#!/usr/bin/python # Copyright (c) 2004 Danilo Segan . # # This file is part of Philips HDD100 Toolkit for Linux. # # This is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or # (at your option) any later version. # # It is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with this; if not, write to the Free Software Foundation, Inc., # 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA # Danilo did the programming, but it was simple thanks to the excellent # documentation over at # # http://solonweb.free.fr/tech.htm # import os, ID3, zlib, struct, sys recordsize = 408 field_offsets = { "hidden" : 0, # 0/1 "reserved1" : 4, # 0 "reserved2" : 8, # 0 "path" : 12, "name" : 68, "bitrate" : 100, # bits per second "samplerate" : 104, # bits per second "duration" : 108, # seconds "artist" : 112, "album" : 192, "genre" : 272, "title" : 312, "tracknumber" : 392, # track number "year" : 396, "size" : 400, # file size in bytes "source" : 404 # reserved, 0 } def get_files(musicdir, files): """Find all the files recursively in directory 'musicdir'. Also modifies record offsets if necessary.""" import mp3 maxpath = 0 maxname = 0 global field_offsets for file in os.listdir(musicdir): fullname = os.path.join(musicdir, file) if os.path.isdir(fullname): get_files(fullname,files) elif fullname[-3:].lower() == 'mp3': header = mp3.detect_mp3(fullname) if not header: print "Problem with %s: doesn't look like MP3 file" % (fullname) #header = { "bitrate" : 128*1000, "sampling_frequency" : 44100, "time" : "180" } continue record = { "file" : fullname, "path" : os.path.dirname(fullname), "name" : os.path.basename(fullname), "bitrate" : header["bitrate"]*1000, "sampling_frequency" : header["sampling_frequency"], "duration" : header["time"] } try: id3info = ID3.ID3(fullname) for key in id3info.keys(): record[key.lower()] = id3info[key] except: pass if 'title' not in record.keys(): (record['title'], ext) = os.path.splitext(record['name']) for key in ['path', 'name', 'album', 'artist', 'title', 'genre']: if not key in record.keys(): record[key] = u"unknown" for key in ['year', 'tracknumber', 'hidden', 'source']: if not key in record.keys(): record[key] = 0 if len(record["name"]) > maxname: maxname = len(record["name"]) if len(record["path"]) > maxpath: maxpath = len(record["path"]) print record['file'] trackid = len(files) record['trackid'] = trackid files.append(record) # #print "Adjusting path length to %d." % maxpath # ln = field_offsets["name"] - field_offsets["path"] # newlen = maxpath * 2 # if newlen > ln: # diff = newlen - ln # if diff % 4 == 2: # diff += 2 # field_offsets["name"] = field_offsets["name"] + diff # else: # diff = 0 # #print "Adjusting name length to %d." % maxname # ln = field_offsets["bitrate"] - field_offsets["name"] # newlen = maxname * 2 # if newlen > ln: # diff = newlen - ln + diff # if diff % 4 == 2: # diff += 2 # for field in ['bitrate', 'samplerate', 'duration', 'artist', 'album', # 'genre', 'title', 'tracknumber', 'year', 'size', 'source']: # field_offsets[field] += diff # #print field_offsets # global recordsize # recordsize += diff def field_format(): format = "III" # hidden, reserved1, reserved2 # file path len = field_offsets["name"] - field_offsets["path"] format += str(len) + "s" # file name len = field_offsets["bitrate"] - field_offsets["name"] format += str(len) + "s" format += "III" # bitrate, samplerate, duration # artist len = field_offsets["album"] - field_offsets["artist"] format += str(len) + "s" # album len = field_offsets["genre"] - field_offsets["album"] format += str(len) + "s" # genre len = field_offsets["title"] - field_offsets["genre"] format += str(len) + "s" # title len = field_offsets["tracknumber"] - field_offsets["title"] format += str(len) + "s" format += "IIII" # track number, year, size (bytes), source return format def write_dat(files): """Writes DB5000.DAT file listing all the MP3 files.""" #try: f = open(os.path.join(datadir, "db5000.dat"), "w") for file in files: format = field_format() path = file['path'] + "\\" path = path.replace("/", "\\") try: data = struct.pack( format, 0, # hidden 0, # reserved 1 0, # reserved 2 path.encode('UTF-16')[2:], # note the [2:] part here, in order to remove BOM file['name'].encode('UTF-16')[2:], int(file['bitrate']), # bitrate int(file['sampling_frequency']), # samplerate int(file['duration']), # duration file['artist'].encode('UTF-16')[2:], file['album'].encode('UTF-16')[2:], file['genre'].encode('UTF-16')[2:], file['title'].encode('UTF-16')[2:], int(file['tracknumber']), # tracknumber 0, #int(file['year']), # year # FIXME os.path.getsize(file['file']), # file size int(0) # source ) f.write(data) except: print "Problem with encoding data for file %s." % file['name'] del(f[data['trackid']]) f.close() #except: # print "Cannot open datadir for writing (%s)." % datadir def printout(files): """Debugging function, print out DAT files.""" global recordsize print recordsize for file in files: f = open(file, "r+") format = field_format() trackid = 0 data = f.read(recordsize) while data: tuple = struct.unpack(format, data) (hidden, res1, res2, path, name, bitrate, samplerate, duration, artist, album, genre, title, tracknumber, year, size, source ) = tuple print "trackid: %d" % trackid print "hidden: %d" % hidden print "res1: %d" % res1 print "res2: %d" % res2 print "path: %s" % path print "name: %s" % name print "bitrate: %d" % bitrate print "samplerate: %d" % samplerate print "duration: %d" % duration print "artist: %s" % artist print "album: %s" % album print "genre: %s" % genre print "title: %s" % title print "track: %d" % tracknumber print "year: %d" % year print "size: %d" % size print "source: %d" % source data = f.read(recordsize) trackid += 1 f.close() def sort_table_by_field(dict, field): keys = {} for elemid in range(0, len(dict)): #print elemid #print dict[elemid]['artist'] #dict[elemid]['trackid'] = elemid if field == 'artist': keys[dict[elemid][field]] = elemid; else: keys[dict[elemid][field]] = elemid; skeys = dict if field=='hidden' or field=='tracknumber' or field=='source': skeys.sort(lambda m,n: cmp(m[field], n[field]) or cmp(n['trackid'], m['trackid'])) else: skeys.sort(lambda m,n: cmp(unicode(str(m[field]), "ISO-8859-1"), unicode(str(n[field]),"ISO-8859-1")) or cmp(n['trackid'], m['trackid'])) return skeys def write_index(files, filename, field): """Create different .IDX files.""" print "Sorting by field '%s'..." % (field) sortedfiles = sort_table_by_field(files, field) id = 0 f = open(filename, "wb") for file in sortedfiles: fileid = file['trackid'] #print "%d: %s" % (fileid, file[field]) try: data = struct.pack("III", 0, int(fileid), zlib.crc32(str(file[field]))) except: data = struct.pack("III", 0, int(fileid), zlib.crc32(str(file['file']))) f.write(data) f.close() files = [] if len(sys.argv) < 3: print "Syntax:\n\t%s [DATADIR] [MUSICDIR] ...\n\n\tDATA dir is where DATA files will be stored.\n\tMUSICDIR is directory with music files, you can have more than one.\n" % sys.argv[0] sys.exit(1) datadir = sys.argv[1] musicdirs = sys.argv[2:] for musicdir in musicdirs: get_files(musicdir, files) # debugging #print files write_dat(files) # fixup track id numbers num = 0 for file in files: file['trackid'] = num #print "%d. %s" % (num, file['title']) num += 1 indexes = { "@DEV" : "hidden", "FNAM" : "name", "FPTH" : "path", "TALB" : "album", "TCON" : "genre", "TIT2" : "title", "TPE1" : "artist", "TRCK" : "tracknumber", "XSRC" : "source" } for name in indexes: filename = os.path.join(datadir, "DB5000_" + name + ".IDX") write_index(files, filename, indexes[name]) count = len(files) print "%d files added." % count f = open(os.path.join(datadir, 'db5000.hdr'), "r+") f.seek(1036) f.write(struct.pack("I", recordsize)) f.seek(1040) f.write(struct.pack("I", count)) f.seek(1604) f.write(struct.pack("I", int((field_offsets["name"] - field_offsets["path"])/2))) f.seek(2148) f.write(struct.pack("I", int((field_offsets["bitrate"] - field_offsets["name"])/2))) #print recordsize #print field_offsets offs = struct.pack("14I", field_offsets["reserved2"], field_offsets["path"], field_offsets["name"], field_offsets["bitrate"], field_offsets["samplerate"], field_offsets["duration"], field_offsets["artist"], field_offsets["album"], field_offsets["genre"], field_offsets["title"], field_offsets["tracknumber"], field_offsets["year"], field_offsets["size"], field_offsets["source"]) f.seek(12964) f.write(offs) # debugging, print out constructed DAT file #printout([os.path.join(datadir,"db5000.dat")]) #printout(["/home/danilo/rad/philips-hdd/data-test/db5000.dat" ])