utils_common.py - main - bioxakep/VoiceTrans

utils_common.py
184 строки7 KB
bioxakep
remove unused things
13 авг 2025, 23:33
13 авг 2025, 23:3362ae8c4
import os
import re
from collections import Counter, defaultdict
from dataclasses import dataclass
from datetime import datetime
from math import ceil
from string import punctuation

NO_PHONE_ABONENT = "NO_PHONE"


@dataclass
class SoundData:
    dt: datetime
    abonent: str
    contact: str
    duration: int


def get_phones(file_name: str):
    phone_pattern = re.compile(r"\d{10,14}")
    phones = phone_pattern.findall(file_name)
    if len(phones) < 2:
        return None, None
    return phones[:2]


def rename_sounds(billing_file_path: str, sounds_dir: str):
    if not os.path.exists(billing_file_path):
        raise FileExistsError("Файл с соединениями не найден.")
    if not os.path.exists(sounds_dir):
        raise FileExistsError("Директория со звуком не найдена.")
    billing_data = defaultdict(SoundData)
    with open(billing_file_path, "r") as billing_file:
        header = list(
            map(lambda x: x.strip('"'), billing_file.readline().strip("\n").split(";"))
        )
        for line in billing_file:
            line_data = dict(
                zip(
                    header,
                    list(map(lambda x: x.strip('"'), line.strip("\n").split(";"))),
                )
            )
            bil_sound_id = line_data["Ид.записи"].lower()
            dt = datetime.strptime(
                line_data["Время начала соединения"], "%d.%m.%Y %H:%M:%S"
            )
            abonent = line_data["Номер абонента"] or NO_PHONE_ABONENT
            contact = line_data["Номер контакта"] or NO_PHONE_ABONENT
            duration = line_data["Длительность, сек"]
            if bil_sound_id in billing_data.keys():
                sound_data = billing_data.get(bil_sound_id)
                updated_data = False
                if (
                    sound_data.abonent == NO_PHONE_ABONENT
                    and abonent != NO_PHONE_ABONENT
                ):
                    sound_data.abonent = abonent
                    updated_data = True
                if (
                    sound_data.contact == NO_PHONE_ABONENT
                    and contact != NO_PHONE_ABONENT
                ):
                    sound_data.contact = contact
                    updated_data = True
                print(f"Exists rec {bil_sound_id}, updated: {updated_data}")
                continue
            print(f"New rec {bil_sound_id} = {dt}")
            billing_data[bil_sound_id] = SoundData(
                dt=dt, abonent=abonent, contact=contact, duration=duration
            )
    for f in os.listdir(sounds_dir):
        wav_sound_id = f.split(".")[0].lower()
        sound_ext = f.split(".")[-1]
        sound_path: str = os.path.join(sounds_dir, f)
        if wav_sound_id not in billing_data.keys():
            print(f"Файл со звуком {wav_sound_id} в билинге не найден.")
            continue
        rec_data = billing_data.get(wav_sound_id)
        dt_str = rec_data.dt.strftime("%Y_%m_%d_%H_%M_%S")
        new_file_name: str = (
            f"{dt_str}_{rec_data.abonent}_{rec_data.contact}_{rec_data.duration}.{sound_ext}"
        )
        new_sound_path: str = os.path.join(sounds_dir, new_file_name)
        if os.path.exists(new_sound_path):
            os.remove(new_sound_path)
        os.rename(sound_path, new_sound_path)
        print(f"Файл {f} переименован в {new_file_name}")
        yield new_sound_path, rec_data.abonent, rec_data.contact


def humanize_time(time_seconds: int | float, clocks=True) -> str:
    time_seconds = ceil(time_seconds)
    if time_seconds == 0:
        return "0 секунд" if not clocks else "00:00:00"
    human_time = ""
    human_time_clocks = ""
    # Дни
    days = time_seconds // 86400
    if days > 0:
        human_time = str(days)
        if days % 10 == 1 and days not in range(11, 20):
            human_time += " день, "
        elif days % 10 in [2, 3, 4]:
            human_time += f" дня, "
        else:
            human_time += f" дней, "
        if clocks:
            human_time_clocks = f"{days}d "
    # Часы
    hours = (time_seconds - days * 86400) // 3600
    if hours > 0:
        human_time += str(hours)
        if hours % 10 == 1 and hours not in range(11, 20):
            human_time += " час, "
        elif hours % 10 in [2, 3, 4] and hours not in range(11, 20):
            human_time += f" часа, "
        else:
            human_time += f" часов, "
        human_time_clocks += ("0" if hours < 10 else f"") + str(hours) + ":"
    else:
        human_time_clocks += "00:"
    # Минуты
    minutes = (time_seconds - days * 86400 - hours * 3600) // 60
    if minutes > 0:
        human_time += str(minutes)
        if minutes % 10 == 1 and minutes not in range(11, 20):
            human_time += " минута, "
        elif minutes % 10 in [2, 3, 4] and minutes not in range(11, 20):
            human_time += f" минуты, "
        else:
            human_time += f" минут, "
        human_time_clocks += ("0" if minutes < 10 else f"") + str(minutes) + ":"
    else:
        human_time_clocks += "00:"
    # Секунды
    seconds = time_seconds - days * 86400 - hours * 3600 - minutes * 60
    if seconds > 0:
        human_time += str(seconds)
        if seconds % 10 == 1 and seconds not in range(11, 20):
            human_time += " секунда"
        elif seconds % 10 in [2, 3, 4] and seconds not in range(11, 20):
            human_time += f" секунды"
        else:
            human_time += f" секунд"
        human_time_clocks += ("0" if seconds < 10 else f"") + str(seconds)
    else:
        human_time_clocks += "00"

    return human_time_clocks if clocks else human_time


def compare_texts(base_text, new_text):
    base_text = base_text.lower()
    new_text = new_text.lower()
    for p in punctuation:
        base_text = base_text.replace(p, "")
        new_text = new_text.replace(p, "")
    base_text_counter = Counter(base_text.split())
    new_text_counter = Counter(new_text.split())
    both_keys = list(set(base_text_counter.keys()) & set(new_text_counter.keys()))
    keys_argument = len(both_keys) / len(
        base_text_counter.keys() | new_text_counter.keys()
    )
    words_argument = len(both_keys)
    for key in both_keys:
        if new_text_counter[key] != base_text_counter[key]:
            words_argument -= 1
    if len(both_keys) > 0:
        words_argument = words_argument / len(both_keys)
    else:
        words_argument = 0
    print(
        f"Сравнение: слова: {keys_argument * 100: 0.2f}",
        f"частота: {words_argument * 100: 0.2f}",
        f"общее: {keys_argument * words_argument * 100: 0.2f}",
    )
    return keys_argument * 100


if __name__ == "__main__":
    print(humanize_time(13123123))
    print(humanize_time(13123123, clocks=False))