scripts/musedata2ly.py

   1 #!@PYTHON@
   2
   3 # musedata = musedata.stanford.edu
   4 # musedata = COBOL for musicians.
   5
   6
   7 # TODO
   8 #
   9 # * clefs,
  10 # * keys,
  11 # * staffs,
  12 # * multiple voices (they use `Backspace' (shudder)
  13 # * tuplets
  14 #
  15
  16 #
  17 # I completely forgot how this was supposed to work --hwn 5/2002
  18 #
  19 #
  20
  21 import re
  22 import sys
  23 import string
  24 import getopt
  25 import os
  26 program_name = 'musedata2ly'
  27 version = '@TOPLEVEL_VERSION@'
  28 if version == '@' + 'TOPLEVEL_VERSION' + '@':
  29         version = '(unknown version)'      # uGUHGUHGHGUGH
  30
  31
  32
  33 ref_header_dict = {
  34         'COM': 'composer',
  35         'OPR': 'collection',
  36         'OTL': 'title',
  37         'OMV': 'subtitle',
  38         'YOR': 'source',
  39         'AGN': 'instrument',
  40         'END': 'encodingdate',
  41         'CDT': 'date',
  42         'OCY': 'composedin',
  43         'AST': 'genre',
  44         'YEC': 'copyright',
  45         'YEM': 'license',
  46         'YEN': 'encodingcountry',
  47         'EED': 'editor',
  48         'SCA': 'opus',
  49         'ONM': 'onm',
  50         'ENC': 'musedataencoder',
  51         'KEY': 'musedatakey',
  52         'AFT': 'musedatastage'
  53         }
  54
  55
  56 class Ref_parser:
  57         def __init__ (self, fn):
  58                 self.dict = {}
  59
  60                 ls = open (fn).readlines ()
  61                 self.parse (ls)
  62         def parse (self,ls):
  63                 for l in ls:
  64                         m = re.match('!!!([A-Z]+):[ \t]+(.*)$',l)
  65                         if m:
  66                                 key = m.group(1)
  67                                 val = m.group (2)
  68                                 val = re.sub ('[ \t]+', ' ', val)
  69                                 try:
  70
  71                                         key =ref_header_dict [key]
  72                                 except KeyError:
  73                                         sys.stderr.write ('\nUnknown ref key \`%s\'' % key)
  74                                 s = ''
  75                                 try:
  76                                         s = self.dict[key]
  77                                 except KeyError:
  78                                         pass
  79
  80                                 s = s + val
  81                                 self.dict[key] = s
  82         def dump( self):
  83                 str = ''
  84                 for (k,v) in self.dict.items ():
  85                         str = str +'  %s = "%s"\n' % (k,v)
  86                 str = '\\header {\n%s}' % str
  87                 return str
  88
  89 verbose = 0
  90
  91
  92 actab = {-2: 'eses', -1: 'es', 0 : '', 1: 'is', 2:'isis'}
  93
  94 def pitch_to_lily_string (tup):
  95         (o,n,a) = tup
  96
  97         nm = chr((n + 2) % 7 + ord ('a'))
  98         nm = nm + actab[a]
  99         if o > 0:
 100                 nm = nm + "'" * o
 101         elif o < 0:
 102                 nm = nm + "," * -o
 103         return nm
 104
 105 def get_key (s):
 106         i = string.atoi (s)
 107         return ''
 108
 109 def get_timesig (s):
 110         return '\\time %s\n' % s
 111
 112
 113 divisions = 4
 114 def get_divisions_per_quarter (s):
 115         divisions = string.atoi (s)
 116         return ''
 117
 118 def get_directive (s):
 119         return '%% %s\n' % s
 120
 121 def get_transposing (s):
 122         return ''
 123
 124 def get_num_instruments (s):
 125         return ''
 126
 127 def get_lilypond_notename (p, ac):
 128         if p > 5:
 129                 p = p - 7
 130         s = chr (p + ord ('c'))
 131         infix = 'i'
 132         if ac < 0:
 133                 infix = 'e'
 134                 ac = -ac
 135
 136         while ac:
 137                 s = s + infix + 's'
 138                 ac = ac - 1
 139         return s
 140 def get_clef ():
 141         return ''
 142
 143 SPACES = ' '
 144 DIGITS = "0123456789"
 145
 146
 147 clef_dict = {
 148 04: 'treble',
 149 13 : 'alto',
 150 22: 'bass',
 151 }
 152 attr_dict = {
 153         'C' : get_clef,
 154         'K' : get_key ,
 155         'T' : get_timesig,
 156         'Q' : get_divisions_per_quarter,
 157         'D' : get_directive,
 158         'X' : get_transposing,
 159         'I': get_num_instruments,
 160         }
 161
 162 class Attribute_set:
 163         def __init__ (self, dict):
 164                 self.dict = dict
 165         def dump (self):
 166                 s = ''
 167                 if self. dict.has_key ('T'):
 168                         s = s+ get_timesig  (self.dict['T'])
 169
 170                 return s
 171
 172
 173 script_table = {
 174 'v': '\\upbow',
 175 'n': '\\downbow',
 176 'o': '\\harmonic',
 177 '0': '"openstring',
 178 'Q': '\\thumb',
 179 '>': '^',
 180 'V': '^',
 181 '.': '.',
 182 '_': '-',
 183 '=': '"det leg"',
 184 'i': '|',
 185 's': '"\\\\textsharp"',
 186 'n': '"\\\\textnatural"',
 187 'b': '"\\\\textflat"',
 188 'F': '\\fermata',
 189 'E': '\\fermata',
 190 }
 191
 192
 193 class Chord:
 194         def __init__ (self):
 195                 self.pitches = []
 196                 self.grace = 0
 197                 self.cue = 0
 198                 self.slurstart = []
 199                 self.slurstop  = []
 200                 self.scripts = []
 201                 self.syllables = []
 202                 self.dots = 0
 203                 self.basic_duration = 4
 204                 self.tied = 0
 205
 206                 self.note_suffix = self.note_prefix = ''
 207                 self.chord_suffix = self.chord_prefix = ''
 208
 209         def add_script (self,s):
 210                 self.scripts.append (s)
 211         def set_duration (self, d):
 212                 self.basic_duration = d
 213         def add_syllable (self, s):
 214                 self.syllables.append (s)
 215         def add_pitch (self,t):
 216                 self.pitches.append (t)
 217
 218         def dump (self):
 219                 str = ''
 220
 221                 sd = ''
 222                 if self.basic_duration == 0.5:
 223                         sd = '\\breve'
 224                 else:
 225                         sd = '%d' % self.basic_duration
 226
 227                 sd = sd + '.' * self.dots
 228
 229                 for p in self.pitches:
 230                         if str:
 231                                 str = str + ' '
 232                         str = str + pitch_to_lily_string (p)
 233
 234                 if len (self.pitches) > 1:
 235                         str = '<%s>' % str
 236                 elif len (self.pitches) == 0:
 237                         str = 'r'
 238
 239                 str = str + sd + '(' * len (self.slurstart) + ')' * len (self.slurstart)
 240                 for s in self.scripts:
 241                         str = str + '-' + s
 242
 243                 str = self.note_prefix +str  + self.note_suffix
 244                 str = self.chord_prefix + str + self.chord_suffix
 245                 return str
 246
 247 class Measure_start:
 248         def dump (self):
 249                 return ' |\n'
 250
 251 class Parser:
 252         def append_entry (self, e):
 253                 self.entries.append (e)
 254         def append_chord (self,c ):
 255                 self.chords.append (c)
 256                 self.entries.append (c)
 257         def last_chord (self):
 258                 return self.chords[-1]
 259         def __init__ (self, fn):
 260                 self.divs_per_q = 1
 261                 self.header_dict = {
 262                         'tagline' :'automatically converted from Musedata',
 263                         'copyright' : 'all rights reserved -- free for noncommercial use'
 264                         #  musedata license (argh)
 265                         }
 266                 self.entries = []
 267                 self.chords = []
 268
 269
 270                 lines = open (fn).readlines ()
 271                 lines = map (lambda x: re.sub ("\r$", '', x), lines)
 272                 lines = self.parse_header (lines)
 273                 lines = self.append_lines (lines)
 274                 str = string.join (lines, '\n')
 275                 lines = re.split ('[\n\r]+', str)
 276                 self.parse_body (lines)
 277
 278         def parse_header (self, lines):
 279                 enter = string.split (lines[3], ' ')
 280                 self.header_dict['enteredby']  = string.join (enter[1:])
 281                 self.header_dict['enteredon'] = enter[0]
 282                 self.header_dict['opus'] = lines[4]
 283                 self.header_dict['source'] = lines[5]
 284                 self.header_dict['title'] = lines[6]
 285                 self.header_dict['subtitle'] = lines[7]
 286                 self.header_dict['instrument']= lines[8]
 287                 self.header_dict['musedatamisc'] =lines[9]
 288                 self.header_dict['musedatagroups'] =lines[10]
 289                 self.header_dict['musedatagroupnumber']=lines[11]
 290                 lines =  lines[12:]
 291                 comment = 0
 292                 while lines:
 293                         if lines[0][0]  == '$':
 294                                 break
 295                         lines = lines[1:]
 296                 return lines
 297
 298         def parse_musical_attributes (self,l):
 299                 atts = re.split('([A-Z][0-9]?):', l)
 300                 atts = atts[1:]
 301                 found = {}
 302                 while len (atts):
 303                         id = atts[0]
 304                         val = atts[1]
 305                         atts = atts[2:]
 306                         found[id] = val
 307
 308                 try:
 309                         self.divs_per_q = string.atoi (found['Q'])
 310                 except KeyError:
 311                         pass
 312
 313                 self.append_entry (Attribute_set (found))
 314         def append_entry (self, e):
 315                 self.entries.append (e)
 316
 317         def parse_line_comment (self,l):
 318                 pass
 319
 320         def parse_note_line (self,l):
 321                 ch = None
 322                 if verbose:
 323                         print DIGITS+DIGITS+DIGITS
 324                         print l
 325                 pi = l[0:5]
 326                 di = l[5:8]
 327                 tied = l[8:9] == '-'
 328
 329                 cue = grace = 0
 330                 if (pi[0] == 'g'):
 331                         grace = 1
 332                         pi = pi[1:]
 333                 elif (pi[0] == 'c'):
 334                         cue = 1
 335                         pi = pi[1:]
 336
 337                 if pi[0] == ' ':
 338                         ch = self.last_chord ()
 339                         pi = pi[1:]
 340                 else:
 341                         ch = Chord ()
 342                         self.append_chord (ch)
 343
 344
 345                 ch.cue = ch.cue or cue
 346                 ch.grace = ch.grace or grace
 347
 348                 while pi[0] in SPACES:
 349                         pi = pi[1:]
 350
 351                 if pi[0] <> 'r':
 352                         name =  ((ord (pi[0]) -ord('A')) + 5) % 7
 353                         alter = 0
 354                         pi = pi[1:]
 355                         while pi and pi[0] in '#f':
 356                                 if pi[0] == '#':
 357                                         alter = alter + 1
 358                                 else:
 359                                         alter = alter - 1
 360                                 pi = pi[1:]
 361
 362                         oct = string.atoi (pi) - 3
 363
 364                         pittup = (oct, name ,alter)
 365                         ch.add_pitch (pittup)
 366
 367                 ch.dots = 0
 368
 369                 dot_str = l[17:18]
 370                 if dot_str  == '.':
 371                         ch.dots = 1
 372                 elif dot_str == ':':
 373                         ch.dots = 2
 374
 375                 base_dur = None
 376                 if ch.cue or ch.grace:
 377                         c = di[2]
 378                         if c == '0':
 379                                 ch.accaciatura = 1
 380                         elif c == 'A':
 381                                 base_dur = 0.5
 382                         else:
 383                                 base_dur = 1 << (9 - (ord (c) - ord ('0')))
 384                 else:
 385                         fact  = (1,1)
 386                         if ch.dots == 1:
 387                                 fact = (2,3)
 388                         elif ch.dots == 2:
 389                                 fact = (4, 7)
 390
 391                         base_dur =  (4 * self.divs_per_q* fact[1]) / (string.atoi (di)* fact[0])
 392                         ch.set_duration (base_dur)
 393
 394                 ch.tied = ch.tied or tied
 395
 396                 if l[26:27] == '[':
 397                         ch.start_beam = 1
 398                 elif l[26:27] == ']':
 399                         ch.end_beam = 1
 400
 401
 402                 additional = l[32:44]
 403                 for c in additional:
 404                         if c in '([{z':
 405                                 ch.slurstart.append( 0)
 406                                 continue
 407                         elif c in ')]}x':
 408                                 ch.slurstop.append( 0)
 409                                 continue
 410
 411                         if c == '*':
 412                                 ch.start_tuplet = 1
 413                                 continue
 414                         elif c == '!':
 415                                 ch.stop_tuplet = 1
 416                                 continue
 417
 418                         if c in DIGITS:
 419                                 ch.add_script (c)
 420                                 continue
 421
 422                         if c == ' ' :
 423                                 continue
 424
 425                         try:
 426                                 scr = script_table[c]
 427                                 ch.add_script (scr)
 428                                 c = None
 429                         except KeyError:
 430                                 sys.stderr.write ("\nFixme: script `%s' not done\n" % c)
 431
 432                 text = l[40:81]
 433                 sylls = string.split (text,'|')
 434
 435                 for syl in sylls:
 436                         ch.add_syllable (syl)
 437
 438
 439         def parse_measure_line (self,l):
 440                 self.append_entry (Measure_start())
 441
 442
 443         def parse_duration (l):
 444                 s = ''
 445                 while l[0] in '0123456789':
 446                         s = s + l[0]
 447                         l= l[1:]
 448                 print l
 449                 num = string.atoi (s)
 450                 den = 4 * divisions
 451
 452                 current_dots = 0
 453                 try_dots = [3, 2, 1]
 454                 for d in try_dots:
 455                         f = 1 << d
 456                         multiplier = (2*f-1)
 457                         if num % multiplier == 0 and den % f == 0:
 458                                 num = num / multiplier
 459                                 den = den / f
 460                                 current_dots = current_dots + d
 461
 462                 if num <> 1:
 463                         sys.stderr.write ('huh. Durations left')
 464                 return '%s%s' % (den, '.' * current_dots)
 465
 466         def append_lines (self,ls):
 467                 nls = []
 468                 for l in ls:
 469                         if l[0] == 'a':
 470                                 nls[-1] = nls[-1]+l[1:]
 471                         else:
 472                                 nls.append(l)
 473                 return nls
 474         def dump (self):
 475                 s = ''
 476                 ln = ''
 477                 for e in self.entries:
 478
 479                         next = ' ' + e.dump()
 480                         if len (ln) + len (next) > 72:
 481                                 s = s +ln +  '\n'
 482                                 ln = ''
 483                         ln = ln + next
 484
 485                 s = s + ln
 486
 487                 s = '\\notes {\n %s \n}' % s
 488                 return s
 489
 490         def parse_body (self,lines):
 491                 comment_switch = 0
 492                 for l in lines:
 493                         if not l:
 494                                 continue
 495
 496                         c = l[0]
 497                         if c == '&':
 498                                 comment_switch = not comment_switch
 499                                 continue
 500
 501                         if comment_switch:
 502                                 continue
 503
 504                         if 0:
 505                                 pass
 506                         elif c == '$':
 507                                 self.parse_musical_attributes (l)
 508                         elif c == '@':
 509                                 self.parse_line_comment (l)
 510                         elif c == '*':
 511                                 self.parse_musical_directions (l)
 512                         elif c in 'ABCDEFGr ':
 513                                 self.parse_note_line (l)
 514                         elif c == 'm':
 515                                 self.parse_measure_line (l)
 516                         elif c == '/':
 517                                 break
 518                         elif c in 'PS':
 519                                 pass                    # ignore sound & print
 520                         else:
 521                                 sys.stderr.write ("\nUnrecognized record `%s'\n"%  l)
 522
 523
 524
 525
 526
 527 def help ():
 528         sys.stdout.write (
 529 """Usage: musedata2ly [OPTIONS]... FILE1 [FILE2 ...]
 530
 531 Convert musedata to LilyPond.
 532
 533 Options:
 534   -h,--help          print this help
 535   -o,--output=FILE   set output filename to FILE
 536   -v,--version       show version information
 537   -r,--ref=REF       read background information from ref-file REF
 538
 539 Musedata (http://www.ccarh.org/musedata/) is an electronic library of
 540 classical music scores, currently comprising XXX scores.  The music is
 541 encoded in so-called Musedata format
 542 (http://www.ccarh.org/publications/books/beyondmidi/online/musedata).
 543 musedata2ly converts a set of musedata files to one .ly file, and will
 544 include a \header field if a .ref file is supplied.
 545
 546 This converter is not complete -- this is left to the user as an excercise.
 547
 548 Report bugs to bug-lilypond@gnu.org.
 549
 550 Written by Han-Wen Nienhuys <hanwen@cs.uu.nl>.
 551
 552 """)
 553
 554
 555 def print_version ():
 556         sys.stdout.write ("""musedata2ly (GNU LilyPond) %s
 557
 558 This is free software.  It is covered by the GNU General Public License,
 559 and you are welcome to change it and/or distribute copies of it under
 560 certain conditions.  Invoke as `midi2ly --warranty' for more information.
 561
 562 Copyright (c) 2000--2004 by Han-Wen Nienhuys <hanwen@cs.uu.nl>.
 563 """ % version)
 564 def identify():
 565         sys.stderr.write ("%s from LilyPond %s\n" % (program_name, version))
 566
 567
 568
 569 (options, files) = getopt.getopt (sys.argv[1:], 'r:vo:h', ['verbose', 'ref=', 'help','version', 'output='])
 570 out_filename = None
 571 ref_file = None
 572 for opt in options:
 573         o = opt[0]
 574         a = opt[1]
 575         if o== '--help' or o == '-h':
 576                 help ()
 577                 sys.exit (0)
 578         elif o == '--version' or o == '-v':
 579                 print_version ()
 580                 sys.exit(0)
 581         elif o == '--ref' or o == '-r':
 582                 ref_file = a
 583         elif o == '--output' or o == '-o':
 584                 out_filename = a
 585         elif o == '--verbose' :
 586                 verbose = 1
 587         else:
 588                 print o
 589                 raise getopt.error
 590
 591 identify()
 592
 593
 594
 595 ly = ''
 596
 597
 598 found_ids = ''
 599
 600 for f in files:
 601         if f == '-':
 602                 f = ''
 603
 604         sys.stderr.write ('Processing `%s\'\n' % f)
 605
 606         e = Parser(f)
 607
 608         id = os.path.basename (f)
 609         id = re.sub ('[^a-zA-Z0-9]', 'x', id)
 610
 611         def num2let (match):
 612                 return chr (ord (match.group ()) - ord('0') + ord('A'))
 613
 614         id = re.sub ('[0-9]', num2let, id)
 615
 616         id = 'voice%s' % id
 617         ly =ly + '\n\n%s = \\context Staff = "%s" %s\n\n' % (id, id, e.dump ())
 618
 619         found_ids = found_ids + '\\%s\n' % id
 620
 621 found_ids = '\n\n\n\\score { << %s >> } ' % found_ids
 622
 623 ly_head = ''
 624 if ref_file:
 625         head = Ref_parser (ref_file)
 626         if not out_filename:
 627                 t = ''
 628                 st = ''
 629                 try:
 630                         t = head.dict['title']
 631                         st= head.dict['subtitle']
 632                 except KeyError:
 633                         pass
 634
 635                 t = t + '-' +st
 636
 637                 t = re.sub ("^ +(.*) +$", r"\1", t)
 638                 t = re.sub ("\\.", '', t)
 639                 out_filename = re.sub ('[^a-zA-Z0-9-]', '-', t)
 640                 out_filename = out_filename+ '.ly'
 641         ly_head = head.dump ()
 642
 643 if not out_filename:
 644         out_filename = 'musedata.ly'
 645
 646 sys.stderr.write ('Writing `%s\'\n' % out_filename)
 647
 648 fo = open (out_filename, 'w')
 649 fo.write ('%% lily was here -- automatically converted by musedata.ly\n')
 650 fo.write(ly_head + ly + found_ids)
 651 fo.close ()
 652