LyricGenius Script - Download and Convert to Text Files

'''
LyricGenius Script: Download and Convert to Text Files
-------------------------------------------------------
This script downloads song lyrics using the 'LyricsGenius' python library and converts them into text files
by removing unnecessary special characters. The purpose of the script is to prepare the lyrics as better data
to be used for AI training.

Script by: Creatide (Sakari Niittymaa - 2023)
'''

import os
import json
from lyricsgenius import Genius

class Lyrics():
    def __init__(self):

        self.artist_name = 'Nirvana'

        # Remove or replace these symbols and words
        self.remove_symbols = {
            'д': 'ä', 'Д': 'Ä',
            'ц': 'ö', 'Ц': 'Ö',
            '/': '', '_': '', '--': '',
            'I:': '', 'II:': '', 'III:': '',
            'San.': '', 'san.': '', 'Säv.': '', 'säv.': '',
            'Kertosäe2:': '', 'Kertosäe:': '' ,
            'Embed': '', 'You might also like': '',
            'CHORUS': '', 'Chorus': '',
            'VERSE': '', 'Verse': '',
            'Intro': '', 'INTRO': '',
            'SOLOS:': '', 'SOLO:': '', '(solo)': '',
            'PRE-:': '', 'PRE-': '',
            'C-PART': '', 'Spoken:': '',
            '(x2)': '', '( x2)': '', '(2x)': '', '(x3)': '', '(3x)': '', '( x3)': '',
            '()': '', '(?)': '', '(???)': '',
            '(original)': '',
            "Refren':": '', "Refrain:": '', "Refren'": '',
        }

        self.json_filename = ''

        # DISABLE DOWNLOADING AND ONLY CLEAN TEXT FILES
        # ---------------------------------------------
        self.download_lyrics()

        # Convert and clean text to files
        self.convert_to_txt()


    # Download Lyrics ---------------------------------------------------------------------------- #
    # https://lyricsgenius.readthedocs.io/en/master/usage.html

    def download_lyrics(self):
        genius = Genius('INSERT_GENIUS_TOKEN_HERE')
        genius.remove_section_headers = True
        genius.response_format = 'plain'
        artist = genius.search_artist(self.artist_name)
        if artist:
            artist.save_lyrics()

    # Extract Lyrics From JSON And Save To Separated Files --------------------------------------- #

    def convert_to_txt(self):

        self.json_filename = self.clean_name('Lyrics_' + self.clean_artist_name(self.artist_name) + '.json')
        self.json_filename = self.json_filename.replace(' ', '')

        # Check if json file exist
        if os.path.exists(self.json_filename):

            json_data = {}

            def save_text_file(data):

                filename = self.clean_name(data['song']) + ' - ' + self.clean_name(data['artist']) + '.txt'
                file_path = os.path.join(os.getcwd(), 'Lyrics', data['artist'])

                if not os.path.exists(file_path):
                    os.makedirs(file_path)

                file_path = os.path.join(file_path, filename)

                # if os.path.exists(file_path):
                with open(file_path, 'w', encoding='utf-8') as f:
                    reformatted_data = ''
                    for lyric in data['lyrics']:
                        reformatted_data += lyric
                    # Try to split string from target word like "Lyrics"
                    if len(reformatted_data.split('Lyrics', 1)) > 1:
                        reformatted_data = reformatted_data.split('Lyrics', 1)[1]
                    # If needed replace character with another
                    for k, v in self.remove_symbols.items():
                        reformatted_data = reformatted_data.replace(k, v)

                    # Save file
                    f.write(str(reformatted_data))

            with open(self.json_filename) as json_file:
                json_data = json.load(json_file)
                print('File Found:', self.json_filename)

            for song in json_data['songs']:
                data = {
                    'artist': song['artist'],
                    'song': song['title'],
                    'lyrics': song['lyrics'],
                }
                print(song['title'])
                save_text_file(data)

    # Clean names to avoid path errors
    def clean_name(self, name):
        return_name = name
        remove_symbols = [*'<>:"/\|?*']
        for i in remove_symbols:
            return_name = return_name.replace(i, '')
        return return_name

    # Clean artist names to make path work
    def clean_artist_name(self, name):
        return_name = name
        remove_symbols = [*'<>:"/\|?*() ']
        for i in remove_symbols:
            return_name = return_name.replace(i, '')
        return return_name

# Main ------------------------------------------------------------------------------------------- #

def main():
    Lyrics()

if __name__ == '__main__':
    main()