working on playlist views

src/db.py

@@ -1,6 +1,10 @@
 import os
+from sys import getsizeof
 import json
 from pymongo import MongoClient
+import pandas as pd
+import numpy as np
+from timeit import default_timer as timer
 
 from dotenv import load_dotenv
 load_dotenv()
@@ -71,6 +75,48 @@ def write_run_record(self, run_record):
  self.runs.insert_one(run_record)
 
  # Playlist
+ def get_playlists(self, name=False):
+ """
+ Returns all playlist ids in stormdb as a list, or as their names if you'd rather
+ """
+ q = {}
+ cols = {"_id":1, "info":1}
+ r = list(self.playlists.find(q, cols))
+
+ if name:
+ return [x["info"]["name"] for x in r]
+ else:
+ return [x["_id"] for x in r]
+
+ def get_playlist_current_info(self, playlist_id):
+ """
+ Returns a playlists full record excluding changelog
+ """
+ q = {"_id":playlist_id}
+ cols = {"changelog":0}
+ r = list(self.playlists.find(q, cols))
+
+ if len(r) == 0:
+ raise Exception(f"{playlist_id} not found.")
+ else:
+ return r[0]
+
+ def get_playlist_changelog(self, playlist_id):
+ """
+ Returns a playlists changelog, a dictionary where each entry is a date.
+ """
+ q = {"_id":playlist_id}
+ cols = {"changelog":1}
+ r = list(self.playlists.find(q, cols))
+
+ if len(r) == 0:
+ raise Exception(f"{playlist_id} not found.")
+ else:
+ if 'changelog' in r[0].keys():
+ return r[0]['changelog']
+ else:
+ raise Exception(f"No changelog found for {playlist_id}, has it been collected more than once?")
+
  def get_playlist_collection_date(self, playlist_id):
  """
  Gets a playlists last collection date.
@@ -392,15 +438,72 @@ def update_artist_albums(self):
 
 class StormAnalyticsDB:
  """
- A StormDB wrapper dedicated to machine learning and general database analytics
+ A StormDB wrapper dedicated to machine learning and general database analytics.
+ Most data will get converted into plot friendly functions, like pandas dataframes.
  """
 
- def __init__(self):
+ def __init__(self, verbose=True):
 
  self.sdb = StormDB()
  #self.sql_db
 
- def gen_playlist_health(self, playlist_id):
+ self.map = {'playlist_track_changes':self.gen_v_playlist_track_changes,
+ 'many_playlist_track_changes':self.gen_v_many_playlist_track_changes}
+ self.print = print if verbose else lambda x: None
+
+ # Get views from StormDB
+ def gen_view(self, name, view_params={}):
+ """
+ Caller function for views (prints and other nice additions)
+ """
+ if name in self.map.keys():
+ self.print(f"Generating View: {name}")
+ self.print(f"Supplied Parameters: {view_params}")
+
+ start = timer()
+ r = self.map[name](**view_params)
+ end = timer()
+
+ self.print("View Complete!")
+ self.print(f"Elapsed Time to Build: {round(end-start, 4)} ms. | File Size: {getsizeof(r)} bytes")
+
+ return r
+
+ else:
+ raise Exception(f"View {name} not in map.")
+
+ def gen_v_many_playlist_track_changes(self, playlist_ids=[], metric='Number of Tracks'):
+ """
+ Cross-Compares many playlist track changes
+ """
+ df = pd.DataFrame()
+
+ if len(playlist_ids) == 0:
+ self.print("No playlists specified, returning all.")
+
+
+ #for playlist_id in playlist_ids:
+
+
+
+ # Single object views - low-level
+ def gen_v_playlist_track_changes(self, playlist_id):
+ """
+ Generates a view of a playlists timely health
+ """
+
+ #playlist_info = self.sdb.get_playlist_current_info()
+ playlist_changelog = self.sdb.get_playlist_changelog(playlist_id)
+
+ # Create Dataframe
+ df = pd.DataFrame(index=list(playlist_changelog.keys()))
+
+ # Compute Metrics
+ for change in playlist_changelog:
+ df.loc[change, 'Number of tracks'] = len(playlist_changelog[change]['tracks'])
+
+ return df
+
+
 
-