1 | #!/usr/bin/env python
|
---|
2 | #
|
---|
3 |
|
---|
4 | import os
|
---|
5 | import os.path
|
---|
6 | import datetime
|
---|
7 | import string
|
---|
8 | import optparse
|
---|
9 | import exceptions
|
---|
10 | import re
|
---|
11 |
|
---|
12 | import eyeD3
|
---|
13 | import eyeD3.tag
|
---|
14 | import eyeD3.frames
|
---|
15 | import eyeD3.utils
|
---|
16 |
|
---|
17 | # We are not running inside the django framework so we need to tell it where
|
---|
18 | # to find our django app framework.
|
---|
19 | #
|
---|
20 | os.environ["DJANGO_SETTINGS_MODULE"] = "mediaserv.settings.main"
|
---|
21 |
|
---|
22 | from django.models.music import *
|
---|
23 |
|
---|
24 | """This program is intended to be invoked periodically on the system hosting
|
---|
25 | the mediaserv app.
|
---|
26 |
|
---|
27 | It will load up the mediaserv models.
|
---|
28 |
|
---|
29 | It will then iterate through the defined MusicRoot's. For each MusicRoot it
|
---|
30 | will scan all the files in it and all of its sub-directories.
|
---|
31 |
|
---|
32 | For every file that is an mp3 it will attempt to parse the ID3 tags for that
|
---|
33 | file and based on those tags it will create Artist, ArtistName, Album, and
|
---|
34 | Track records in our database.
|
---|
35 |
|
---|
36 | For every track that we encounter we will update the 'last_scanned' field.
|
---|
37 |
|
---|
38 | After we finish scanning a MusicRoot we ask for all of the tracks whose
|
---|
39 | last_scanned field is before its MusicRoot's last_scan_started date. We will
|
---|
40 | then remove those tracks from the system.
|
---|
41 | """
|
---|
42 |
|
---|
43 | quote_pattern = re.compile("'")
|
---|
44 |
|
---|
45 | ############################################################################
|
---|
46 | #
|
---|
47 | def q( string ):
|
---|
48 | """Simple function to take the given string and append single quotes ( ' )
|
---|
49 | around it. it will also go through the string first and double any single
|
---|
50 | quotes in it so that they are SQL safe.
|
---|
51 | """
|
---|
52 |
|
---|
53 | return "'%s'" % quote_pattern.sub("''", string)
|
---|
54 |
|
---|
55 | ############################################################################
|
---|
56 | #
|
---|
57 | def is_simple_charset( value ):
|
---|
58 | """Given a string return True if it only contains printable ascii
|
---|
59 | characters, otherwise return false.
|
---|
60 |
|
---|
61 | This is our simple test to see if a string is a fancy UTF-8 one or
|
---|
62 | something else, or is a simple plain ASCII string.
|
---|
63 | """
|
---|
64 | for char in value:
|
---|
65 | if char not in string.printable:
|
---|
66 | return False
|
---|
67 | return True
|
---|
68 |
|
---|
69 | ############################################################################
|
---|
70 | #
|
---|
71 | def create_artist(id3_artist):
|
---|
72 | """We are called with the name of an artist. We create a new Arist object
|
---|
73 | and ArtistName and tie them together.
|
---|
74 |
|
---|
75 | If the ArtistName is SJIS or has 8bit characters then we also flag this
|
---|
76 | name as not having a simple character set.
|
---|
77 | """
|
---|
78 |
|
---|
79 | art = artists.Artist(date_added = datetime.datetime.now())
|
---|
80 | art.save()
|
---|
81 | art.add_artistname("'%s'" % id3_artist, preference = 0,
|
---|
82 | simple_char_set = is_simple_charset(id3_artist))
|
---|
83 | return art
|
---|
84 |
|
---|
85 | ############################################################################
|
---|
86 | #
|
---|
87 | def create_album(id3_album):
|
---|
88 | """Create a new album object with the given name.
|
---|
89 | """
|
---|
90 | alb = albums.Album(name = "'%s'" % id3_album,
|
---|
91 | date_added = datetime.datetime.now())
|
---|
92 | alb.save()
|
---|
93 | return alb
|
---|
94 |
|
---|
95 | ############################################################################
|
---|
96 | #
|
---|
97 | def add_file(filename, af, music_root, verbosity = "quiet"):
|
---|
98 | """We have a file. We know it is an audio/mpeg file. We know we do not have
|
---|
99 | this file in our track db. Query the file for its info and add the
|
---|
100 | requisite objects to our db.
|
---|
101 | """
|
---|
102 |
|
---|
103 | tag = af.getTag()
|
---|
104 |
|
---|
105 | playtime = af.getPlayTime()
|
---|
106 | (vbr, bitrate) = af.getBitRate()
|
---|
107 | if vbr == 0:
|
---|
108 | vbr = False
|
---|
109 | else:
|
---|
110 | vbr = True
|
---|
111 |
|
---|
112 | # Let us pull out all the id3 tags we have, filling in default info
|
---|
113 | #
|
---|
114 | if tag is None:
|
---|
115 | # The file has no id3 info. We create a title based on the file name
|
---|
116 | # and everything else is empty.
|
---|
117 | #
|
---|
118 | track = tracks.Track(title = "'%s'" % os.path.basename(filename),
|
---|
119 | filename = filename,
|
---|
120 | last_scanned = datetime.datetime.now(),
|
---|
121 | play_time = playtime, bit_rate = bitrate,
|
---|
122 | vbr = vbr, musicroot_id = music_root.id)
|
---|
123 | track.save()
|
---|
124 | return
|
---|
125 |
|
---|
126 |
|
---|
127 | # Get the title, if it does not exist, use the file name.
|
---|
128 | #
|
---|
129 | title = tag.getTitle()
|
---|
130 | if not title:
|
---|
131 | title = filename
|
---|
132 |
|
---|
133 | # At this point we have the information to create the basic track object.
|
---|
134 | # The rest of the fields are optional.
|
---|
135 | #
|
---|
136 | track = tracks.Track(title = q(title), filename = filename,
|
---|
137 | last_scanned = datetime.datetime.now(),
|
---|
138 | play_time = playtime, bit_rate = bitrate, vbr = vbr,
|
---|
139 | musicroot_id = music_root.id)
|
---|
140 |
|
---|
141 | # Get the artist tag. See if we have an artist with this exact name.
|
---|
142 | # If we do not, then create a new artist.
|
---|
143 | #
|
---|
144 | id3_artist = tag.getArtist()
|
---|
145 | if id3_artist:
|
---|
146 | try:
|
---|
147 | artist = artistnames.get_object(name__exact = \
|
---|
148 | "'%s'" % id3_artist).get_artist()
|
---|
149 | except artistnames.ArtistNameDoesNotExist:
|
---|
150 | artist = create_artist(id3_artist)
|
---|
151 | track.artist_id = artist.id
|
---|
152 |
|
---|
153 | # See if we can get the track number & disc number (note: getTrackNum() &
|
---|
154 | # getDiscNum() return a tuple (track num, total tracks) (disc num, total
|
---|
155 | # discs) so we only one element 0 of the tuple.
|
---|
156 | #
|
---|
157 | id3_tracknum = tag.getTrackNum()[0]
|
---|
158 | if id3_tracknum:
|
---|
159 | track.track_number = id3_tracknum
|
---|
160 | id3_discnum = tag.getDiscNum()[0]
|
---|
161 | if id3_discnum:
|
---|
162 | track.disc_number = id3_discnum
|
---|
163 |
|
---|
164 | # The album is like the artist. We see if we can find an album that exists
|
---|
165 | # with the exact name. If we can then we use it. If we can not then we
|
---|
166 | # create a new album and use that.
|
---|
167 | #
|
---|
168 | id3_album = tag.getAlbum()
|
---|
169 | if id3_album:
|
---|
170 | try:
|
---|
171 | album = albums.get_object(name__exact = "'%s'" % id3_album)
|
---|
172 | except albums.AlbumDoesNotExist:
|
---|
173 | album = create_album(id3_album)
|
---|
174 | track.album_id = album.id
|
---|
175 |
|
---|
176 | # Going to skip genre for now. Just really do not care much about it.
|
---|
177 | #
|
---|
178 | track.save()
|
---|
179 |
|
---|
180 | # Depending on the verbosity print out info on the track we just scanned.
|
---|
181 | #
|
---|
182 | if verbosity == "terse":
|
---|
183 | print "Added track: %s" % track
|
---|
184 | elif verbose == "verbose":
|
---|
185 | print "Added track: %s .. (and other info)" % track
|
---|
186 |
|
---|
187 | print "Track %s, last scanned: %s" % (track, track.last_scanned)
|
---|
188 | return
|
---|
189 |
|
---|
190 | ############################################################################
|
---|
191 | #
|
---|
192 | #
|
---|
193 | def scan_file(filename, music_root, verbosity = "quiet"):
|
---|
194 | """This function is given a file name as an actual absolute file path.
|
---|
195 | We will now use the eyeD3 library to scan any id3 tags it may have and
|
---|
196 | based on the content of those tags create tracks, artists, artistnames, and
|
---|
197 | albums in our database.
|
---|
198 |
|
---|
199 | If a track object already exists for this file we will see if any of the
|
---|
200 | records in our database need to be updated (and update them.)
|
---|
201 | """
|
---|
202 |
|
---|
203 | # If the file is not an mp3 file we just return.
|
---|
204 | #
|
---|
205 | if not eyeD3.tag.isMp3File(filename):
|
---|
206 | if verbosity == "verbose":
|
---|
207 | print "Skipping file %s (not an audio/mpeg file)" % filename
|
---|
208 | return
|
---|
209 |
|
---|
210 | if verbosity == "verbose":
|
---|
211 | print " Scanning file %s" % filename
|
---|
212 |
|
---|
213 | try:
|
---|
214 | af = eyeD3.tag.Mp3AudioFile(filename)
|
---|
215 | except Exception, e:
|
---|
216 | print "Unable to parse file: %s" % filename
|
---|
217 | return
|
---|
218 | tag = af.getTag()
|
---|
219 | if tag is None:
|
---|
220 | if verbosity == "verbose":
|
---|
221 | print "File %s had no id3 tag information. Filling in defaults" % \
|
---|
222 | filename
|
---|
223 |
|
---|
224 | # First see if a track already exists that refers to this exact same
|
---|
225 | # file. This is because files are the acutal item that identifies a
|
---|
226 | # track. If the file already exists then we already have this track in our
|
---|
227 | # db. We just need to make sure that all the fields we have in the db match
|
---|
228 | # the ones in this file.
|
---|
229 | #
|
---|
230 | try:
|
---|
231 | track = tracks.get_object(filename__exact = filename)
|
---|
232 | except tracks.TrackDoesNotExist:
|
---|
233 | add_file(filename, af, music_root, verbosity)
|
---|
234 | return
|
---|
235 |
|
---|
236 | # This track already existed in our db. Check to see if any of its id3 tags
|
---|
237 | # differ from what we already have in the db. If they do, update the db.
|
---|
238 | #
|
---|
239 | print "We would normually update track %s, but we are skipping it for " \
|
---|
240 | "now" % os.path.basename(filename)
|
---|
241 | track.last_scanned = datetime.datetime.now()
|
---|
242 | track.save()
|
---|
243 | #compare_update_file(filename, af, music_root, verbosity = verbosity)
|
---|
244 | return
|
---|
245 |
|
---|
246 | ############################################################################
|
---|
247 | #
|
---|
248 | #
|
---|
249 | def run(verbosity = "quiet"):
|
---|
250 | """This is the function that actually does the work of scanning all of our
|
---|
251 | MusicRoots for .mp3 files.
|
---|
252 |
|
---|
253 | It expects a single argument: a string that indicates the verbosity
|
---|
254 | level. This mean be either 'verbose', 'terse', or 'quiet.' If not specified
|
---|
255 | it will default to 'quiet.'
|
---|
256 | """
|
---|
257 |
|
---|
258 | # Get the list of defined MusicRoots. These had better point to real
|
---|
259 | # directories!
|
---|
260 | #
|
---|
261 | music_roots = musicroots.get_list()
|
---|
262 | for music_root in music_roots:
|
---|
263 |
|
---|
264 | # We first mark that we actually started to scan this music root.
|
---|
265 | #
|
---|
266 | music_root.last_scan_started = datetime.datetime.now()
|
---|
267 | music_root.save()
|
---|
268 |
|
---|
269 | # music_root = musicroots.get_object(pk = music_root.id)
|
---|
270 |
|
---|
271 | if verbosity == "verbose":
|
---|
272 | print "Started scanning MusicRoot %s at %s" % \
|
---|
273 | (music_root.directory, music_root.last_scan_started)
|
---|
274 |
|
---|
275 | # Then the magic walk happens
|
---|
276 | #
|
---|
277 | for root, dirs, files in os.walk(music_root.directory):
|
---|
278 | if verbosity == "verbose":
|
---|
279 | print "Scanning directory: %s" % root
|
---|
280 |
|
---|
281 | for f in files:
|
---|
282 | check_file = os.path.join(root, f)
|
---|
283 | scan_file(check_file, music_root, verbosity = verbosity)
|
---|
284 |
|
---|
285 | # Okay. Our magic walk happeend. Now we need to delete any tracks that
|
---|
286 | # had been a part of this music root but were not scanned in this run
|
---|
287 | #
|
---|
288 | missing_tracks = \
|
---|
289 | music_root.get_track_list(last_scanned__lt = \
|
---|
290 | music_root.last_scan_started)
|
---|
291 |
|
---|
292 | print "\n\n** Music root last scanned: %s" % music_root.last_scan_started
|
---|
293 | for track in missing_tracks:
|
---|
294 | # If it is a member of any playlists remove it..
|
---|
295 | #
|
---|
296 | print "Track %s last scanned: %s" % (track, track.last_scanned)
|
---|
297 | if verbosity == "verbose" or verbosity == "terse":
|
---|
298 | print "Track %s not found in scan. Deleting from MusicRoot " \
|
---|
299 | "%s" % (track, music_root)
|
---|
300 | track.set_playlists([])
|
---|
301 | track.delete()
|
---|
302 |
|
---|
303 | # Done scanning a music root. Indicate when we finished scanning it.
|
---|
304 | #
|
---|
305 | music_root.last_scan_finished = datetime.datetime.now()
|
---|
306 | music_root.save()
|
---|
307 |
|
---|
308 | return
|
---|
309 |
|
---|
310 | ############################################################################
|
---|
311 | #
|
---|
312 | #
|
---|
313 | def setup_option_parser():
|
---|
314 | """This function uses the python OptionParser module to define an option
|
---|
315 | parser for parsing the command line options for this script. This does not
|
---|
316 | actually parse the command line options. It returns the parser object that
|
---|
317 | can be used for parsing them.
|
---|
318 | """
|
---|
319 | parser = optparse.OptionParser(usage = "%prog [options]",
|
---|
320 | version = "%prog 1.0")
|
---|
321 | parser.add_option("-v", "--verbosity", type="choice", dest="verbosity",
|
---|
322 | default="terse", choices = ["verbose", "terse",
|
---|
323 | "quiet"],
|
---|
324 | help = """Controls how talkative the script is about what
|
---|
325 | it is doing. In 'verbose' mode it will tell you
|
---|
326 | every track it finds. In 'terse' mode it will only tell
|
---|
327 | you about tracks that are changed, added or removed.
|
---|
328 | In 'quiet' mode it will say nothing. DEFAULT:
|
---|
329 | '%default'""")
|
---|
330 | return parser
|
---|
331 |
|
---|
332 | ############################################################################
|
---|
333 | #
|
---|
334 | def main():
|
---|
335 | """The main routine. This is invoked if this file is run as a program
|
---|
336 | instead of being imported as a library.
|
---|
337 |
|
---|
338 | If you are running this as a module you should not invoke the 'main()'
|
---|
339 | function but should instead invoke the 'run()' function.
|
---|
340 | """
|
---|
341 |
|
---|
342 | parser = setup_option_parser()
|
---|
343 | (opts, args) = parser.parse_args()
|
---|
344 |
|
---|
345 | run(opts.verbosity)
|
---|
346 |
|
---|
347 | ###########
|
---|
348 | #
|
---|
349 | # The work starts here
|
---|
350 | #
|
---|
351 |
|
---|
352 | if __name__ == "__main__":
|
---|
353 | main()
|
---|
354 |
|
---|
355 | #
|
---|
356 | #
|
---|
357 | #
|
---|
358 | ###########
|
---|