Scraping reviews left on any app from the GPlay store
This code is meant to be used for educational purposes and I take no responsibility for what you do with the code.
# Find app package names at https://www.fileplanet.com/apk
# Google play scraper library https://github.com/facundoolano/google-play-scraper
from google_play_scraper import app
from google_play_scraper import Sort, reviews, reviews_all
import pandas as pd
import csv
import os, globdef extract_comments(language, country, app_package_name):
it = iter(range(1,6)) # to iterate the following loop over the selected star ratings
for x in it: # (1,6) being star ratings from 1 to 5
result = reviews_all(
app_package_name,
sleep_milliseconds=0, # defaults to 0
lang=language, # defaults to 'en'
country=country, # defaults to 'us'
sort=Sort.NEWEST, # defaults to Sort.MOST_RELEVANT
filter_score_with=x # defaults to None(means all score)
)
if result == []: # Skips x value in case there are no reviews at this score
continue
else:
toCSV = result
# Add line saying if result for x=x are 0, then skip to next value of x
keys = toCSV[0].keys()
filename = 'reviews_' + language + '_' + str(x) + '.csv' # eg. 'reviews_it_5.csv'
with open(filename, 'w', encoding='utf8', newline='') as output_file:
dict_writer = csv.DictWriter(output_file, fieldnames=toCSV[0].keys())
dict_writer.writeheader()
dict_writer.writerows(toCSV)The above code creates a separate CSV for each star rating
# Merge CSVs in one
# https://blog.softhints.com/how-to-merge-multiple-csv-files-with-python/
path_1 = ''
path_2 = 'D:/Your location of choice'
def merge_CSVs(language):
filename = 'reviews_' + language
all_files = glob.glob(os.path.join(path_1, filename + '*.csv'))
df_from_each_file = (pd.read_csv(f, sep=',') for f in all_files)
df_merged = pd.concat(df_from_each_file, ignore_index=True)
df_merged.to_csv(path_2 + '/All_reviews_' + language + '.csv', index=False)# Add language column
def add_language(language):
filename = 'All_reviews_' + language + '.csv'
all_reviews_df = pd.read_csv(path_2 + '/' + filename)
all_reviews_df['language'] = language
all_reviews_df.insert(loc=len(all_reviews_df.columns), column='platform', value='Android') # adding platform column
all_reviews_df.insert(3, 'title', '') # adding title column to make df equal to iOS df
all_reviews_df.rename(columns={'at': 'date'}, inplace=True) # changing column name from 'at' to 'date'
return all_reviews_df, filename# Export results to csv
def export(all_reviews_df, filename):
all_reviews_df.to_csv(path_2 + '/' + filename, index=False)# Delete remaining single rating files
def remove_single_CSVs():
all_single_ratings = glob.glob(os.path.join(path_1, 'reviews*.csv'))
for x in all_single_ratings:
os.remove(x)# Master command
def master_command(language, country, app_package_name):
extract_comments(language, country, app_package_name)
merge_CSVs(language)
all_reviews_df, filename = add_language(language)
export(all_reviews_df, filename)
remove_single_CSVs() master_command('en', 'US', 'com.spotify.music') # (language, country, package name)As an example we have the Spotify package name inputted. However, you can of course put in any package name. You can usually find it in the URL, and it's the part that starts with com as in the example below
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent