utils.py

import random
from unicodedata import normalize


PERM_OWNER = 100
PERM_ADMIN = 50
PERM_USER = 10

PERMISSIONS = [
    PERM_OWNER,
    PERM_ADMIN,
    PERM_USER,
]


AFFIRMATIONS = [
    "Cool",
    "Nice",
    "Doing great",
    "Awesome",
    "Okey dokey",
    "Neat",
    "Whoo",
    "Wonderful",
    "Splendid",
    "Splendiferous",
    "Superb",
    "Brilliant",
    "Cool beans",
    "Sweet",
    "Marvelous",
    "Amazing",
    "Jolly good",
    "Great",
    "Swell",
    "Gorgeous",
    "Fantastic",
    "Alrighty",
    "Excellent",
    "Very good",
]

EXCLAMATIONS = [
    "Oh no!",
    "Oh dear!",
    "Bummer!",
    "Yikes!",
    "Aww...",
    "Damn...",
    "Aw no!",
    ">.<",
    "Whoopsie!",
    "Oh oh...",
    "!!!",
    "Oh fudge!",
    "Ye gods!",
    "Fiddlesticks!",
    "Zounds!"
    "Egads!",
    "Blast!",
    "Blazes!",
    "Holey moley",
]

GENERIC_RESPONSES = [
    "I see.",
    "Oh yeah.",
    "Okay",
    "Oh, I see",
    "So it has come to this...",
    "lol",
    "Whoa",
    "Really?",
    "Tell me more about that",
    "and then what happened?",
    "Oh, can you repeat that?",
    "No way!",
    "I knew it!",
    "huh, interesting",
    "Who knew?",
    "Indeed",
    "Quite so",
    "So the rumors are true",
    "but why?",
    "As foretold!",
    "As it was written",
    "That makes sense",
    "Very well",
    "Noted",
    "I shall consider that",
    "yeah, sure, whatever",
    "hmm?",
    "That checks out",
    "I had no idea!",
    "Indubitably",
    "???",
    "I thought so too",
    "My word!",
    "Oh fudge",
    "Pardon?",
    "Certainly",
    "I'm afraid I don't have the answer to that",
    "Who told you?",
    "How unexpected",
    "This is getting out of hand",
    "And then, the wolves came",
    "I think you have the wrong number",
    "nah",
    "I don't know, probably",
    "Sorry, can't talk right now, my hands are wet",
    "Do you take me for some conjurer or cheap tricks?",
    "I'll need to investigate that",
    "Scientists are still debating about this, actually",
    "That's just propaganda",
    "Affirmative",
    "I recommend you take a nap instead",
    "Fascinating",
    "How peculiar",
    "no hablo ingles",
]


def get_affirmation():
    return random.choice(AFFIRMATIONS)

def get_exclamation():
    return random.choice(EXCLAMATIONS)

def get_generic_response():
    return random.choice(GENERIC_RESPONSES)


def fuzzy_match(query, items, name_getter, min_words=2, min_chars_first_word=3, min_chars_subseq_words=2, min_chars_total=5):
    matches_found = []
    query_parts = to_words(query, ",")
    for part_index, part in enumerate(query_parts):
        query_words = part.strip().lower().split(' ')
        for item in items:
            name_words = to_words(name_getter(item))
            for q, query_word in enumerate(query_words):
                for n, name_word in enumerate(name_words):
                    match = len(commonprefix(query_word, name_word))
                    # Heuristic. Start scan from X matching characters onward.
                    if match >= min(min_chars_first_word, len(name_word)):
                        matches = []
                        for nn, next_name_word in enumerate(name_words[n:]):
                            if q+nn < len(query_words):
                                match = len(commonprefix(next_name_word, query_words[q+nn]))
                                # Heuristic. Consider it a match if at least X characters match
                                if match >= min(min_chars_subseq_words, len(next_name_word)):
                                    matches.append(match)
                        # Heuristic. If the product has multiple words, at least X must match.
                        if len(matches) >= min(min_words, len(name_words)):
                            # Heuristic. In total, at least X characters must match (if the product has that many).
                            if sum(matches) >= min(sum([len(w) for w in name_words]), min_chars_total):
                                if not any([match['item'] == item for match in matches_found]):
                                    matches_found.append({
                                        'len': len(matches),
                                        'sum': sum(matches),
                                        'part': part_index,
                                        'word': q,
                                        'item': item,
                                    })

    matches_found.sort(key=lambda m: (m['len'], m['sum']), reverse=True)
    return matches_found


def to_words(string, split_at=" "):
    return normalize('NFD', string.lower()).encode('ascii', 'ignore').decode('ascii').split(split_at)


def commonprefix(a, b):
    if a > b:
        return commonprefix(b, a)
    for i, c in enumerate(a):
        if c != b[i]:
            return a[:i]
    return a