MISP
/
misp-warninglists
mirror of https://github.com/MISP/misp-warninglists
2
0
Fork 0
Code Issues Releases Wiki Activity

Merge pull request #5 from claudex/alexa 

Alexa
pull/7/head
Alexandre Dulaunoy 2016-08-03 16:00:09 +02:00 committed by GitHub
parent 41ce6cd77d 16a4265688
commit 7a52d7bf8d
1 changed files with 5 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
tools/generate-alexa.py
View File

@ -6,15 +6,18 @@ import zipfile
import datetime import datetime
import json import json
alexa_url = "http://s3.amazonaws.com/alexa-static/top-1mcsv.zip" alexa_url = "http://s3.amazonaws.com/alexa-static/top-1m.csv.zip"
alexa_file = "top-1m.csv.zip" alexa_file = "top-1m.csv.zip"
user_agent = {"User-agent":"Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:46.0) Gecko/20100101 Firefox/46.0"} user_agent = {"User-agent":"Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:46.0) Gecko/20100101 Firefox/46.0"}
r = requests.get(alexa_url, headers=user_agent) r = requests.get(alexa_url, headers=user_agent)
with open(alexa_file, 'wb') as fd:
for chunk in r.iter_content(4096):
fd.write(chunk)
with zipfile.ZipFile(alexa_file, 'r') as alexa_lists: with zipfile.ZipFile(alexa_file, 'r') as alexa_lists:
for name in alexa_lists.namelist(): for name in alexa_lists.namelist():
if name == "top-1m.csv": if name == "top-1m.csv":
with alexa_lists.open(name) as top: with alexa_lists.open(name) as top:
top1000 = top.readlines()[0:999] top1000 = top.readlines()[:1000]
else: else:
continue continue