tweet-printer/printer.py

#!/usr/bin/env python3

"""
TODO: Pseude Retweets (aka Commments). Tweets in which the only URL is to another Tweet.
TODO: Figure out if/when to show unshortened URLs in links contained in Tweets.
TODO: Figure out bimap mode so stuff that is not latin-1 encodable can still be printed.
"""

__version__ = "0.0.1"

import argparse
import datetime
import html
import logging
import sys

import toml
import tweepy

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("tweet-printer")

def load_config():
    fn = "config.toml"
    cfg = {}
    try:
        with open(fn, "r") as f:
            cfg = toml.load(f)
    except FileNotFoundError:
            logger.error(f"Can't open config file {fn}")
            sys.exit(1)
    except toml.TomlDecodeError as ex:
            logger.error(f"Can't parse config file {fn}: {ex}")
            sys.exit(1)

    if "auth" not in cfg:
        logger.error("Can't find auth section in configuration data.")
        sys.exit(1)
    authdata = cfg["auth"]
    if "consumer" not in authdata:
        logger.error("Can't find consumer auth section in configuration data.")
        sys.exit(1)

    consumerdata = authdata["consumer"]
    if "key" not in consumerdata or not consumerdata["key"]:
        logger.error("Can't find consumer key in configuration data.")
        sys.exit(1)
    if "secret" not in consumerdata or not consumerdata["secret"]:
        logger.error("Can't find consumer secret in configuration data.")
        sys.exit(1)

    if "access" not in authdata:
        logger.error("Can't find access auth section of configuration data.")
        sys.exit(1)
    accessdata = authdata["access"]
    if "token" not in accessdata:
        logger.error("Can't find access token in configuration data.")
        sys.exit(1)
    if "secret" not in accessdata:
        logger.error("Can't find access secret in configuration data.")
        sys.exit(1)

    return cfg


def format_header(handle, name, dt, is_rt=None):
    space_total = 28
    header = ""
    if is_rt is None:
        is_rt = False
    if is_rt:
        header += "RT by "
    header += f"@{handle} "[:28]
    space_for_name = space_total-len(header)-2

    shortened_by=0

    if len(name)<=space_for_name:
        header+=f"({name})"
    else:
        shortened_name_len = space_for_name-3
        if shortened_name_len>0:
            shortened_by = len(name)-shortened_name_len
            header+=f"({name[:shortened_name_len]}...)"

    header = header.ljust(28)
    header += dt.strftime(" %Y-%m-%d %H:%M:%S")
    #if shortened_by>0:
    #    header += f" name shoretened by {shortened_by}"
    return header


class Printer:
    def __init__(self, port):
        self.port = port

    def __enter__(self):
        self.printer = open(self.port, "wb")
        self.printer.write(bytes([0x1b, 0x40]))         # Initialize printer
        self.printer.write(bytes([0x1b, 0x74, 0x06]))   # Set character code table to CP1252/Latin-1
        return self

    def __exit__(self, exc_type, exc_value, traceback):
        self.printer.close()

    def write(self, data, newline=True, cut=False):
        if type(data) is str:
            data = data.encode("latin-1")
        elif type(data) is not bytes:
            logger.error("Unsupported type for Printer.write: "+str(type(data)))
            return
        self.printer.write(data)
        if newline:
            self.printer.write(bytes([13, 10]))
        if cut:
            self.printer.write(bytes([0x1d, 0x56,  # Feed and Cut
                                      0x42,  # Feed to cut position with extra length and cut
                                      10  # extra*Y Unit exta space
                                      ]))

    def flush(self):
        self.printer.flush()


class StreamListener(tweepy.StreamListener):
    def __init__(self, title, ignore_rt=False, printer=None):
        super().__init__()
        self.title = title
        self.seen_tweets = set()
        self.ignore_rt = ignore_rt
        self.printer = printer

    def on_status(self, status):

        if status.id in self.seen_tweets:
            print("********", status.id, "Already seen this tweet before")
            return True

        #with open("status.log", "at+") as f:
        #    f.write(repr(status))
        #    f.write("\n\n")

        rt = getattr(status, "retweeted_status", None)

        if not rt:
            print("********", status.id)
            self.seen_tweets.add(status.id)
            text = status.extended_tweet["full_text"] if status.truncated else status.text
            text = html.unescape(text)
            #print("[{}] New tweet from @{}: {}".format(self.title, status.user.screen_name, text))
            header = format_header(status.user.screen_name, status.user.name, status.created_at)
            print(header)
            print(text)
            if self.printer:
                try:
                    encoded_header = header.encode("latin-1")
                    encoded_text = text.encode("latin-1")
                    self.printer.write(encoded_header)
                    self.printer.write(encoded_text, True, True)
                    self.printer.flush()
                except UnicodeEncodeError:
                    print("Can't print. Encoding issue.")
        else:
            if self.ignore_rt:
                #print("******** Ignoring Retweet")
                return True
            if rt.id in self.seen_tweets:
                print("********", rt.id, "Already seen this retweeted before")
                return True
            print("********", rt.id)
            self.seen_tweets.add(rt.id)
            rt_text = rt.extended_tweet["full_text"] if rt.truncated else rt.text
            rt_text = html.unescape(rt_text)
            #print("[{}] New retweet by @{}:".format(self.title, status.user.screen_name))
            #print("@{}: {}".format(rt.user.screen_name, rt_text))
            header = format_header(status.user.screen_name, status.user.name, status.created_at, is_rt=True)
            rt_header = format_header(rt.user.screen_name, rt.user.name, rt.created_at)
            print(header)
            print(rt_header)
            print(rt_text)
        return True

    def on_error(self, status_code):
        if status_code == 420:
            return False


def check_rate_limits(api):
    rate_limits = api.rate_limit_status()
    for category, endpoints in rate_limits["resources"].items():
        for endpoint, data in endpoints.items():
            limit = int(data["limit"])
            remaining = int(data["remaining"])
            reset = data["reset"]
            if limit!=remaining and endpoint not in ["/application/rate_limit_status"]:
                print(endpoint, f"{remaining}/{limit}",
                      "Reset in:", datetime.datetime.fromtimestamp(reset)-datetime.datetime.now())


def main():
    parser = argparse.ArgumentParser(description="Prints tweets to a thermal printer")
    parser.add_argument("-p", "--printer", type=str, metavar="printer",
                        help="Which thermal printer to use (e.g. /dev/usb/lp1). If empty nothing will be printed.")
    rt_handling = parser.add_mutually_exclusive_group()
    rt_handling.add_argument("-n", "--no-retweets", action="store_true", default=None,
                        help="Don't show retweets (default when monitoring hashtags)")
    rt_handling.add_argument("-r", "--retweets", action="store_true", default=None,
                        help="Show retweets (default when showing user timeline)")
    parser.add_argument(metavar="keyword", nargs="*", type=str, dest="keywords",
                        help="What to track. You can specify hashtags or keywords here. If empty the user timeline will be monitored instead")

    args = parser.parse_args()

    config = load_config()

    consumer_key = config["auth"]["consumer"]["key"]
    consumer_secret = config["auth"]["consumer"]["secret"]

    access_token = config["auth"]["access"]["token"]
    access_secret = config["auth"]["access"]["secret"]

    auth = tweepy.OAuthHandler(consumer_key, consumer_secret)
    auth.set_access_token(access_token, access_secret)

    api = tweepy.API(auth)

    check_rate_limits(api)

    show_rts = None
    if len(args.keywords) == 0:
        show_rts = False if args.no_retweets else True
    else:
        show_rts = True if args.retweets else False

    stream_name = ",".join(args.keywords) if args.keywords else "Timeline"

    def start_streaming(listener, api):
        stream = tweepy.Stream(listener=listener, auth=api.auth)
        if args.keywords:
            stream.filter(track=args.keywords)
        else:
            stream.userstream()

    try:
        if args.printer:
            with Printer(args.printer) as thermal:
                logger.info("Streaming {} to thermal printer {}...".format(stream_name, thermal.port))
                listener = StreamListener(stream_name, ignore_rt=not show_rts, printer=thermal)
                start_streaming(listener, api)
        else:
            logger.info("Streaming {}...".format(stream_name))
            listener = StreamListener(stream_name, ignore_rt=not show_rts)
            start_streaming(listener, api)

    except KeyboardInterrupt:
            pass


if __name__ == "__main__":
    main()