CIRCL
/
lookyloo
mirror of https://github.com/CIRCL/lookyloo
2
0
Fork 0
Code Issues Releases Wiki Activity

Merge pull request #128 from FafnerKeyZee/patch-7 

Update stats.py
pull/129/head
Raphaël Vinot 2020-11-20 15:47:24 +01:00 committed by GitHub
parent 7dc38d0ff2 10541a2e75
commit 0ad2c4d573
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 16 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
tools/stats.py
View File

@ -9,8 +9,8 @@ stats = {}
today = datetime.date.today()
calendar_week = today.isocalendar()[1]
weeks_stats = {calendar_week - 1: {'analysis': 0, 'uniq_urls': set()},
calendar_week: {'analysis': 0, 'uniq_urls': set()}}
weeks_stats = {calendar_week - 1: {'analysis': 0, 'analysis_with_redirects': 0, 'redirects': 0, 'uniq_urls': set()},
calendar_week: {'analysis': 0, 'analysis_with_redirects': 0, 'redirects': 0, 'uniq_urls': set()}}
def uniq_domains(uniq_urls):
@ -29,12 +29,18 @@ for uuid in lookyloo.capture_uuids:
if date.year not in stats:
stats[date.year] = {}
if date.month not in stats[date.year]:
stats[date.year][date.month] = {'analysis': 0, 'uniq_urls': set()}
stats[date.year][date.month] = {'analysis': 0, 'analysis_with_redirects' :0, 'redirects': 0, 'uniq_urls': set()}
stats[date.year][date.month]['analysis'] += 1
if len(cache['redirects']) > 0:
stats[date.year][date.month]['analysis_with_redirects'] += 1
stats[date.year][date.month]['redirects'] += len(cache['redirects'])
stats[date.year][date.month]['uniq_urls'].update(cache['redirects'])
stats[date.year][date.month]['uniq_urls'].add(cache['url'])
if date.isocalendar()[1] in weeks_stats:
weeks_stats[date.isocalendar()[1]]['analysis'] += 1
if len(cache['redirects']) > 0:
weeks_stats[date.isocalendar()[1]]['analysis_with_redirects'] += 1
weeks_stats[date.isocalendar()[1]]['redirects'] += len(cache['redirects'])
weeks_stats[date.isocalendar()[1]]['uniq_urls'].update(cache['redirects'])
weeks_stats[date.isocalendar()[1]]['uniq_urls'].add(cache['url'])
@ -42,6 +48,8 @@ print('Statistics for the last two weeks:')
for week_number, week_stat in weeks_stats.items():
print(f'Week {week_number}:')
print(' Number of analysis:', week_stat['analysis'])
print(' Number of analysis with redirects:', week_stat['analysis_with_redirects'])
print(' Number of redirects:', week_stat['redirects'])
print(' Number of unique URLs:', len(week_stat['uniq_urls']))
domains = uniq_domains(week_stat['uniq_urls'])
print(' Number of unique domains:', len(domains))
@ -50,19 +58,18 @@ for week_number, week_stat in weeks_stats.items():
for year, data in stats.items():
print('Year:', year)
yearly_analysis = 0
yearly_urls = 0
yearly_domains = 0
yearly_redirects = 0
for month in sorted(data.keys()):
stats = data[month]
print(' ', calendar.month_name[month])
print("\tNumber of analysis :", stats['analysis'])
print("\tNumber of analysis with redirects:", stats['analysis_with_redirects'])
print("\tNumber of redirects :", stats['redirects'])
print('\tNumber of unique URLs:', len(stats['uniq_urls']))
domains = uniq_domains(stats['uniq_urls'])
print('\tNumber of unique domains:', len(domains))
yearly_analysis += stats['analysis']
yearly_urls += len(stats['uniq_urls'])
yearly_domains += len(domains)
yearly_redirects += stats['redirects']
print(" Sum analysis:", yearly_analysis)
print(" Sum unique urls:", yearly_urls)
print(" Sum unique domains:", yearly_domains)
print(" Sum redirects:", yearly_redirects)