-
Notifications
You must be signed in to change notification settings - Fork 0
/
loader.py
406 lines (327 loc) · 13.8 KB
/
loader.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
#!/usr/bin/python3
import sys
import json
import pandas as pd
import psycopg2
import json
from psycopg2 import sql
from psycopg2.extensions import ISOLATION_LEVEL_AUTOCOMMIT
from collections import Counter
from collections import defaultdict
HELP_TEXT = ('USAGE: \033[1mloader.py\033[0m dataset_base_path\n' +
'\tdataset_base_path: path to the extracted google play store dataset folder')
# dataset url:
# https://www.kaggle.com/lava18/google-play-store-apps/
# file names without base path
APPS = 'googleplaystore.csv'
REVIEWS = 'googleplaystore_user_reviews.csv'
DB_CONFIG_PATH = 'db_config.json'
# schemes of the database tables
TABLE_SCHEMA_FILE = 'db_schema.json'
def create_connection(db_config):
con = None
cur = None
# create db connection
try:
con = psycopg2.connect(
"dbname='" + db_config['db_name'] + "' user='"
+ db_config['username'] + "' host='" + db_config['host']
+ "' password='" + db_config['password'] + "'")
except:
try:
print("user='" + db_config['username'] + "' host='" + db_config['host']
+ "' password='" + db_config['password'] + "'")
con = psycopg2.connect(
"user='" + db_config['username'] + "' host='" + db_config['host']
+ "' password='" + db_config['password'] + "'")
print('get here')
cur = con.cursor()
con.set_isolation_level(ISOLATION_LEVEL_AUTOCOMMIT)
cur.execute(sql.SQL("CREATE DATABASE {}").format(
sql.Identifier(db_config['db_name'])))
cur.close()
con.close()
con = psycopg2.connect(
"dbname='" + db_config['db_name'] + "' user='"
+ db_config['username'] + "' host='" + db_config['host']
+ "' password='" + db_config['password'] + "'")
except:
print('ERROR: Can not connect to database')
return
cur = con.cursor()
return con, cur
def is_valid_str(term):
if type(term) == str:
if len(term) > 0:
return True
def parse_list(text):
return text.split(';')
def disable_triggers(schema_info, con, cur):
for table_name in schema_info.keys():
cur.execute('ALTER TABLE ' + table_name + ' DISABLE trigger ALL;')
con.commit()
return
def enable_triggers(schema_info, con, cur):
for table_name in schema_info.keys():
cur.execute('ALTER TABLE ' + table_name + ' ENABLE trigger ALL;')
con.commit()
return
def create_schema(schema_info, con, cur):
query_drop = "DROP TABLE IF EXISTS " + ', '.join(
[key for key in schema_info]) + ';'
queries_create = []
for (name, schema) in schema_info.items():
queries_create.append("CREATE TABLE " + name + " " + schema + ";")
# run queries
for query in [query_drop] + queries_create:
cur.execute(query)
con.commit()
def tokenize_category(cat_str):
return ' '.join(cat_str.lower().split('_'))
# Takes the DataFrame from the movie file and extract all relevant information.
def extract_app_data(df_apps):
# define columns which information is useful
RELEVANT_COLUMNS = ['App', 'Category', 'Type', 'Content Rating', 'Genres']
# reduce data frame to relevant columns
apps_reduced = df_apps[RELEVANT_COLUMNS]
# extract data and create output dictonary
extracted_apps = dict()
extracted_categories = dict()
extracted_price_types = dict()
extracted_content_ratings = dict()
extracted_genres = dict()
app_name_lookup = dict()
category_name_lookup = dict()
price_type_lookup = dict()
rating_value_lookup = dict()
genre_name_lookup = dict()
id = 0
for line in apps_reduced.iterrows():
# line[0]: line number line[1]: content
if line[1]['App'] in app_name_lookup:
continue
# add simple values
values = dict()
if not is_valid_str(line[1]['App']):
continue
values['name'] = line[1]['App']
# 1:n foreign key relations
# TODO
if is_valid_str(line[1]['Category']):
category_value = tokenize_category(line[1]['Category'])
if not category_value in category_name_lookup:
cat_id = len(category_name_lookup)
category_name_lookup[category_value] = cat_id
extracted_categories[cat_id] = {
'id': cat_id, 'name': category_value}
category_id = category_name_lookup[category_value]
values['category_id'] = category_id
if is_valid_str(line[1]['Type']): # TODO
price_type_value = str(line[1]['Type'])
if not price_type_value in price_type_lookup:
price_type_id = len(price_type_lookup)
price_type_lookup[price_type_value] = price_type_id
extracted_price_types[price_type_id] = {
'id': price_type_id, 'name': price_type_value}
price_type_id = price_type_lookup[price_type_value]
values['price_type'] = price_type_id
if is_valid_str(line[1]['Content Rating']):
rating_value = str(line[1]['Content Rating'])
if not rating_value in rating_value_lookup:
rating_id = len(rating_value_lookup)
rating_value_lookup[rating_value] = rating_id
extracted_content_ratings[rating_id] = {
'id': rating_id, 'rating': rating_value}
rating_id = rating_value_lookup[rating_value]
values['content_rating'] = rating_id
# n:n foreign key relations
values['genres'] = set()
if is_valid_str(line[1]['Genres']):
genres_list_value = parse_list(line[1]['Genres'])
for genre_value in genres_list_value:
if not genre_value in genre_name_lookup:
genre_id = len(genre_name_lookup)
genre_name_lookup[genre_value] = genre_id
extracted_genres[genre_id] = {
'id': genre_id, 'name': genre_value}
genre_id = genre_name_lookup[genre_value]
values['genres'].add(genre_id)
extracted_apps[id] = values
app_name_lookup[values['name']] = id
id += 1
return {
'extracted_apps': extracted_apps,
'extracted_categories': extracted_categories,
'extracted_price_types': extracted_price_types,
'extracted_content_ratings': extracted_content_ratings,
'extracted_genres': extracted_genres
}
def extract_reviews(df_reviews):
RELEVANT_COLUMNS = ['App', 'Translated_Review']
reviews_reduced = df_reviews[RELEVANT_COLUMNS]
extracted_reviews = dict()
app_review_lookup = defaultdict(set)
id = 0
for line in reviews_reduced.iterrows():
values = dict()
if not is_valid_str(line[1]['App']):
continue
if not is_valid_str(line[1]['Translated_Review']):
continue
values['app'] = str(line[1]['App'])
values['review'] = str(line[1]['Translated_Review'])
extracted_reviews[id] = values
app_review_lookup[values['app']].add(id)
id += 1
return extracted_reviews, app_review_lookup
def process_buffers(buffers, con, cur, batch_size):
for buffer, query in buffers.values():
if len(buffer) >= batch_size:
cur.executemany(query, buffer)
con.commit()
buffer.clear()
return
def flush_buffers(buffers, con, cur, batch_size):
for buffer, query in buffers.values():
cur.executemany(query, buffer)
con.commit()
buffer.clear()
return
def get_db_literal(value):
if value == None:
return None
else:
return str(value)
def insert_apps_data(data, con, cur, batch_size):
QUERY_INSERT_APPS = (
"INSERT INTO apps (id, name, category_id, price_type, content_rating) VALUES %s")
QUERY_INSERT_CATEGORIES = "INSERT INTO categories (id, name) VALUES %s"
QUERY_INSERT_TYPES = "INSERT INTO price_types (id, name) VALUES %s"
QUERY_INSERT_RATINGS = "INSERT INTO content_ratings (id, rating) VALUES %s"
QUERY_INSERT_GENRES = "INSERT INTO genres (id, name) VALUES %s"
QUERY_INSERT_GENRES_RELATION = (
"INSERT INTO apps_genres (app_id, genre_id) VALUES %s")
apps_data = data['extracted_apps']
buffers = {
'apps_content': (list(), QUERY_INSERT_APPS),
'genres_relation': (list(), QUERY_INSERT_GENRES_RELATION)
}
for app_id, app_values in apps_data.items():
buffers['apps_content'][0].append([(
get_db_literal(app_id), get_db_literal(app_values['name']),
get_db_literal(app_values['category_id']),
get_db_literal(app_values['price_type']),
get_db_literal(app_values['content_rating']))])
for genre_id in app_values['genres']:
buffers['genres_relation'][0].append(
[(get_db_literal(app_id), get_db_literal(genre_id))])
process_buffers(buffers, con, cur, batch_size)
flush_buffers(buffers, con, cur, batch_size)
category_data = data['extracted_categories']
buffers = {'category_content': (list(), QUERY_INSERT_CATEGORIES)}
for category_id, category_values in category_data.items():
buffers['category_content'][0].append(
[(get_db_literal(category_id), get_db_literal(category_values['name']))])
process_buffers(buffers, con, cur, batch_size)
flush_buffers(buffers, con, cur, batch_size)
price_type_data = data['extracted_price_types']
buffers = {'price_type_content': (list(), QUERY_INSERT_TYPES)}
for price_type_id, price_type_values in price_type_data.items():
buffers['price_type_content'][0].append(
[(get_db_literal(price_type_id), get_db_literal(price_type_values['name']))])
process_buffers(buffers, con, cur, batch_size)
flush_buffers(buffers, con, cur, batch_size)
content_rating_data = data['extracted_content_ratings']
buffers = {'rating_content': (list(), QUERY_INSERT_RATINGS)}
for rating_id, rating_values in content_rating_data.items():
buffers['rating_content'][0].append(
[(get_db_literal(rating_id), get_db_literal(rating_values['rating']))])
process_buffers(buffers, con, cur, batch_size)
flush_buffers(buffers, con, cur, batch_size)
genre_data = data['extracted_genres']
buffers = {'genres_content': (list(), QUERY_INSERT_GENRES)}
for genre_id, genre_values in genre_data.items():
buffers['genres_content'][0].append(
[(get_db_literal(genre_id), get_db_literal(genre_values['name']))])
process_buffers(buffers, con, cur, batch_size)
flush_buffers(buffers, con, cur, batch_size)
return
def insert_review_data(data, con, cur, batch_size):
QUERY_INSERT_REVIEWS = "INSERT INTO reviews (id, app_id, review) VALUES %s"
buffers = {'review_content': (list(), QUERY_INSERT_REVIEWS)}
for review_id, review_values in data.items():
buffers['review_content'][0].append([(
get_db_literal(review_id),
get_db_literal(review_values['app_id']),
get_db_literal(review_values['review']))])
process_buffers(buffers, con, cur, batch_size)
flush_buffers(buffers, con, cur, batch_size)
return
def match_apps_reviews(apps_data, extracted_reviews, app_review_lookup):
review_app_lookup = dict() # app id -> review id
extracted_apps = apps_data['extracted_apps']
# match reviews to apps
to_remove = list()
id = 0
for app_id, app_values in extracted_apps.items():
if not app_values['name'] in app_review_lookup:
to_remove.append(app_id)
continue
for review_id in app_review_lookup[app_values['name']]:
extracted_reviews[review_id]['app_id'] = app_id
# remove apps without any review
for app_id in to_remove:
del extracted_apps[app_id]
# remove reviews without any app
to_remove = list()
for review_id, review_values in extracted_reviews.items():
if not 'app_id' in review_values:
to_remove.append(review_id)
for review_id in to_remove:
del extracted_reviews[review_id]
return apps_data, extracted_reviews
def get_df(path):
df = pd.read_csv(path)
df = df.drop_duplicates()
return df
def main(argc, argv):
if argc != 2:
print(HELP_TEXT)
return
dataset_base_path = argv[1] + '/'
apps_path = dataset_base_path + APPS
reviews_path = dataset_base_path + REVIEWS
df_apps = get_df(apps_path)
print('Read apps')
df_reviews = get_df(reviews_path)
print('Read reviews')
print('Extract apps data from csv ...')
extracted_apps = extract_app_data(df_apps)
print('Extract review data from csv ...')
extracted_reviews, app_review_lookup = extract_reviews(df_reviews)
print('Match apps and reviews ...')
extracted_apps, extracted_reviews = match_apps_reviews(
extracted_apps, extracted_reviews, app_review_lookup)
print('Connect to database ...')
f_db_config = open(DB_CONFIG_PATH, 'r')
db_config = json.load(f_db_config)
f_db_config.close()
con, cur = create_connection(db_config)
batch_size = db_config['batch_size']
# get schema
print('Read schema file ...')
schema_file = open(TABLE_SCHEMA_FILE, 'r')
schema_info = json.load(schema_file)
schema_file.close()
print('Create Schema ...')
create_schema(schema_info, con, cur)
print('Insert data into database ...')
disable_triggers(schema_info, con, cur)
print('Insert app data ...')
insert_apps_data(extracted_apps, con, cur, batch_size)
print('Insert reviews ...')
insert_review_data(extracted_reviews, con, cur, batch_size)
enable_triggers(schema_info, con, cur)
print('Done.')
if __name__ == "__main__":
main(len(sys.argv), sys.argv)