From 60e1fb2e74daa4e947e8837d3d59119727c88384 Mon Sep 17 00:00:00 2001
From: "Maxence G. de Montauzan" <maxence@gdemontauzan.fr>
Date: Fri, 13 Aug 2021 02:30:39 +0200
Subject: [PATCH] (back) Suggester: Improve processor - more generic

Process album & artist with a calculated fields
Separate main
Show progress

(cherry picked from commit fc8407cc6a51fe18b14169b3a3f0e4fc363beb4f)
---
 suggester.py | 70 ++++++++++++++++++++++++++--------------------------
 1 file changed, 35 insertions(+), 35 deletions(-)

diff --git a/suggester.py b/suggester.py
index 3cd76e4..288b9f5 100644
--- a/suggester.py
+++ b/suggester.py
@@ -1,5 +1,6 @@
 import requests
 import json
+import sys
 
 ELS_URL ='http://localhost:9200'
 INDEX = 'itunes-suggest'
@@ -26,15 +27,15 @@ def get_tokens(data: str) -> list:
         raise NoGoodDataException('Data is not correct to get tokens')
     return [t['token'] for t in r.json()['tokens']]
 
-def post_document(artist: str = None, artist_sugget: list = None, album: str = None, album_suggest: list = None) -> bool:
+def post_document(name: str, input: list, field_name: str) -> bool:
+    suggest_name = field_name + '_suggest'
     element = {
-        "artist_suggest" : artist_sugget,
-        "artist": artist,
-        "album": album,
-        "album_suggest": album_suggest}
+        field_name: name,
+        suggest_name: input
+    }
 
     # Filter empty keys
-    element = {k: v for k, v in element.items() if v}
+    # element = {k: v for k, v in element.items() if v}
 
     url = '{}/{}/_doc'.format(ELS_URL, INDEX)
     resp = requests.post(url, json=element)
@@ -48,36 +49,35 @@ def post_document(artist: str = None, artist_sugget: list = None, album: str = N
     # print('Post_element - Element created: ' + el_id)
     return el_id
 
+def process_file(file_name: str, field_name: str) -> int:
+    print('Process file: ' + file_name)
+    with open(file_name, 'r') as o_file:
+            lines = o_file.readlines()
+
+    count = 0
+    i = 0
+    for line in lines:
+        i += 1
+        sys.stdout.write(str(int((i/len(lines))*100)) + '%')
+        sys.stdout.flush()
+        sys.stdout.write("\b" * (40+1)) # return to start of line, after '['
+        data = json.loads(line)
+        if "Artist" in data:
+            try :
+                input = get_tokens(data[field_name])
+                post_document(name=data[field_name], input=input, field_name=field_name.lower())
+                count += 1
+            except NoGoodDataException:
+                print('ERROR WITH DATA')
+                print(str(data))
+    print('File processed\n')
+
+    return count
+
+
 if __name__ == '__main__':
     # Using readlines()
-    with open('/home/budd/workspace/iTunes/es-artists.json', 'r') as artist_file:
-        artists_lines = artist_file.readlines()
-
-    with open('/home/budd/workspace/iTunes/es-albums.json', 'r') as artist_file:
-        albums_lines = artist_file.readlines()
-
-    # Strips the newline character
     count = 0
-    for line in artists_lines:
-        data = json.loads(line)
-        if "Artist" in data:
-            try :
-                artist_input = get_tokens(data['Artist'])
-                post_document(artist=data['Artist'], artist_sugget=artist_input)
-                count += 1
-            except NoGoodDataException:
-                print('ERROR WITH DATA')
-                print(str(data))
-
-    for line in albums_lines:
-        data = json.loads(line)
-        if "Artist" in data:
-            try :
-                album_input = get_tokens(data['Album'])
-                post_document(album=data['Album'], album_suggest=album_input)
-                count += 1
-            except NoGoodDataException:
-                print('ERROR WITH DATA')
-                print(str(data))
-
+    count += process_file('/home/budd/workspace/iTunes/es-albums.json', 'Album')
+    count += process_file('/home/budd/workspace/iTunes/es-artists.json', 'Artist')
     print('Created documents: ' + str(count))