def IPv_analysis(IPv_type, exp_n, res_directory, coll, from_d, to_d, ext):
input_file_path = res_directory + exp_n + '/5.split_data_for_analysis/' + IPv_type + '/' + coll + '_' + from_d + '-' + to_d + ext
output_file_path = res_directory + exp_n + '/6.more_specifics_analysis/' + IPv_type + '/' + coll + '_' + from_d + '-' + to_d + '.csv'
write_flag = f.overwrite_file(output_file_path)
if write_flag:
print "Loading " + input_file_path + "..."
df = f.read_file(file_ext, input_file_path)
df_sort = df.sort_values(by=['MONITOR', 'PREFIX', 'TIME'])
df_sort = df_sort.reset_index(drop=True)
df_sort = df_sort.drop(['Unnamed: 0'], axis=1)
print "Data loaded successfully"
# 1.Prefix visibility analysis
print 'Getting visibility per prefix...'
monitors, prefixes, visibilities_per_prefix, updates_per_prefix, ASes = prefix_visibility_analysis(
df_sort, exp_n)
df_prefixes_per_monitor = pd.DataFrame({
'MONITOR': monitors,
'PREFIX': prefixes
})
# 2.Clustering prefixes into more specifics, least_specifics and uniques (non-specifics)
pref_types, deeps = clustering_prefixes(df_prefixes_per_monitor)
df_visibility_per_prefix = pd.DataFrame({
'MONITOR': monitors,
'PREFIX': prefixes,
'VISIBILITY': visibilities_per_prefix,
'UPDATES': updates_per_prefix,
'TYPE': pref_types,
'DEEP': deeps,
'ORIGIN': ASes
})
# 3.Clustering more specifics prefixes into TOP, single level and more specifics of other more specifics
df_more_specifics = df_visibility_per_prefix[
df_visibility_per_prefix['TYPE'] == 'more_specific']
df_more_specifics = df_more_specifics.reset_index(drop=True)
df_more_specifics = df_more_specifics.drop(['Unnamed: 0'], axis=1)
pref_types, deeps = clustering_prefixes(df_more_specifics)
# Replace types for more detailed types
df_more_specifics['TYPE'] = pref_types
df_others = df_visibility_per_prefix[
df_visibility_per_prefix['TYPE'] != 'more_specific']
df_others = df_others.reset_index(drop=True)
df_others = df_others.drop(['Unnamed: 0'], axis=1)
df_visibility_per_prefix = df_more_specifics.append(df_others,
ignore_index=True)
output_file_path = res_directory + exp_n + '/6.more_specifics_analysis/' + IPv_type + '/' + collector + '_' + from_d + '-' + to_d + '.csv'
f.save_file(df_visibility_per_prefix, ext, output_file_path)
def saveFile(self):
#Get the current textbox
textbox = self.getTabText()
#Save the file by creating a new fill
filename = tk.filedialog.asksaveasfilename()
saved_text = textbox.get('1.0', tk.END)
fm.save_file(filename, saved_text)
#Add the file name to the title of the tab
self.notebook.tab(textbox, text=filename)
step_dir = '/5.split_data_for_analysis/IPv6'
exp.per_step_dir(exp_name, step_dir)
input_file_path = result_directory + exp_name + '/4.concatenate_RIB_data/' + collector + '_' + from_date + '-' + to_date + file_ext
output_file_path = result_directory + exp_name + step_dir + '/' + collector + '_' + from_date + '-' + to_date + file_ext
write_flag = f.overwrite_file(output_file_path)
if write_flag == 1:
print "Loading " + input_file_path + "..."
df_advises = f.read_file(file_ext, input_file_path)
print "Data loaded successfully"
print "Splitting {} advises...".format(len(df_advises))
df_IPv4_updates, df_IPv6_updates = separate_IPv_types(df_advises)
df_IPv4_updates = df_IPv4_updates.drop(['Unnamed: 0'], axis=1)
df_IPv6_updates = df_IPv6_updates.drop(['Unnamed: 0'], axis=1)
step_dir = '/5.split_data_for_analysis/IPv4'
output_file_path = result_directory + exp_name + step_dir + '/' + collector + '_' + from_date + '-' + to_date + file_ext
f.save_file(df_IPv4_updates, file_ext, output_file_path)
step_dir = '/5.split_data_for_analysis/IPv6'
output_file_path = result_directory + exp_name + step_dir + '/' + collector + '_' + from_date + '-' + to_date + file_ext
f.save_file(df_IPv6_updates, file_ext, output_file_path)
print('UNAVAILABLE file: ' + file_path +
'updates.' + year_str + month_str +
dd_str + '.' + hh_str + mm_str)
# DATA FIELDS
times = []
types = []
s_IPs = []
s_AS = []
prefixes = []
AS_PATHs = []
# dump data into several lists
for i in range(len(update_lines)):
dump_into_lists(update_lines[i], times, types, s_IPs, s_AS,
prefixes, AS_PATHs, from_time, to_time)
print(' Data saved as lists!')
df_update = pd.DataFrame({
'TIME': times,
'TYPE': types,
'MONITOR': s_IPs,
'AS': s_AS,
'PREFIX': prefixes,
'AS_PATH': AS_PATHs
})
print(' Data Frame created!')
f.save_file(df_update, file_ext, output_file_path)
step_dir = '/7.combined_data'
exp.per_step_dir(exp_name, step_dir)
IP_versions = ['IPv4', 'IPv6']
collectors = [
'rrc00', 'rrc01', 'rrc04', 'rrc05', 'rrc07', 'rrc10', 'rrc11', 'rrc12',
'rrc13'
]
frames = []
for IP_v in IP_versions:
output_file_path = result_directory + exp_name + '/7.combined_data/' + IP_v + '/' + from_date + '-' + to_date + '.csv'
write_flag = f.overwrite_file(output_file_path)
if write_flag:
for coll in collectors:
df = pd.read_csv(result_directory + exp_name +
'/6.more_specifics_analysis/' + IP_v + '/' +
coll + '_' + from_date + '-' + to_date +
'.csv')
df = df.drop(['Unnamed: 0'], axis=1)
df['collector'] = coll
df['IPv'] = IP_v
frames.append(df)
result = pd.concat(frames)
result = result.reset_index(drop=True)
output_file_path = result_directory + exp_name + step_dir + '/' + from_date + '-' + to_date + '.csv'
f.save_file(result, file_ext, output_file_path)
df_time_mm = df_time_s // 60
df_time_list = df_time_mm.tolist()
df_type = df['TYPE']
df_type_list = df_type.tolist()
state_indexes = get_state_indexes(df_type_list)
print(len(state_indexes))
df_monitor = df['MONITOR']
df_monitor_list = df_monitor.tolist()
affected_messages = []
print('\nSearching affected messages...')
df_clean = df
affected_indexes = []
for i in reversed(state_indexes):
affected_indexes += get_affected_message_indexes_per_STATE(
i, df_monitor_list, df_type_list, df_time_list)
df_clean = df_clean.drop(df.index[affected_indexes])
df_clean = df_clean.reset_index(drop=True)
df_clean = df_clean.drop(['Unnamed: 0'], axis=1)
f.save_file(df_clean, file_ext, output_file_path)
print('Clean data saved')
df_RIBS = pd.DataFrame({
'TIME': times,
'TYPE': types,
'MONITOR': s_IPs,
'AS': s_AS,
'PREFIX': prefixes,
'AS_PATH': AS_PATHs
})
print(' Data Frame created!')
# Load clean data
input_file_path = result_directory + exp_name + '/3.data_cleaning/' + collector + '_' + from_date + '-' + to_date + file_ext
output_file_path = result_directory + exp_name + step_dir + '/' + collector + '_' + from_date + '-' + to_date + file_ext
write_flag = f.overwrite_file(output_file_path)
print "Loading " + input_file_path + "..."
df_updates = f.read_file(file_ext, input_file_path)
print "Data loaded successfully"
print "Concatenating RIBs to updates"
list_complete = [df_RIBS, df_updates]
df_complete = pd.concat(list_complete)
df_complete = df_complete.drop(['Unnamed: 0'], axis=1)
f.save_file(df_complete, file_ext, output_file_path)
experiments = getattr(exp, 'experiments')
experiment = experiments[exp_name]
from_date = experiment['initDay']
to_date = experiment['endDay']
result_directory = experiment['resultDirectory']
file_ext = experiment['resultFormat']
step_dir = '/2.sort_data_for_cleaning'
exp.per_step_dir(exp_name, step_dir)
input_file_path = result_directory + exp_name + '/1.load_data/' + collector + '_' + from_date + '-' + to_date + file_ext
output_file_path = result_directory + exp_name + step_dir + '/' + collector + '_' + from_date + '-' + to_date + file_ext
write_flag = f.overwrite_file(output_file_path)
if write_flag:
print('Loading ' + input_file_path + ' ...')
df = f.read_file(file_ext, input_file_path)
df_sort = df.sort_values(by=['MONITOR', 'TIME'])
df_sort = df_sort.reset_index(drop=True)
df_sort = df_sort.drop(['Unnamed: 0'], axis=1)
f.save_file(df_sort, file_ext, output_file_path)
check_output(input_file_path, output_file_path)