def get_playlist(self):
# Retrieve HTML source if it has not been retrieved already
if not self.html_src:
source = self.retrieve_html_source()
else:
source = self.html_src
#split to find the playlist name
name_source = source.split(r'<h1 class="main">')[1]
name_source = name_source.split('</span>')[0]
playlist_name = re.findall(r'\">(.*)</a>', name_source)[0]
# Remove everything before the playlist section
songs_source = source.split("<tbody data-bind=\"foreach: tracks\"")[1]
# Divide up into songs
songs = songs_source.split("</tr>")
# Create a array of dictionaries of all the songs
songs_dict = []
for song in songs:
try:
title = re.findall(r'<td.*>(.*)<\/div>', song, re.S)[0]
artist = re.findall(r'spotify:artist:.*>(.*)<\/a>', song)[0]
album = re.findall(r'spotify:album.*>(.*)<\/a>', song)[0]
song_time = re.findall(r'tl-time\">([\w|:]*)<\/td>', song,
re.S)[0]
title = re.sub(r" - \w* Edit", "", title, re.IGNORECASE)
title = re.sub(r" -.*Version.*", "", title, re.IGNORECASE)
title = re.sub(r" -.*Remaster(ed)?.*", "", title,
re.IGNORECASE)
title = re.sub(r" \(Remaster(ed)?\) *", "", title,
re.IGNORECASE)
title = re.sub(r" -.*Anniversary Mix.*", "", title,
re.IGNORECASE)
song_dict = {
'title': Util.html_to_ascii(title),
'artist': Util.html_to_ascii(artist),
'album': Util.html_to_ascii(album),
'time': Util.time_in_seconds(song_time),
}
songs_dict.append(song_dict)
except IndexError:
pass
return [playlist_name, songs_dict]
def _get_search_info(self, song_search_url):
"""
Downloads the page source of the song_search_url, and returns a list of dictionaries containing
the information for each search result. The dictionaries contain 'title', 'url', and 'time' (in seconds) fields.
:param song_search_url: The url of a search for a song
:return: A list of dictionaries, each containing the 'title', 'url', and 'time' (in seconds) info of each search result
"""
with urllib.request.urlopen(song_search_url) as response:
html = response.read()
# decodes html source from binary bytes to string
search_source = html.decode("UTF-8", "ignore")
# parse source for vid info
search_info = []
# Isolate the list of results in the source
results_source = re.split(
r"<ol id=\"item-section-.*?\" class=\"item-section\">",
search_source)[1]
results_source = re.split(
r"<div class=\"branded-page-box search-pager.*\">", results_source,
1)[0]
# split by video in list, returns the type of entry (video, playlist, channel)
results_source = re.split(
r"<li><div class=\"yt-lockup yt-lockup-tile yt-lockup-(.*?) vve-check clearfix.*?\"",
results_source)[1:]
index = 0
while len(search_info) < self.MAX_NUM_SEARCH_RESULTS and index < len(
results_source) - 1:
source_type = results_source[index]
source = results_source[index + 1]
if source_type == "video":
video_url = re.findall(r"href=\"\/watch\?v=(.*?)\"", source)[0]
video_url = self.SONG_URL_RESULT_ROOT + video_url
video_title = re.findall(r"title=\"(.*?)\"", source)[2]
video_title = Util.html_to_ascii(video_title)
video_time = re.findall(r"Duration: (\d+:\d+)", source)[0]
video_time = re.split(r":", video_time)
video_time = int(video_time[0]) * 60 + int(video_time[1])
search_info.append({
"url": video_url,
"title": video_title,
"time": video_time
})
index += 2
return search_info
def _get_search_info(self, song_search_url):
"""
Downloads the page source of the song_search_url, and returns a list of dictionaries containing
the information for each search result. The dictionaries contain 'title', 'url', and 'time' (in seconds) fields.
:param song_search_url: The url of a search for a song
:return: A list of dictionaries, each containing the 'title', 'url', and 'time' (in seconds) info of each search result
"""
with urllib.request.urlopen(song_search_url) as response:
html = response.read()
# decodes html source from binary bytes to string
search_source = html.decode("UTF-8", "ignore")
# parse source for vid info
search_info = []
# Isolate the list of results in the source
results_source = re.split(r"<div class=\"searchResultGroupHeading\">",
search_source)[1]
results_source = re.split(r"</ul>", results_source, 1)[0]
# split by search result
results_source = re.split(r"<div class=\"searchItem\">",
results_source)[1:]
# This code theoretically works, but urllib can't access all of Soundclouds
# websource because it thinks it's an invalid browser of something
index = 0
while len(search_info) < self.MAX_NUM_SEARCH_RESULTS and index < len(
results_source):
source = results_source[index]
artist = re.findall(
r"<span class=\"soundTitle_+usernameText\">(.*)</span>",
source)[0]
title = re.findall(r"<span class=\"\">(.*)</span>", source)[0]
url = re.findall(
r"<a class=\"soundTitle_+title sc-link-dark\" href=\"(.*)\">",
source)[0]
title = Util.html_to_ascii(artist + " " + title)
url = self.SONG_URL_RESULT_ROOT + url
search_info.append({"url": url, "title": title, "time": None})
index += 1
return search_info