: dict): @@ -170,23 +211,38 @@ def store(folder: str, config: dict): table_schema = config["postgres"]["schema"] upload: dict[str, list] = {} + latest_dates: dict[str, pd.Timestamp] = {} completed = [] - for file_path in path.rglob("*.pkl"): - table_name = table_name_mapping.get(file_path.parent.name) + files = list(path.rglob("*.pkl")) + list(path.rglob("*.csv")) + for file_path in files: + + table_name = table_name_mapping.get(file_path.parent.parent.name) if not table_name: continue - data = pd.read_pickle(file_path) + file_name = str(file_path.name) + data = pd.read_pickle(file_path) if file_name.endswith(".pkl") else pd.read_csv(file_path) if isinstance(data, dict): data = pd.DataFrame([data]) + for column in data.columns: + if "timestamp" not in column: + continue + data[column] = pd.to_datetime(data[column]) + if table_name not in upload: upload[table_name] = [] + if "timestamp" in data.columns: + latest_dates.update(data.groupby("chat_id")["timestamp"].max().to_dict()) + upload[table_name].append(data) completed.append(str(file_path)) + if latest_dates: + save_file(config["files"]["current_state"], latest_dates) + prefix = "POSTGRES_" params = { key.replace(prefix, "").lower(): value