import logging import re parlog = logging VALID_PULSES = set([4, 8, 12, 16, 24, 32, 48, 64, 96, 192]) int_parser = lambda x: int(x.strip()) if x.strip() else None bool_parser = lambda x: True if x.strip() == 'YES' else False str_parser = lambda x: x.strip() if x.strip() else None float_parser = lambda x: float(x.strip()) if x.strip() else None def kv_parser(k_parser, v_parser): def parser(x): if not x: return (None, None) k, v = x.split('=', 1) return k_parser(k), v_parser(v) return parser def list_parser(x_parser): def parser(l): l_strip = l.strip() if len(l_strip) == 0: return [] else: return [x_parser(x) for x in l_strip.split(',')] return parser def bpms_parser(x): bpms = list_parser(kv_parser(float_parser, float_parser))(x) if len(bpms) == 0: raise ValueError('No BPMs found in list') if bpms[0][0] != 0.0: raise ValueError('First beat in BPM list is {}'.format(bpms[0][0])) # make sure changes are nonnegative, take last for equivalent beat_last = -1.0 bpms_cleaned = [] for beat, bpm in bpms: if beat == None or bpm == None: raise ValueError('Empty BPM found') if bpm <= 0.0: raise ValueError('Non positive BPM found {}'.format(bpm)) if beat == beat_last: bpms_cleaned[-1] = (beat, bpm) continue bpms_cleaned.append((beat, bpm)) if beat <= beat_last: raise ValueError('Descending list of beats in BPM list') beat_last = beat if len(bpms) != len(bpms_cleaned): parlog.warning('One or more (beat, BPM) pairs begin on the same beat, using last listed') return bpms_cleaned def stops_parser(x): stops = list_parser(kv_parser(float_parser, float_parser))(x) beat_last = -1.0 for beat, stop_len in stops: if beat == None or stop_len == None: raise ValueError('Bad stop formatting') if beat < 0.0: raise ValueError('Bad beat in stop') if stop_len == 0.0: continue if beat <= beat_last: raise ValueError('Nonascending list of beats in stops') beat_last = beat return stops def notes_parser(x): pattern = r'([^:]*):' * 5 + r'([^;:]*)' notes_split = re.findall(pattern, x) if len(notes_split) != 1: raise ValueError('Bad formatting of notes section') notes_split = notes_split[0] if (len(notes_split) != 6): raise ValueError('Bad formatting within notes section') # parse/clean measures measures = [measure.splitlines() for measure in notes_split[5].split(',')] measures_clean = [] for measure in measures: measure_clean = filter(lambda pulse: not pulse.strip().startswith('//') and len(pulse.strip()) > 0, measure) measures_clean.append(list(measure_clean)) if len(measures_clean) > 0 and len(measures_clean[-1]) == 0: measures_clean = measures_clean[:-1] # check measure lengths for measure in measures_clean: if len(measure) == 0: raise ValueError('Found measure with 0 notes') if not len(measure) in VALID_PULSES: parlog.warning('Nonstandard subdivision {} detected, allowing'.format(len(measure))) chart_type = str_parser(notes_split[0]) if chart_type not in ['dance-single', 'dance-double', 'dance-couple', 'lights-cabinet']: raise ValueError('Nonstandard chart type {} detected'.format(chart_type)) return (str_parser(notes_split[0]), str_parser(notes_split[1]), str_parser(notes_split[2]), int_parser(notes_split[3]), list_parser(float_parser)(notes_split[4]), measures_clean ) def unsupported_parser(attr_name): def parser(x): raise ValueError('Unsupported attribute: {} with value {}'.format(attr_name, x)) return None return parser ATTR_NAME_TO_PARSER = { 'title': str_parser, 'subtitle': str_parser, 'artist': str_parser, 'titletranslit': str_parser, 'subtitletranslit': str_parser, 'artisttranslit': str_parser, 'genre': str_parser, 'credit': str_parser, 'banner': str_parser, 'background': str_parser, 'lyricspath': str_parser, 'cdtitle': str_parser, 'music': str_parser, 'offset': float_parser, 'bpms': bpms_parser, 'stops': stops_parser, 'samplestart': float_parser, 'samplelength': float_parser, 'displaybpm': str_parser, 'selectable': bool_parser, 'bgchanges': str_parser, 'bgchanges2': str_parser, 'fgchanges': str_parser, 'keysounds': str_parser, 'musiclength': float_parser, 'musicbytes': int_parser, 'attacks': str_parser, 'timesignatures': list_parser(kv_parser(float_parser, kv_parser(int_parser, int_parser))), 'warps': str_parser, 'notes': notes_parser } ATTR_MULTI = ['notes'] def parse_sm_txt(sm_txt): attrs = {attr_name: [] for attr_name in ATTR_MULTI} for attr_name, attr_val in re.findall(r'#([^:]*):([^;]*);', sm_txt): attr_name = attr_name.lower() if attr_name not in ATTR_NAME_TO_PARSER: parlog.warning('Found unexpected attribute {}:{}, ignoring'.format(attr_name, attr_val)) continue attr_val_parsed = ATTR_NAME_TO_PARSER[attr_name](attr_val) if attr_name in attrs: if attr_name not in ATTR_MULTI: if attr_val_parsed == attrs[attr_name]: continue else: raise ValueError('Attribute {} defined multiple times'.format(attr_name)) attrs[attr_name].append(attr_val_parsed) else: attrs[attr_name] = attr_val_parsed for key in list(attrs): if attrs[key] is None or attrs[key] == []: attrs.pop(key) # for attr_name, attr_val in attrs.items(): # if attr_val == None or attr_val == []: # del attrs[attr_name] return attrs TIME_SIG={ 'timesignatures': list_parser(kv_parser(float_parser, kv_parser(int_parser, int_parser))), } def extract_time_signature(sm_txt): attrs = {attr_name: [] for attr_name in ATTR_MULTI} for attr_name, attr_val in re.findall(r'#([^:]*):([^;]*);', sm_txt): attr_name = attr_name.lower() if attr_name not in TIME_SIG: # parlog.warning('Found unexpected attribute {}:{}, ignoring'.format(attr_name, attr_val)) continue attr_val_parsed = TIME_SIG[attr_name](attr_val) if attr_name in attrs: if attr_name not in ATTR_MULTI: if attr_val_parsed == attrs[attr_name]: continue else: raise ValueError('Attribute {} defined multiple times'.format(attr_name)) attrs[attr_name].append(attr_val_parsed) else: attrs[attr_name] = attr_val_parsed for key in list(attrs): if attrs[key] is None or attrs[key] == []: attrs.pop(key) # for attr_name, attr_val in attrs.items(): # if attr_val == None or attr_val == []: # del attrs[attr_name] return attrs