Scrape tracks

Some scrolling is still needed to actually get the whole track list

scrape.py

@@ -47,6 +47,19 @@ class User:
     following_count = None
     follower_count = None
 
+class Track:
+    title = None
+    artist = None
+    date = None
+    tag = None # We'll only use the first tag for now, saves opening each track
+
+    plays = None
+    likes = None
+    reposts = None
+    comments = None
+
+    tracks = {}
+
 # Gets a user's username, track count, following count, follower count, and like count
 def Get_Basic_Info(user):
     # Get the user's profile page and soup it
@@ -81,6 +94,64 @@ def Get_Basic_Info(user):
         user.like_count = like_count_elem.string.split(' ')[0]
     print('  Likes: ' + str(user.like_count))
 
+def Get_Tracks_Info(user):
+    # Get the user's tracks page and soup it
+    print('GET: https://soundcloud.com/' + user.url_username + '/tracks')
+    driver.get('https://soundcloud.com/' + user.url_username + '/tracks')
+    soup = BeautifulSoup(driver.page_source, 'lxml')
+
+    tracks = {}
+
+    track_elems = soup.find_all(class_='soundList__item')
+    for track_elem in track_elems:
+        # Initialise a track object to hold our data
+        track = Track()
+
+        # Grab the title
+        title_elem = soup.find(class_='soundTitle__title')
+        if title_elem:
+            track.title = title_elem.span.string.strip()
+
+        # Grab who uploaded it (usually the artist unless it's a label)
+        artist_elem = soup.find(class_='soundTitle__usernameText')
+        if artist_elem:
+            track.artist = artist_elem.string.strip()
+
+        # Check if we have already gotten this track before
+        if track.artist + track.title in tracks.keys():
+            continue
+
+        # Grab when it was uploaded
+        date_elem = soup.find(class_='soundTitle__uploadTime')
+        if date_elem:
+            track.date = date_elem.time['datetime']
+
+        # Grab the first tag featured on the list item
+        tag_elem = soup.find(class_='soundTitle__tagContent')
+        if tag_elem:
+            track.tag = tag_elem.string
+
+        # Grab the play and comment counts (either of these may or may not be present)
+        play_comment_elems = soup.find_all(class_='sc-ministats-item')
+        if play_comment_elems:
+            for elem in play_comment_elems:
+                num, unit = elem['title'].split(' ')
+                if unit == 'plays':
+                    track.plays = num
+                elif unit == 'comments':
+                    track.comments = num
+
+        # Grab the like count
+        track.likes = soup.find(class_='sc-button-like').string.strip()
+
+        # Grab the repost count
+        track.reposts = soup.find(class_='sc-button-repost').string.strip()
+
+        # Finally add the track to the dictionary
+        tracks[track.artist + track.title] = track
+
+    user.tracks = tracks
+
 user_dict = {}
 users_to_process = ['thomotron', 'lacheque', 'slynk', 'bossfightswe']
 
@@ -100,6 +171,9 @@ for user_str in users_to_process:
     # Get their basic info
     Get_Basic_Info(user)
 
+    # Get their tracks
+    Get_Tracks_Info(user)
+
     # Finally add the user to the dictionary
     user_dict[user.url_username] = user