def _get_things(self,
method,
thing,
thing_type,
params=None,
cacheable=True):
"""Returns a list of the most played thing_types by this thing, in a
tuple with the total number of pages of results. Includes an MBID, if
found.
"""
doc = self._request(self.ws_prefix + "." + method, cacheable, params)
toptracks_node = doc.getElementsByTagName('toptracks')[0]
total_pages = int(toptracks_node.getAttribute('totalPages'))
seq = []
for node in doc.getElementsByTagName(thing):
title = _extract(node, "name")
artist = _extract(node, "name", 1)
mbid = _extract(node, "mbid")
playcount = _number(_extract(node, "playcount"))
thing = thing_type(artist, title, self.network)
thing.mbid = mbid
seq.append(TopItem(thing, playcount))
return seq, total_pages
def crawl_tag(song_hash_file):
network = pylast.LastFMNetwork(api_key=API_KEY)
taghash = {}
tags = []
taginfo = []
with open(song_hash_file) as fin:
for line in fin:
[sid, title, artist] = line.strip().split('\t')
#if int(sid)>3:
# break
print line.strip()
title = title.replace('\/','/')
title = title.replace('w/ ','feat. ')
tags.append({})
track = network.get_track(artist, title)
try:
toptags = track.get_top_tags()
except:
if title.find('(feat.')!=-1:
track = network.get_track(artist, title[:title.find('(feat.')])
elif title.find('feat.')!=-1:
track = network.get_track(artist, title[:title.find('feat.')])
else:
print sid, ':', title, 'not found'
continue
try:
toptags = track.get_top_tags()
except:
print sid, ':', title, 'not found'
continue
for tag in toptags:
if tag.item.name is None or tag.weight is None:
continue
if tag.item.name not in taghash:
taghash[tag.item.name] = len(taghash)
tags[-1][taghash[tag.item.name]] = int(tag.weight)
errlist = []
for k in taghash:
print k
tag = network.get_tag(k)
taginfo.append({'tid':taghash[k], 'name':k})
try:
info = tag._request('tag.getInfo', True)
reach = pylast._extract(info, 'reach')
taggings = pylast._extract(info, 'taggings')
if reach is None or taggings is None:
errlist.append(k)
else:
taginfo[-1]['reach'] = reach
taginfo[-1]['taggings'] = taggings
except:
errlist.append(k)
print 'errlist',len(errlist),errlist
taginfo.sort(key=lambda k:k['tid'])
return tags, taginfo
def _extract_track_data(track: Node):
track_mbid = pylast._extract(track, "mbid")
track_title = pylast._extract(track, "name")
timestamp = dt.datetime.fromtimestamp(
int(track.getElementsByTagName("date")[0].getAttribute("uts")))
artist_name = pylast._extract(track, "artist")
artist_mbid = track.getElementsByTagName("artist")[0].getAttribute("mbid")
album_title = pylast._extract(track, "album")
album_mbid = track.getElementsByTagName("album")[0].getAttribute("mbid")
# TODO: could call track/album/artist.getInfo here, and get more info?
# Handle missing titles
if album_title is None:
album_title = "(unknown album)"
# If we don't have mbids, synthesize them
if not artist_mbid:
artist_mbid = "md5:" + hashlib.md5(
artist_name.encode("utf8")).hexdigest()
if not album_mbid:
h = hashlib.md5()
h.update(artist_mbid.encode("utf8"))
h.update(album_title.encode("utf8"))
album_mbid = "md5:" + h.hexdigest()
if not track_mbid:
h = hashlib.md5()
h.update(album_mbid.encode("utf8"))
h.update(track_title.encode("utf8"))
track_mbid = "md5:" + h.hexdigest()
return {
"artist": {
"id": artist_mbid,
"name": artist_name
},
"album": {
"id": album_mbid,
"title": album_title,
"artist_id": artist_mbid
},
"track": {
"id": track_mbid,
"album_id": album_mbid,
"title": track_title
},
"play": {
"track_id": track_mbid,
"timestamp": timestamp
},
}
def get_similar(self, limit=None):
if limit:
doc = self._request(self.ws_prefix + '.getSimilar', True, params={'artist': self.artist.name,
'track': self.title,
'limit': limit})
else:
doc = self._request(self.ws_prefix + '.getSimilar', True)
seq = []
for node in doc.getElementsByTagName(self.ws_prefix):
title = pylast._extract(node, 'name')
artist = pylast._extract(node, 'name', 1)
match = pylast._number(pylast._extract(node, "match"))
seq.append(pylast.SimilarItem(pylast.Track(artist, title, self.network), match))
return seq
def check_album(artist, album):
album_object = network.get_album(artist, album)
corrected_album_name = pylast._extract(album_object._request(album_object.ws_prefix + ".getCorrection"), "name")
if not corrected_album_name:
print('No suggested album.')
return album
if (album == corrected_album_name):
# No change, return immediately and don't waste server hits
return album
elif album == "" and not corrected_album_name:
# Blank album
return album
corrected_album_object = network.get_album(artist, corrected_album_name)
if (album_object != corrected_album_object):
response = input(f'Should album {album} be {corrected_album_name}? (y/N)').lower()
if (response == "y"):
album = corrected_album_name
elif (album != corrected_album_name):
print(f"Suggested album name correction {corrected_album_name} does not affect scrobble, recommend changing.")
response = input(f'Should album {album} be {corrected_album_name}? (Y/n)').lower()
if (response == "n"):
pass
else:
album = corrected_album_name
return album
def tag(self, irc, msg, args, tag):
"""<tag>
Returns some info on the tag <tag>."""
try:
summaryLength = 260
outputString = "%s%s%s%s"
summaryString = "\"%s\" "
artistsString = " Top artists: %s."
#tracksString = " Top tracks: %s."
#tagsString = " Similar tags: %s."
sString = "'s%s"
nothingString = " has no top artists."
urlString = " %s"
tagRef = network.get_tag(tag)
tag = tagRef.get_name(properly_capitalized = True)
tag = tag[0].upper() + tag[1:]
summary = pylast._extract(tagRef._request("tag.getInfo", True), "summary")
if summary != None and summary != "":
summary = re.sub("\<[^<]+\>", "", summary)
summary = re.sub("\s+", " ", summary)
summary = summary[:summaryLength] + "..." if (summary[:summaryLength] != summary) else summary
topArtists = tagRef.get_top_artists()
if len(topArtists) > 3:
topArtists = topArtists[:3]
topArtistsText = commaAndify([x.item.get_name() for x in topArtists])
if len(topArtistsText) > 0:
topArtistsText = artistsString % topArtistsText
#topTracks = tagRef.get_top_tracks()
#if len(topTracks) > 3:
# topTracks = topTracks[:3]
#topTracksText = commaAndify([x.item.get_artist().get_name()
# + " - " + x.item.get_name() for x in topTracks]).encode("utf-8")
#if len(topTracksText) > 0:
# topTracksText = tracksString % topTracksText
#similarTags = tagRef.get_similar()
#if len(similarTags) > 3:
# similarTags = similarTags[:3]
#similarTagsText = commaAndify([x.get_name() for x in similarTags])
#if len(similarTagsText) > 0:
# similarTagsText = tagsString % similarTagsText
data = sString % (topArtistsText)
if len(topArtistsText) is 0:
data = nothingString
else:
data = data[:3] + data[3].lower() + data[4:]
url = tagRef.get_url()
if len(url) > 0:
url = urlString % url
summary = summaryString % summary if (summary != None and summary != "") else ""
output = outputString % (summary, tag, data, url)
irc.reply(output.encode("utf-8"))
except pylast.WSError, e:
irc.error(str(e))
def get_child(cls, response, field):
""" Retrieves child element from response and casts it according
to field type in DB model.
Parameters
----------
response : XML string
Response from Last.fm
field : DB attribute.
Attribute object contains name, which matches child in XML
tree, and db_value, which casts string to appropriate type.
"""
return field.db_value(pylast._extract(response, field.name))
def _get_things(self, method, thing, thing_type, params=None, cacheable=True):
"""Returns a list of the most played thing_types by this thing, in a
tuple with the total number of pages of results. Includes an MBID, if
found.
"""
doc = self._request(self.ws_prefix + "." + method, cacheable, params)
toptracks_node = doc.getElementsByTagName("toptracks")[0]
total_pages = int(toptracks_node.getAttribute("totalPages"))
seq = []
for node in doc.getElementsByTagName(thing):
title = _extract(node, "name")
artist = _extract(node, "name", 1)
mbid = _extract(node, "mbid")
playcount = _number(_extract(node, "playcount"))
thing = thing_type(artist, title, self.network)
thing.mbid = mbid
seq.append(TopItem(thing, playcount))
return seq, total_pages
def scrape_artisttags(self, artist, doc = None):
""" Iterator over top tags associated with artist
Parameters
----------
artist : pylast.Artist
Artist object.
doc : xml
Answer to artist.getInfo request. Used to aviod
repeating request when underlying caching is disabled.
"""
if not doc:
doc = artist._request('artist.getInfo', True)
for tag in doc.getElementsByTagName("tag"):
yield pylast._extract(tag, "name")
