def upload_logset():
#created specifies if the new_logset is a pre-existing object or new
if request.form['name'] == "":
return jsonify(dict(status = 'Error', message = "Provide a name for the logset"))
new_logset, created = LogsetMetadata.objects.get_or_create(\
name = request.form['name']\
, creator_name = current_user.name\
, users_with_access = [current_user.name]\
)
uploaded_file = request.files['file']
if uploaded_file:
filename = secure_filename(uploaded_file.filename)
uploaded_file.save(os.path.join(app.config['UPLOAD_FOLDER'], filename))
lp = LogParser()
file_location = app.config['UPLOAD_FOLDER'] + filename
collection_name = current_user.name + '_' + request.form['name']
if lp.load_apache_log_file_into_DB(file_location, collection_name) is False:
return jsonify(dict(status = 'Error', message='The data was not stored'))
else:
try:
new_logset.save()
return jsonify(dict(status = 'Success', message = 'Upload successful'))
except db.NotUniqueError:
return jsonify(dict(status = 'Error', message = 'Logset by this name already exists'))
else:
return jsonify(dict(status = 'Error', message = 'No file provided'))
from log_analyzer import LogAnalyzer, LogParser
# from json import json
if __name__ == '__main__':
lp = LogParser()
lp.load_apache_log_file_into_DB('access.2','access_log')
# lp.load_apache_log_file_into_DB('access_log_2','access_log')
# lp.load_apache_log_file_into_DB('access_log_3','access_log')
# lp.load_apache_log_file_into_DB('access_log_4','access_log')
# la = LogAnalyzer(db = 'test', collection = 'access_log')
# print la.get_log_data(collection_name = 'access_log', page_number = 5)
# print la.get_log_data('access_log')
# print la.count_hits('access_log')
# df = la.load_apache_logs_into_DataFrame()
# data = la.group_by(df,['request_country'])
# print json.dumps(la.daily_bandwidth_sums())
# print la.sum(data, 'request_size')
# print (la.count(data, ['request_country'])['request_country']) #group_by os and browser and return the count of referer based on that groups. notice the field selection after the count, otherwise all the fields will be printed
# print la.count(data, 'referer')
# data = la.group_by(df, 'referer')
# data = la.count(data, 'referer')
# print la.to_dict(data)
# #print la.to_dict(data, 'referer', 'count')
# data = la.group_by(df, 'os')
# data = la.count(data, 'os')
# print data
# data = la.group_by(df, 'request_country')
# data = la.count(data, 'request_country')
# print data
# data = la.group_by(df, 'browser')
# data = la.count(data, 'browser')