epicyon/filters.py

__filename__ = "filters.py"
__author__ = "Bob Mottram"
__license__ = "AGPL3+"
__version__ = "1.2.0"
__maintainer__ = "Bob Mottram"
__email__ = "bob@libreserver.org"
__status__ = "Production"
__module_group__ = "Moderation"

import os
from utils import acct_dir


def add_filter(base_dir: str, nickname: str, domain: str, words: str) -> bool:
    """Adds a filter for particular words within the content of a incoming posts
    """
    filtersFilename = acct_dir(base_dir, nickname, domain) + '/filters.txt'
    if os.path.isfile(filtersFilename):
        if words in open(filtersFilename).read():
            return False
    try:
        with open(filtersFilename, 'a+') as filtersFile:
            filtersFile.write(words + '\n')
    except OSError:
        print('EX: unable to append filters ' + filtersFilename)
    return True


def add_global_filter(base_dir: str, words: str) -> bool:
    """Adds a global filter for particular words within
    the content of a incoming posts
    """
    if not words:
        return False
    if len(words) < 2:
        return False
    filtersFilename = base_dir + '/accounts/filters.txt'
    if os.path.isfile(filtersFilename):
        if words in open(filtersFilename).read():
            return False
    try:
        with open(filtersFilename, 'a+') as filtersFile:
            filtersFile.write(words + '\n')
    except OSError:
        print('EX: unable to append filters ' + filtersFilename)
    return True


def remove_filter(base_dir: str, nickname: str, domain: str,
                  words: str) -> bool:
    """Removes a word filter
    """
    filtersFilename = acct_dir(base_dir, nickname, domain) + '/filters.txt'
    if not os.path.isfile(filtersFilename):
        return False
    if words not in open(filtersFilename).read():
        return False
    newFiltersFilename = filtersFilename + '.new'
    try:
        with open(filtersFilename, 'r') as fp:
            with open(newFiltersFilename, 'w+') as fpnew:
                for line in fp:
                    line = line.replace('\n', '')
                    if line != words:
                        fpnew.write(line + '\n')
    except OSError as ex:
        print('EX: unable to remove filter ' + filtersFilename + ' ' + str(ex))
    if os.path.isfile(newFiltersFilename):
        os.rename(newFiltersFilename, filtersFilename)
        return True
    return False


def remove_global_filter(base_dir: str, words: str) -> bool:
    """Removes a global word filter
    """
    filtersFilename = base_dir + '/accounts/filters.txt'
    if not os.path.isfile(filtersFilename):
        return False
    if words not in open(filtersFilename).read():
        return False
    newFiltersFilename = filtersFilename + '.new'
    try:
        with open(filtersFilename, 'r') as fp:
            with open(newFiltersFilename, 'w+') as fpnew:
                for line in fp:
                    line = line.replace('\n', '')
                    if line != words:
                        fpnew.write(line + '\n')
    except OSError as ex:
        print('EX: unable to remove global filter ' +
              filtersFilename + ' ' + str(ex))
    if os.path.isfile(newFiltersFilename):
        os.rename(newFiltersFilename, filtersFilename)
        return True
    return False


def _is_twitter_post(content: str) -> bool:
    """Returns true if the given post content is a retweet or twitter crosspost
    """
    if '/twitter.' in content or '@twitter.' in content:
        return True
    elif '>RT <' in content:
        return True
    return False


def _is_filtered_base(filename: str, content: str) -> bool:
    """Uses the given file containing filtered words to check
    the given content
    """
    if not os.path.isfile(filename):
        return False

    try:
        with open(filename, 'r') as fp:
            for line in fp:
                filterStr = line.replace('\n', '').replace('\r', '')
                if not filterStr:
                    continue
                if len(filterStr) < 2:
                    continue
                if '+' not in filterStr:
                    if filterStr in content:
                        return True
                else:
                    filterWords = filterStr.replace('"', '').split('+')
                    for word in filterWords:
                        if word not in content:
                            return False
                    return True
    except OSError as ex:
        print('EX: _is_filtered_base ' + filename + ' ' + str(ex))
    return False


def is_filtered_globally(base_dir: str, content: str) -> bool:
    """Is the given content globally filtered?
    """
    globalFiltersFilename = base_dir + '/accounts/filters.txt'
    if _is_filtered_base(globalFiltersFilename, content):
        return True
    return False


def is_filtered_bio(base_dir: str,
                    nickname: str, domain: str, bio: str) -> bool:
    """Should the given actor bio be filtered out?
    """
    if is_filtered_globally(base_dir, bio):
        return True

    if not nickname or not domain:
        return False

    accountFiltersFilename = \
        acct_dir(base_dir, nickname, domain) + '/filters_bio.txt'
    return _is_filtered_base(accountFiltersFilename, bio)


def is_filtered(base_dir: str, nickname: str, domain: str,
                content: str) -> bool:
    """Should the given content be filtered out?
    This is a simple type of filter which just matches words, not a regex
    You can add individual words or use word1+word2 to indicate that two
    words must be present although not necessarily adjacent
    """
    if is_filtered_globally(base_dir, content):
        return True

    if not nickname or not domain:
        return False

    # optionally remove retweets
    removeTwitter = acct_dir(base_dir, nickname, domain) + '/.removeTwitter'
    if os.path.isfile(removeTwitter):
        if _is_twitter_post(content):
            return True

    accountFiltersFilename = \
        acct_dir(base_dir, nickname, domain) + '/filters.txt'
    return _is_filtered_base(accountFiltersFilename, content)