Finalize scripts

.gitattributes

@@ -1,2 +1,5 @@
 *.txt filter=lfs diff=lfs merge=lfs -text
 *.ERB filter=lfs diff=lfs merge=lfs -text
+*.csv filter=lfs diff=lfs merge=lfs -text
+*.json filter=lfs diff=lfs merge=lfs -text
+

convo-dataset/dataset.py

@@ -1,36 +0,0 @@
-import abc
-import os
-import json
-
-from tqdm import tqdm
-
-from . import util
-
-class Dataset(abc.ABC):
-    @abc.abstractmethod
-    def name(self):
-        """
-        Returns the name of the dataset
-        """
-        pass
-
-    @abc.abstractmethod
-    def download(self):
-        """
-        Downloads the dataset
-        """
-        pass
-
-    @abc.abstractmethod
-    def clean(self):
-        """
-        Cleans the dataset
-        """
-        pass
-
-    @abc.abstractmethod
-    def size(self):
-        """
-        Returns the size of the dataset
-        """
-        pass

convo-dataset/util.py

@@ -1,8 +0,0 @@
-import requests
-
-def download_file(url, filename):
-    with requests.get(url, stream=True) as r:
-        r.raise_for_status()
-        with open(filename, 'wb') as f:
-            for chunk in r.iter_content(chunk_size=8192): 
-                f.write(chunk)

utils/compile.py

@@ -0,0 +1,71 @@
+import random
+import glob
+import os
+import json
+
+import argparse
+
+from tqdm import tqdm
+
+import umi_parse
+import cornell_movie_dialogue_parse
+import eratoho_parse
+import valhalla_parse
+import discord_parse
+
+def compile_raw():
+    print('-- Parsing Umineko Data --\n')
+    umi_parse.parse()
+    print('\n-- Parsing Cornell Movie Dialogue Data --\n')
+    cornell_movie_dialogue_parse.parse()
+    print('\n-- Parsing Eratoho Data --\n')
+    eratoho_parse.parse()
+    print('\n-- Parsing Valhalla Data --\n')
+    valhalla_parse.parse()
+    print('\n-- Parsing Discord Data --\n')
+    discord_parse.parse()
+
+def compile_mtf_jax():
+    # compiles dataset into a single text file to be tokenized by the mtf jax repo
+    
+    # get all the files with *.txt in ./data.
+    files = glob.glob('data/*/*.txt')
+    with open('output.txt', 'w', encoding='utf-8') as f:
+        for file in tqdm(files):
+            with open(file, 'r', encoding='utf-8') as f2:
+                f.write(f2.read().replace('\n\n', '\n'))
+                f.write('\n')
+    
+    # remove all double newlines
+    lines = ''
+    with open('output.txt', 'r', encoding='utf-8') as f:
+        lines = f.read().replace('\n\n', '\n')
+    with open('output.txt', 'w', encoding='utf-8') as f:
+        f.write(lines)
+
+def compile_gpt_neo():
+    # compile each file into a json lines file
+    files = glob.glob('data/*/*.txt')
+    # shuffle the files
+    random.shuffle(files)
+    with open('output.jsonl', 'w', encoding='utf-8') as f:
+        for file in tqdm(files):
+            with open(file, 'r', encoding='utf-8') as f2:
+                f.write(json.dumps({'text': f2.read().replace('\n\n', '\n')}))
+                f.write('\n')
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description='Process raw data')
+    parser.add_argument('-d', '--dont_compile', action='store_true', help='dont compile raw data', default=False)
+    parser.add_argument('-m', '--mtf_jax', action='store_true', help='compile raw data into a single text file to be tokenized by the mtf jax repo')
+    parser.add_argument('-g', '--gpt_neo', action='store_true', help='compile raw data into a single json lines file')
+    args = parser.parse_args()
+
+    if not args.dont_compile:
+        compile_raw()
+    if args.mtf_jax:
+        compile_mtf_jax()
+    if args.gpt_neo:
+        compile_gpt_neo()
+
+    print('Done!')

utils/cornell_movie_dialogue_parse.py

@@ -1,4 +1,6 @@
 # Parses the Cornell Movie Dialogue Corpus https://www.cs.cornell.edu/~cristian/Cornell_Movie-Dialogs_Corpus.html
+import os
+import json
 import argparse
 import re
 import tqdm
@@ -54,7 +56,7 @@ def get_movie_metadata(filename):
 
 
 def format_movie_metadata(metadata):
-    return f'[Title: \'{metadata["title"]}\'; Year: {metadata["year"]}; Rating: {metadata["rating"]}; Num_votes: {metadata["num_votes"]}; Genres: {metadata["genres"]};]'
+    return f'[Title: {metadata["title"]}; Genres: {metadata["genres"].replace("[", "").replace("]", "")}]'.replace('\'', '')
 
 
 def construct_dialogue(lineIDs, lines):
@@ -65,10 +67,21 @@ def construct_dialogue(lineIDs, lines):
         dialogue.append(lines[lineID])
     return '\n'.join(dialogue)
 
+def dump_stats(length, movie_id):
+    stats = {'Cornell Movie-Dialogue Corpus': {}}
+    if os.path.exists('stats.json'):
+        stats = json.load(open('stats.json', 'r', encoding='utf-8'))
+        if 'Cornell Movie-Dialogue Corpus' not in stats:
+            stats['Cornell Movie-Dialogue Corpus'] = {}
+    stats['Cornell Movie-Dialogue Corpus'][str(movie_id)] = length
+    with open('stats.json', 'w', encoding='utf-8') as f:
+        json.dump(stats, f)
 
 # parses movie_conversations.txt and puts the results in out_filename
-def parse_conversations(conversation_filename, out_filename, lines, movie_metadata):
-    with open(conversation_filename, 'r', encoding='utf-8') as conversation_file, open(out_filename, 'w', encoding='utf-8') as out_file:
+def parse_conversations(conversation_filename, out_dir, lines, movie_metadata, stats=True):
+    if not os.path.exists(out_dir):
+        os.makedirs(out_dir)
+    with open(conversation_filename, 'r', encoding='utf-8') as conversation_file:
         for row in tqdm.tqdm(conversation_file, desc='Constructing dialogue'):
             row = row.split(' +++$+++ ')
             if len(row) != 4:
@@ -80,22 +93,33 @@ def parse_conversations(conversation_filename, out_filename, lines, movie_metada
                 dialogue = construct_dialogue(lineIDs, lines)
             except InvalidLineError:
                 continue
-            out_file.write(f'⁂\n{format_movie_metadata(movie_metadata[movieID])}\n⁂\n')
-            out_file.write(f'{dialogue}\n')
-
-
-parser = argparse.ArgumentParser(description='Process Cornell Movie Dialogue Corpus', formatter_class=argparse.ArgumentDefaultsHelpFormatter)
-parser.add_argument('-i', '--in_dir', type=str, help='directory to process', default='.')
-parser.add_argument('-o', '--out_file', type=str, help='file to output', default='cornell_movie_dialogue.txt')
-parser.add_argument('-r', '--raw_name', action='store_true', help='use speaker tags as they appear in the dataset instead of normalized names')
-args = parser.parse_args()
-
-if __name__ == '__main__':
+            with open(f'{out_dir}/{movieID}.txt', 'w', encoding='utf-8') as out_file:
+                out_file.write(f'⁂\n{format_movie_metadata(movie_metadata[movieID])}\n⁂\n')
+                out_file.write(f'{dialogue}\n')
+                if stats:
+                    dump_stats(len(dialogue.split('\n')), movieID)
+
+def parse(args = None):
+    if args == None:
+        parser = argparse.ArgumentParser()
+        parser.add_argument('--in_dir', default='./raw/cornell movie-dialogs corpus')
+        parser.add_argument('--out_dir', default='./data/cornell movie-dialogs corpus')
+        parser.add_argument('--raw_name', action='store_true', default=False)
+        parser.add_argument('--stats', action='store_true', default=True)
+        args = parser.parse_args()
     try:
         lines = get_lines(args.in_dir + '/movie_lines.txt', args.raw_name);
         movie_metadata = get_movie_metadata(args.in_dir + '/movie_titles_metadata.txt')
-        parse_conversations(args.in_dir + '/movie_conversations.txt', args.out_file, lines, movie_metadata)
+        parse_conversations(args.in_dir + '/movie_conversations.txt', args.out_dir, lines, movie_metadata)
     except (FileNotFoundError, InvalidFormatError) as e:
         print(e)
         exit(1)
 
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser(description='Process Cornell Movie Dialogue Corpus', formatter_class=argparse.ArgumentDefaultsHelpFormatter)
+    parser.add_argument('-i', '--in_dir', type=str, help='directory to process', default='.')
+    parser.add_argument('-o', '--out_file', type=str, help='file to output', default='cornell_movie_dialogue.txt')
+    parser.add_argument('-r', '--raw_name', action='store_true', help='use speaker tags as they appear in the dataset instead of normalized names')
+    parser.add_argument('-s', '--stats', action='store_true', help='store stats')
+    args = parser.parse_args()
+    parse(None)

utils/discord_parse.py

@@ -71,13 +71,15 @@ def dump_stats(s):
     stats = {}
     if os.path.exists('stats.json'):
         stats = json.load(open('stats.json', 'r', encoding='utf-8'))
+        if 'discord' not in stats:
+            stats['discord'] = {}
     else:
         stats = {'discord': {}}
     stats['discord'][s[0]['guild']+' - '+s[0]['name']] = s[1]
     with open('stats.json', 'w', encoding='utf-8') as f:
         json.dump(stats, f)
 
-if __name__ == '__main__':
+def parse(args=None):
     parser = argparse.ArgumentParser(description='Process Discord JSONs')
     parser.add_argument('-f', '--file', type=str, help='file to process', required=False)
     parser.add_argument('-a', '--anonymous', action='store_true', help='anonymous author')
@@ -85,7 +87,7 @@ if __name__ == '__main__':
     parser.add_argument('-o', '--out_dir', type=str, help='directory to output', required=False, default='./data/discord')
     parser.add_argument('-d', '--dl', type=str, help='json file containing channel IDs to download', required=False)
     parser.add_argument('-t', '--token', type=str, help='discord auth token', required=False)
-    parser.add_argument('-s', '--stats', action='store_true', help='write to stats')
+    parser.add_argument('-s', '--stats', action='store_true', help='write to stats', default=True)
     args = parser.parse_args()
 
     if args.dl:
@@ -110,4 +112,7 @@ if __name__ == '__main__':
                     dump_stats(s)
             except json.JSONDecodeError:
                 print(f'JSON Validation error in "{file}", skipping.')
-                continue
+                continue
+
+if __name__ == '__main__':
+    parse()

utils/eratoho_parse.py

@@ -7,9 +7,6 @@ import re
 
 from clean import clean
 
-def format_channel(metadata):
-	return f'[Name: #{metadata["name"]}; Description: {metadata["topic"]}; Guild: {metadata["guild"]}]'
-
 def worker_parse(filename, out_filename=None, **kwargs):
 	filtered_lines = []
 	with open(filename, 'r', encoding='utf-8') as f:
@@ -72,16 +69,16 @@ def worker_parse(filename, out_filename=None, **kwargs):
 					
 
 def dump_stats(length, character, index):
-	stats = {}
+	stats = {'eratoho': {}}
 	if os.path.exists('stats.json'):
 		stats = json.load(open('stats.json', 'r', encoding='utf-8'))
-	else:
-		stats = {'eratoho': {}}
+	if 'eratoho' not in stats:
+		stats['eratoho'] = {}
 	stats['eratoho'][str(character)+' - '+str(index)] = length
 	with open('stats.json', 'w', encoding='utf-8') as f:
 		json.dump(stats, f)
 
-if __name__ == '__main__':
+def parse():
 	parser = argparse.ArgumentParser(description='Process Discord JSONs')
 	parser.add_argument('-i', '--in_dir', type=str, help='directory to process', required=False, default='./raw/eratoho')
 	parser.add_argument('-o', '--out_dir', type=str, help='directory to output', required=False, default='./data/eratoho')
@@ -102,8 +99,6 @@ if __name__ == '__main__':
 		'clownpiece': 'Clownpiece'
 	}
 
-	print('wat')
-
 	if args.in_dir and args.out_dir:
 		if not os.path.exists(args.out_dir):
 			os.mkdir(args.out_dir)
@@ -111,4 +106,7 @@ if __name__ == '__main__':
 		for file in files:
 			lines = worker_parse(file, out_filename=args.out_dir+'/'+file.split('/')[-1].replace('.ERB', '.txt'), character=names[file.split('/')[-1].split('_')[0]], index=int(file.split('/')[-1].split('_')[1].split('.')[0]))
 			if args.stats:
-				dump_stats(lines, names[file.split('/')[-1].split('_')[0]].split('.')[0], file.split('/')[-1].split('_')[1].split('.')[0])
+				dump_stats(lines, names[file.split('/')[-1].split('_')[0]].split('.')[0], file.split('/')[-1].split('_')[1].split('.')[0])
+
+if __name__ == '__main__':
+	parse()

utils/friends_dialogue_parse.py

@@ -24,7 +24,7 @@ def valid_csv(dict_reader):
 
 
 def format_conversation_metadata(utterance):
-    return f'[Season: {utterance["season_id"]}; Episode: {utterance["episode_id"]}; Conversation: {utterance["scene_id"]};]'
+    return f'[Title: Friends Season {utterance["season_id"]}]'
 
 
 def format_utterance(utterance):

utils/hourai_touhoudialogue_parser.py

@@ -0,0 +1,219 @@
+import os
+import json
+import tqdm
+import re
+
+from clean import clean
+
+authors = {
+    'Alice': 'Alice Margatroid',
+    'Hourai': 'Hourai',
+    'Shanghai': 'Shanghai',
+    'Marisa': 'Marisa Kirisame',
+    'Patchouli': 'Patchouli Knowledge',
+    'Yuuka': 'Yuuka Kazami',
+    'Cirno': 'Cirno',
+    'Remilia': 'Remilia Scarlet',
+    'Koakuma': 'Koakuma',
+    'Medicine': 'Medicine Melancholy',
+    'Shinki': 'Shinki',
+    '???': '???',
+    'Flandre': 'Flandre Scarlet',
+    'Rumia': 'Rumia',
+    'Meiling': 'Hong Meiling',
+    'Sakuya': 'Sakuya Izayoi',
+    'Reimu': 'Reimu Hakurei',
+    'Yukari': 'Yukari Yakumo',
+    'Chen': 'Chen',
+    'Sanae': 'Sanae Kochiya',
+    'Ran': 'Ran Yakumo',
+    'Letty': 'Letty Whiterock',
+    'Lyrica': 'Lyrica Prismriver',
+    'Lunasa': 'Lunasa Prismriver',
+    'Merlin': 'Merlin Prismriver',
+    'Sisters': 'Lyrica, Lunasa, and Merlin',
+    'Youmu': 'Youmu Konpaku',
+    'Yuyuko': 'Yuyuko Saigyouji',
+    'Marisa\nYuyuko': 'Marisa and Yuyuko',
+    'Reimu\nYuyuko': 'Reimu and Yuyuko',
+    'Sakuya\nYuyuko': 'Sakuya and Yuyuko',
+    'Keine': 'Keine Kamishirasawa',
+    'Kaguya': 'Kaguya Houraisan',
+    'Mokou': 'Fujiwara no Mokou',
+    'Wriggle': 'Wriggle Nightbug',
+    'Mystia': 'Mystia Lorelei',
+    'Reisen': 'Reisen Udongein Inaba',
+    'Eirin': 'Eirin Yagokoro',
+    '\xa0': '???',
+    'Aya': 'Aya Shameimaru',
+    'Tewi': 'Tewi Inaba',
+    'Komachi': 'Komachi Onozuka',
+    'Eiki': 'Eiki Shiki',
+    'Kanako': 'Kanako Yasaka',
+    'Suwako': 'Suwako Moriya',
+    'Minoriko': 'Minoriko Aki',
+    'Hina': 'Hina Kagiyama',
+    'Nitori': 'Nitori Kawashiro',
+    'Iku': 'Iku Nagae',
+    'Tenshi': 'Tenshi Hinanawi',
+    'Suika': 'Suika Ibuki',
+    'Koishi': 'Koishi Komeiji',
+    'Satori': 'Satori Komeiji',
+    'Yamame': 'Yamame Kurodani',
+    'Parsee': 'Parsee Mizuhashi',
+    'Yuugi': 'Yuugi Hoshiguma',
+    'Rin': 'Rin Kaenbyou',
+    'Utsuho': 'Utsuho Reiuji',
+    'Kogasa': 'Kogasa Tatara',
+    'Nue': 'Nue Houjuu',
+    'Nazrin': 'Nazrin',
+    'Ichirin': 'Ichirin Kumoi',
+    'Minamitsu': 'Minamitsu Murasa',
+    'Shou': 'Shou Toramaru',
+    'Byakuren': 'Byakuren Hijiri',
+    'Master Big Catfish': 'Master Big Catfish',
+    'Hatate': 'Hatate Himekaidou',
+    'Luna': 'Luna Child',
+    'Star': 'Star Sapphire',
+    'Sunny': 'Sunny Milk',
+    'Mamizou': 'Mamizou Futatsuiwa',
+    'Kyouko': 'Kyouko Kasodani',
+    'Yoshika': 'Yoshika Miyako',
+    'Seiga': 'Seiga Kaku',
+    'Futo': 'Mononobe no Futo',
+    'Miko': 'Toyosatomimi no Miko',
+    'Hijiri': 'Byakuren Hijiri',
+    'Kokoro': 'Hata no Kokoro',
+    '<miko2>': 'Toyosatomimi no Miko\'s Clone',
+    'Benben': 'Benben Tsukumo',
+    'Yatsuhashi': 'Yatsuhashi Tsukumo',
+    'Raiko': 'Raiko Horikawa',
+    'Wakasagihime': 'Wakasagihime',
+    'Sekibanki': 'Sekibanki',
+    'Kagerou': 'Kagerou Imaizumi',
+    'Seija': 'Seija Kijin',
+    'Shinmyoumaru': 'Shinmyoumaru',
+    'Kasen': 'Kasen Ibaraki',
+    'Usami': 'Sumireko Usami',
+    'Doremy': 'Doremy Sweet',
+    'Junko': 'Junko',
+    'Hecatia': 'Hecatia Lapislazuli',
+    'Seiran': 'Seiran',
+    'Ringo': 'Ringo',
+    'Sagume': 'Sagume Kishin',
+    'Clownpiece': 'Clownpiece',
+    'Udonge': 'Reisen Udongein Inaba',
+    'Joon': 'Joon Yorigami',
+    'Shion': 'Shion Yorigami',
+    'Joon\nShion': 'Joon and Shion',
+    'Usami2': 'Sumireko Usami\'s Clone',
+    'Usami3': 'Sumireko Usami\'s Other Clone',
+    'Shinmyoumaru2': 'Shinmyoumaru\'s Clone',
+    'Mai': 'Mai',
+    'Satono': 'Satono Nishida',
+    'Okina': 'Okina Matara',
+    'Eternity': 'Eternity Larva',
+    'Nemuno': 'Nemuno Sakata',
+    'Aunn': 'Aunn Komano',
+    'Narumi': 'Narumi Yatadera',
+    'Sumireko': 'Sumireko Usami',
+    'Kutaka': 'Kutaka Niwatari',
+    'Marisa(Wolf)': 'Marisa Kirisame',
+    'Saki': 'Saki Kurokuma',
+    'Eika': 'Eika Ebisu',
+    'Urumi': 'Urumi Ushizaki',
+    'Yachie': 'Yachie Kicchou',
+    'Mayumi': 'Mayumi Joutouguu',
+    'Keiki': 'Keiki Haniyasushin',
+    'Marisa(Otter)': 'Marisa Kirisame',
+    'Marisa(Eagle)': 'Marisa Kirisame',
+    'Reimu(Wolf)': 'Reimu Hakurei',
+    'Reimu(Otter)': 'Reimu Hakurei',
+    'Reimu(Eagle)': 'Reimu Hakurei',
+    'Youmu(Wolf)': 'Youmu Konpaku',
+    'Youmu(Otter)': 'Youmu Konpaku',
+    'Youmu(Eagle)': 'Youmu Konpaku',
+    'Tsukasa': 'Tsukasa Kudamaki',
+    'Momoyo': 'Momoyo Himemushi',
+    'Misumaru': 'Misumaru Tamatsukuri',
+    'Mike': 'Mike Goutokuji',
+    'Takane': 'Takane Yamashiro',
+    'Sannyo': 'Sannyo Komakusa',
+    'Megumu': 'Megumu Iizunamaru',
+    'Chimata': 'Chimata Tenkyuu',
+}
+
+# regex to remove strings like <balloon$a11x3>
+exp = re.compile(r'<.*?>')
+
+def worker_parse_hourai(filename, out_filename=None):
+    # parse csv file
+    messages = []
+
+    with open(filename, 'r', encoding='utf-8') as f:
+        data = f.read()
+        lines = data.split('\n')
+        for line in lines:
+            if line.startswith('name,line'):
+                continue
+            line = line.split(',')
+            if len(line) < 2:
+                continue
+            author = line[0]
+            text = clean(line[1].replace('\"', ''))
+            messages.append(f'{authors[author]}: {text}')
+    
+    # write to file
+    with open(out_filename, 'w', encoding='utf-8') as f:
+        txt = '\n'.join(messages)
+        f.write('⁂\n[Title: Hourai; Genre: Touhou Project]\n⁂\n')
+        f.write(txt)
+    
+    return len(messages)
+
+def worker_parse_dialogue(filename, out_filename=None):
+    # open json
+    messages = []
+    with open(filename, 'r', encoding='utf-8') as f:
+        data = json.load(f)
+        for message in data:
+            # first, check if author is in dict values
+            author = message['Name']
+            # iterate over values
+            for key, value in authors.items():
+                if value == author:
+                    author = key
+                    break
+            author = authors[author]
+            text = clean(message['Text']).replace('&#160;?"@', '').replace('&amp;', '').replace('&#160;', '').replace('\ ', '').replace('\. ', '\n')
+            for key, value in authors.items():
+                text = text.replace(key, value)
+            text = exp.sub('', text)
+            messages.append(f'{author}: {text}')
+    
+    # write to file
+    with open(out_filename, 'w', encoding='utf-8') as f:
+        txt = '\n'.join(messages)
+        f.write('⁂\n[Title: The Touhou Project]\n⁂\n')
+        f.write(txt.replace('\n\n', '\n'))
+    
+    return len(messages)
+
+def dump_stats(length, name):
+    stats = {'misc': {}}
+    if os.path.exists('stats.json'):
+        stats = json.load(open('stats.json', 'r', encoding='utf-8'))
+    if 'misc' not in stats:
+        stats['misc'] = {}
+    stats['misc'][name] = length
+    with open('stats.json', 'w', encoding='utf-8') as f:
+        json.dump(stats, f)
+
+def parse():
+    if not os.path.exists('./data/misc'):
+        os.mkdir('./data/misc')
+    dump_stats(worker_parse_hourai('./raw/misc/lines.csv', './data/misc/lines.txt'), 'hourai')
+    dump_stats(worker_parse_dialogue('./raw/misc/dialogue.json', './data/misc/dialogue.txt'), 'dialogue')
+
+if __name__ == '__main__':
+    parse()

utils/stats.py

@@ -13,10 +13,14 @@ if __name__ == '__main__':
     print('-- Convo1 Dataset Stats --')
     print(f'Datasets Available: {len(stats)}')
 
+    total_total = 0
     for dataset in stats.keys():
         total = 0
         print(f'\n-- Dataset Name: {dataset} --')
         for guild in sorted(stats[dataset].items(), key=lambda x: x[1], reverse=True):
             print(f'{guild[0]}: {guild[1]}')
             total += guild[1]
-        print(f'\nTotal Messages in {dataset}: {total}\n')
+        print(f'\nTotal Messages in {dataset}: {total}\n')
+        total_total += total
+    
+    print(f'\nTotal Messages in Convo Dataset: {total_total}\n')

utils/{07_parse.py → umi_parse.py}

@@ -1,4 +1,4 @@
-import parse
+import parse as _parse
 import os
 import json
 
@@ -22,7 +22,7 @@ def worker_parse(filename, out_filename=None, stats=True, index=0):
 
     jap_characters = ['。', '「', '」', '？', '！', '、', '…', 'ー']
 
-    author = parse.parse('\ufeff== {0} ==\n', lines[0])[0]
+    author = _parse.parse('\ufeff== {0} ==\n', lines[0])[0]
 
     with open(out_filename, 'w') as f:
         for line in lines:
@@ -33,7 +33,7 @@ def worker_parse(filename, out_filename=None, stats=True, index=0):
                 continue
             if line.startswith('\n==') or line.startswith('=='):
                 try:
-                    author = parse.parse('== {0} ==\n', line)[0]
+                    author = _parse.parse('== {0} ==\n', line)[0]
                 except:
                     author = 'Narrator'
                 continue
@@ -72,13 +72,19 @@ def worker_parse(filename, out_filename=None, stats=True, index=0):
         f.write(metadata)
         f.write(clean_07(lines))
 
-if __name__ == '__main__':
+def parse():
     files = [
         './raw/visualnovel/ep1-4_EN__JP.txt',
         './raw/visualnovel/ep5-8_EN__JP.txt',
     ]
 
     output_files = [ i.replace('raw', 'data') for i in files ]
+
+    if not os.path.exists('./data/visualnovel'):
+        os.mkdir('./data/visualnovel')
     
     for i in range(len(files)):
-        worker_parse(files[i], output_files[i], stats=True, index=i)
+        worker_parse(files[i], output_files[i], stats=True, index=i)
+
+if __name__ == '__main__':
+    parse()

utils/valhalla_parse.py

@@ -31,5 +31,8 @@ def worker_parse(filename, out_filename, stats=True):
             dump_stats(len(msgs), day)
 
 
+def parse():
+    worker_parse('./raw/visualnovel/valhalla.txt', './data/visualnovel/valhalla.txt')
+
 if __name__ == '__main__':
-    worker_parse('./raw/visualnovel/valhalla.txt', './data/visualnovel/valhalla.txt')
+    parse()