Audio analyzer

audio-analyzer.py

@@ -1,21 +1,64 @@
 import json
 import sys
-from collections import Counter
 
-from nltk import RegexpTokenizer
+from nltk import RegexpTokenizer, OrderedDict
 from nltk.stem.snowball import RussianStemmer
 
-counter = Counter()
+genres = {
+	1: "Rock",
+	2: "Pop",
+	3: "Rap & Hip - Hop",
+	4: "Easy Listening",
+	5: "Dance & House",
+	6: "Instrumental",
+	7: "Metal",
+	21: "Alternative",
+	8: "Dubstep",
+	9: "Jazz & Blues",
+	10: "Drum & Bass",
+	11: "Trance",
+	12: "Chanson",
+	13: "Ethnic",
+	14: "Acoustic & Vocal",
+	15: "Reggae",
+	16: "Classical",
+	17: "Indie Pop",
+	19: "Speech",
+	22: "Electropop & Disco",
+	18: "Other"
+}
+
+
+def dictWithoutOneKey(d, key):
+	new_d = d.copy()
+	new_d.pop(key)
+	return new_d
+
+
+audioStats = dict()
+
 tokenizer = RegexpTokenizer(r"[A-Za-zА-Яа-я]+")
 stemmer = RussianStemmer()
 
-musicFileName = sys.argv[0]
+musicFileName = sys.argv[1]
-with open(musicFileName, "r") as file:
+with open(musicFileName, "r", encoding="utf8") as file:
 	for line in file:
 		jsonData = json.loads(line, encoding="utf8")
-		for song in jsonData.values()[0]:
+		for song in list(jsonData.values())[0]:
-			key = "".join([stemmer.stem(token).lower() for token in tokenizer.tokenize("{} {}".format(song["artist"], song["title"]))])
+			songName = "{} - {}".format(song["artist"], song["title"])
-			counter[key] += 1
+			filteredSongName = "".join([stemmer.stem(token).lower() for token in tokenizer.tokenize(songName)])
+			if len(filteredSongName) > 1:
+				audioStatsItem = audioStats.get(filteredSongName, {
+					"name": songName,
+					"url": song["url"],
+					"genre": genres.get(song["genre_id"], "Other"),
+					"count": 0
+				})
+				audioStatsItem["count"] += 1
+				audioStats[filteredSongName] = audioStatsItem
 
-for item in counter.most_common():
+destFileName = sys.argv[2]
-	print(item)
+with open(destFileName, "w", encoding="utf8") as file:
+	sortedSongs = [item[1] for item in sorted(audioStats.items(), key=lambda item: item[1]["count"], reverse=True)]
+	data = OrderedDict([(item["name"], dictWithoutOneKey(item, "name")) for item in sortedSongs])
+	file.write(json.dumps(data, ensure_ascii=False, indent=4))