Skip to content

Instantly share code, notes, and snippets.

@kamleong
Last active May 17, 2024 21:49
Show Gist options
  • Save kamleong/531d80b3de933f86179a49173ddf3245 to your computer and use it in GitHub Desktop.
Save kamleong/531d80b3de933f86179a49173ddf3245 to your computer and use it in GitHub Desktop.
file.py
#!/usr/bin/env python3
import sys
print(sys.argv[0]) ## __file__
filepath = sys.argv[1] if len(sys.argv) > 1 else r"C:\Windows\win.ini"
print(filepath)
#f2 = open(r"B:\a.txt","a+")
#f2.write("Hello \n")
#f2.close()
dictObj = {}
import re
with open(filepath, encoding="utf-8") as f1:
count = 0
while True:
line = f1.readline()
if not line: break
count = count + 1
line = line.strip()
if line=="": continue
print(line)
if line.startswith("http") :
m = re.search(r'http[s]?://([^\/\:]+)', line)
host = m.group(1)
#print( host )
if host not in dictObj : dictObj[host] = 0
dictObj[host] = dictObj[host] + 1
# if
# while
dictObj = dict(sorted(dictObj.items(), key=lambda x:x[1], reverse=True))
#print(dictObj)
count = 0
for k in dictObj :
count = count + 1
v = dictObj[k]
print(k + " [" + str(v) + "]")
#if count > 5 : break
# for
# with
quit()
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment