def final(silent_skip: bool, process_: bool):
"""Generates the latest application static data.json file, used by the backend API."""
descriptions = load_file(os.path.join(DATA_DIR, 'descriptions.json'), True)
seasons = [{'season_id': season, 'episodes': []} for season in range(1, 10)]
if process_:
logger.info('Processing before building final.json')
try:
process(["--all"])
except:
pass
for season_id, episode_id in get_episodes():
# Load data file
try:
episode_data = load_file(get_filepath(season_id, episode_id, 'processed'), True)
except FileNotFoundError:
if not silent_skip:
logger.warning(
f'No data for Season {season_id}, Episode {episode_id} available. Null data inserted.')
episode_data = None
description = descriptions[season_id - 1][episode_id - 1]
seasons[season_id - 1]['episodes'].append(
{
'title': description['title'].strip(),
'description': description['description'].strip(),
'episode_id': episode_id,
'characters': get_appearances(season_id, episode_id),
'scenes': episode_data
}
)
logger.info('Saving to data.json')
save_file(os.path.join(DATA_DIR, 'data.json'), seasons, True)
def character():
"""
Uses algolia.json to build a characters.json file, a masterlist of quotes separated by the speaker.
Speakers not considered 'main characters' are excluded from the list.
This file also pulls information to build character descriptions and other relevant information.
"""
data = load_file(os.path.join(DATA_DIR, 'algolia.json'), True)
descriptions = load_file(os.path.join(DATA_DIR, 'character_descriptions.json'), True)
key_list = [('speaker',), ('text',), ('season',), ('episode_rel', 'episode'), ('section_rel', 'scene'),
('quote_rel', 'quote')]
master = map(lambda item: algolia_transform(item, key_list), filter(lambda item: True, data))
# Separate the quotes based on speaker
char_data = defaultdict(list)
for quote in master:
char_data[character_id(quote['speaker'])].append(quote)
final_data = {}
for character, quotes in char_data.items():
final_data[character] = {'quotes': quotes, 'summary': None, 'name': None}
if character in descriptions.keys():
final_data[character]['name'] = descriptions[character].get('name')
final_data[character]['summary'] = descriptions[character].get('summary')
# Save to characters.json
save_file(os.path.join(DATA_DIR, 'characters.json'), final_data, True)
def algolia(silent_skip: bool, process_: bool):
"""
Generates algolia.json, a all encompassing file for Algolia's search index.
"""
if process_:
logger.info('Processing before building algolia.json')
try:
process(["--all", '--silent'])
except:
pass
data = []
episode_num_abs, section_num_abs, quote_num_abs = 0, 0, 0
for season, episode in get_episodes():
episode_num_abs += 1
try:
episode_data = load_file(get_filepath(season, episode, 'processed'), True)
except FileNotFoundError:
if not silent_skip:
logger.warning(f'Skipping Season {season}, Episode {episode}. No episode data file found.')
else:
for section_num_rel, section in enumerate(episode_data, start=1):
section_num_abs += 1
for quote_num_rel, quote in enumerate(section['quotes'], start=1):
quote_num_abs += 1
# Relative position
quote['quote_rel'] = quote_num_rel
quote['section_rel'] = section_num_rel
quote['episode_rel'] = episode
# Absolute position
quote['quote_abs'] = quote_num_abs
quote['section_abs'] = section_num_abs
quote['episode_abs'] = episode_num_abs
quote['season'] = season
quote['is_deleted'] = 'deleted' in section.keys()
quote['deleted_section'] = section.get('deleted')
data.append(quote)
logger.info(f'Saving {len(data):,} quotes to algolia.json')
save_file(os.path.join(DATA_DIR, 'algolia.json'), data, True)
def process(season: Optional[int], episode: Optional[int], all_: bool, report: bool):
"""
Processes manually processed raw quote data into JSON.
"""
episodes: List[Tuple[int, int]]
if all_:
episodes = list(get_episodes())
elif season:
if episode:
if verify_episode(season, episode):
episodes = [(season, episode)]
else:
logger.error(f'Season {season}, Episode {episode} is not a valid combination.')
return
else:
episodes = list(get_episodes(season=season))
logger.info(f'Processing Season {season}...')
else:
if episode:
logger.info('You must specify more than just an episode.')
else:
logger.info('You must specify which episodes to process.')
logger.info('Check --help for more information on this command.')
return
quote: Union[str, List[str]]
section_num: int
for _season, _episode in episodes:
sections = []
try:
preprocessed_data = load_file(get_filepath(_season, _episode, 'raw'))
for section_num, raw_section in enumerate(re.split('^-', preprocessed_data, flags=re.MULTILINE), start=1):
section = {
'quotes': []
}
section_data = list(raw_section.strip().split('\n'))
if section_data[0].startswith('!'):
section['deleted'] = int(re.search('!(\d+)', section_data.pop(0)).group(1))
for quote in section_data:
quote = quote.split('|', 1)
section['quotes'].append(
{
'speaker': quote[0],
'text': quote[1]
}
)
sections.append(section)
except FileNotFoundError:
logger.info(f'Skipped Season {_season}, Episode {_episode}, no file found.')
continue
except:
logger.exception(f'Skipped Season {_season}, Episode {_episode}: Malformed data.')
if quote:
logger.info(
f'Last quote seen "{quote if type(quote) is str else "|".join(quote)}" in section {section_num}')
else:
# Save processed data
save_file(get_filepath(_season, _episode, 'processed'), sections, True)
if report:
deleted_count = [0, set()]
quote_count = 0
speakers = set()
for section in sections:
quote_count += len(section['quotes'])
if 'deleted' in section.keys():
deleted_count[0] += 1
deleted_count[1].add(section['deleted'])
for quote in section['quotes']:
speakers.add(quote['speaker'])
logger.debug(f'{quote_count} quotes.')
logger.debug(f'{deleted_count[0]} different deleted sections, {len(deleted_count[1])} unique.')
logger.info(f'{len(speakers)} Speakers:')
logger.info(', '.join(speakers))
def preprocess(season: int, episode: int, all: bool, overwrite: bool, silent_skip: bool, silent_skip_missing: bool,
silent_skip_existing: bool):
"""
Pre-processes raw HTML files into mangled custom quote data.
Custom quote data requires manual inspection and formatting, making it a dangerous operation that may overwrite
precious quote data.
"""
print(silent_skip_existing)
episodes: List[Tuple[int, int]]
if all:
episodes = list(get_episodes())
elif season:
if episode:
if verify_episode(season, episode):
episodes = [(season, episode)]
else:
logger.error(f'Season {season}, Episode {episode} is not a valid combination.')
return
else:
episodes = list(get_episodes(season=season))
logger.info(f'Preprocessing Season {season}...')
else:
if episode:
logger.info('You must specify more than just an episode.')
else:
logger.info('You must specify which episodes to pre-process.')
logger.info('Check --help for more information on this command.')
return
for season, episode in episodes:
# Overwrite protection
save_path = get_filepath(season, episode, 'raw')
if os.path.exists(save_path) and not overwrite:
if (not silent_skip) or (not silent_skip_existing):
logger.info(f'Skipping Season {season}, Episode {episode}, file already exists. Skipping processing.')
continue
try:
page_data = load_file(get_filepath(season, episode, 'html'), False)
except FileNotFoundError:
if not silent_skip or not silent_skip_missing:
logger.warning(f'No data for Season {season}, Episode {episode} available. Skipping processing.')
else:
soup = BeautifulSoup(page_data, "html.parser")
data = []
sections = soup.find_all(attrs={"class": "quote"})
for section in sections:
for br in section.find_all('br'):
br.replace_with("\n" + br.text)
for line in section.get_text().split('\n'):
data.append(line.strip())
data.append('-')
data.pop(-1)
data = '\n'.join(data)
save_file(save_path, data, False)