common.py

#!/usr/bin/env python3
# -*- coding: UTF-8 -*-
"""The common component contains the logger, error logging, and
flair sanitizing functions that are used by both routines.
There are no functions that connect to Reddit in this component.
"""
import datetime
import logging
import re
import time

from settings import INFO, FILE_ADDRESS

logger = None

"""INITIALIZATION INFORMATION"""


def main_error_log(entry):
    """A function to save detailed errors to a log for later review.
    This is easier to check for issues than to search through the entire
    events log, and is based off of a basic version of the function
    used in Wenyuan/Ziwen.

    :param entry: The text we wish to include in the error log entry.
                  Typically this is the traceback entry.
    :return: Nothing.
    """

    # Open the file for the error log in appending mode.
    # Then add the error entry formatted our way.
    with open(FILE_ADDRESS.error, "a+", encoding="utf-8") as f:
        error_date_format = datetime.datetime.utcnow().strftime("%Y-%m-%dT%I:%M:%SZ")
        bot_format = "Artemis v{}".format(INFO.version_number)
        entry = entry.replace("\n", "\n    ")  # Indent the code.
        f.write(
            "\n---------------\n### {} ({})\n{}\n".format(error_date_format, bot_format, entry)
        )

    return


"""LOGGER SETUP"""


def start_logger(file_path=FILE_ADDRESS.logs):
    """The main logging system used by the bot. Allows for separate
    file paths to be passed to it.
    :param file_path: Expressed in terms of FILE_ADDRESS.xxx,
                      where xxx is the file name.
    :return: Nothing.
    """

    global logger

    # Set up the logger. By default only display INFO or higher levels.
    log_format = "%(levelname)s: %(asctime)s - [Artemis] v{} %(message)s"
    logformatter = log_format.format(INFO.version_number)
    logging.basicConfig(format=logformatter, level=logging.INFO)

    # Set the logging time to UTC.
    logging.Formatter.converter = time.gmtime
    logger = logging.getLogger(__name__)

    # Define the logging handler (the file to write to.)
    # By default only log INFO level messages or higher.
    handler = logging.FileHandler(file_path, "a", "utf-8")
    handler.setLevel(logging.INFO)

    # Set the time format in the logging handler.
    d = "%Y-%m-%dT%H:%M:%SZ"
    handler.setFormatter(logging.Formatter(logformatter, datefmt=d))
    logger.addHandler(handler)

    return logger


"""OTHER FUNCTIONS"""


def flair_sanitizer(text_to_parse, change_case=True):
    """This is a small function that sanitizes the input from the user
    for flairs and from flair dictionaries' text in order to make them
    consistent. This includes removing extraneous characters,
    lower-casing and stripping, and removing Reddit and Unicode emoji.

    :param text_to_parse: The text we want to convert and clean up.
    :param change_case: Whether or not we want to change the
                        capitalization of the text. Generally we want to
                        change it if it's for a case-insensitive
                        situation like matching people's messages.
                        Otherwise, if we're just displaying the options
                        available, we *do not* want to change case.
    :return: The sanitized text.
    """
    # Here we REMOVE the brackets and characters that may be in post
    # flairs so that they can match.
    deleted_characters = ["[", "]", ">", "•"]
    for character in deleted_characters:
        if character in text_to_parse:
            text_to_parse = text_to_parse.replace(character, "")

    # Here we REPLACE some problematic characters that may cause
    # rendering issues, namely vertical pipes in tables.
    replaced_characters = {"|": "◦"}
    for character in replaced_characters:
        if character in text_to_parse:
            text_to_parse = text_to_parse.replace(character, replaced_characters[character])

    # Process the text further. If changing case is desired, change it.
    # In case people keep the Reddit emoji text in, delete it.
    text_to_parse = text_to_parse.strip()
    if change_case:
        text_to_parse = text_to_parse.lower()
    text_to_parse = re.sub(r":\S+:", "", text_to_parse)

    # Account for Unicode emoji by deleting them as well.
    # uFE0F is an invisible character marking emoji.
    reg = re.compile(
        u"[\U0001F300-\U0001F64F"
        u"\U0001F680-\U0001F6FF"
        u"\U0001F7E0-\U0001F7EF"
        u"\U0001F900-\U0001FA9F"
        u"\uFE0F\u2600-\u26FF\u2700-\u27BF]",
        re.UNICODE,
    )
    text_to_parse = reg.sub("", text_to_parse).strip()

    return text_to_parse


def markdown_escaper(input_text):
    """Small function that escapes out special characters in Markdown
    so they display as intended. Primarily intended for use in titles.
    :param input_text: The text we want to work with.
    :return: `input_text`, but with the characters escaped.
    """
    characters_to_replace = ["[", "]", "`", "*", "_"]

    for character in characters_to_replace:
        input_text = input_text.replace(character, r"\{}".format(character))

    return input_text


def flair_template_checker(input_text):
    """Small function that checks whether a given input is valid as a
    Reddit post flair ID.
    """
    try:
        regex_pattern = r"^[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12}$"
        valid = re.search(regex_pattern, input_text)
    except TypeError:
        return False

    if valid:
        return True
    else:
        return False


start_logger()