From 2c7417cb5aeea5daa669efe1d4ec2538ac19431e Mon Sep 17 00:00:00 2001
From: Xevion <xevioni@yandex.com>
Date: Sun, 27 Oct 2019 16:32:53 -0500
Subject: [PATCH] finish work for plotting with copy function

---
 process.py | 56 +++++++++++++++++++++++++++++++++++++++---------------
 1 file changed, 41 insertions(+), 15 deletions(-)

diff --git a/process.py b/process.py
index 2e78e30..6e19e68 100644
--- a/process.py
+++ b/process.py
@@ -5,11 +5,14 @@ import logging
 import collections
 import numpy as np
 import dateutil.parser
+import PIL.Image as Image
 import matplotlib.pyplot as plt
 
+# Gets all files in tracks folder, returns them in parsed JSON
 def get_files():
     folder = os.path.join(sys.path[0], 'tracks')
     files = []
+    print(os.listdir(folder))
     for file in os.listdir(folder):
         with open(os.path.join(os.path.join(folder, file))) as file:
             files.append(
@@ -17,12 +20,14 @@ def get_files():
             )
     return files
 
+# Simple function to combine a bunch of items from different files
 def combine_files(files):
     items = []
     for file in files:
         items.extend(file['items'])
     return items
 
+# Prints the data in a interesting format
 def print_data(data):
     for i, item in enumerate(data):
         date = dateutil.parser.parse(item['added_at'])
@@ -32,26 +37,45 @@ def print_data(data):
         print('[{}] {} "{}" by {}'.format(date, explicit, track_name, artists))
 
 def process_data(data):
-    explicit = collections.defaultdict(int)
-    safe = collections.defaultdict(int)
+    # Process the data by Month/Year, then by Safe/Explicit
+    scores = {}
     for item in data:
-        date = dateutil.parser.parse(item['added_at'])
-        date = date.strftime('%b %Y')
-        if item['track']['explicit']:
-            explicit[date] += 1
-        else:
-            safe[date] += 1
-    times = [dateutil.parser.parse(item['added_at']) for item in data]
-    sort_func = lambda x : dateutil.parser.parse(x[0]).epoch
+        date = dateutil.parser.parse(item['added_at']).strftime('%b %Y')
+        if date not in scores.keys():
+            scores[date] = [0, 0]
+        scores[date][1 if item['track']['explicit'] else 0] += 1
     
+    # Create simplified arrays for each piece of data
+    months = list(scores.keys())[::-1]
+    safe, explicit = [], []
+    for item in list(scores.values())[::-1]:
+        safe.append(item[0])
+        explicit.append(item[1])
 
-    n = len(data)
+    # Done processing date properly, start plotting work
+    n = len(scores.values())
     ind = np.arange(n)
-    width = 0.35
-    plt.ylabel('Month')
-    plt.xlabel('Song by Type')
-    plt.xticks(ind, )
+    width = 0.55
+
+    p1 = plt.bar(ind, explicit, width)
+    p2 = plt.bar(ind, safe, width, bottom=explicit) # bottom= just has the bar sit on top of the explicit
+
+    plt.title('Song by Safe/Explicit')
+    plt.ylabel('Song Count')
+    plt.xlabel('Month')
+    plt.xticks(ind, months, rotation=270) # Rotation 90 will have the 
+    plt.legend((p1[0], p2[0]), ('Explicit', 'Safe'))
     # plt.show()
+    copy(months, safe, explicit)
+
+# Simple method for exporting data to a table like format
+# Will paste into Excel very easily
+def copy(months, safe, explicit):
+    from pyperclip import copy
+    top = 'period\tsafe\texplicit\n'
+    copy(top + '\n'.join([
+        f'{item[0]}\t{item[1]}\t{item[2]}' for item in zip(months, safe, explicit)
+    ]))
 
 def main():
     logging.basicConfig(level=logging.INFO)
@@ -60,7 +84,9 @@ def main():
     logging.info(f"Read and parse {len(files)} track files")
     logging.info("Combining into single track file for ease of access")
     data = combine_files(files)
+    data.sort(key=lambda item : dateutil.parser.parse(item['added_at']).timestamp(), reverse=True)
     logging.info(f'File combined with {len(data)} items')
     logging.info('Processing file...')
     process_data(data)
+
 main()
\ No newline at end of file