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

chg: Store iframe info and PNG

pull/14/merge
Raphaël Vinot 2018-02-09 18:29:22 +01:00
parent 92ffd1b7da
commit 510c35b57f
1 changed files with 9 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
lookyloo/__init__.py
View File

@ -19,6 +19,7 @@ import pathlib
from zipfile import ZipFile, ZIP_DEFLATED from zipfile import ZipFile, ZIP_DEFLATED
from io import BytesIO from io import BytesIO
import base64
app = Flask(__name__) app = Flask(__name__)
@ -68,19 +69,23 @@ def scrape():
depth = request.form.get('depth') depth = request.form.get('depth')
if depth is None: if depth is None:
depth = 1 depth = 1
items = crawl(SPLASH, url, depth) items = crawl(SPLASH, url, depth, log_enabled=True, log_level='INFO')
if not items: if not items:
# broken # broken
pass pass
width = len(str(len(items))) width = len(str(len(items)))
i = 1
dirpath = os.path.join(HAR_DIR, datetime.now().isoformat()) dirpath = os.path.join(HAR_DIR, datetime.now().isoformat())
os.makedirs(dirpath) os.makedirs(dirpath)
for item in items: for i, item in enumerate(items):
harfile = item['har'] harfile = item['har']
png = base64.b64decode(item['png'])
child_frames = item['childFrames']
with open(os.path.join(dirpath, '{0:0{width}}.har'.format(i, width=width)), 'w') as f: with open(os.path.join(dirpath, '{0:0{width}}.har'.format(i, width=width)), 'w') as f:
json.dump(harfile, f) json.dump(harfile, f)
i += 1 with open(os.path.join(dirpath, '{0:0{width}}.png'.format(i, width=width)), 'wb') as f:
f.write(png)
with open(os.path.join(dirpath, '{0:0{width}}.frames.json'.format(i, width=width)), 'w') as f:
json.dump(child_frames, f)
return tree(0) return tree(0)
return render_template('scrape.html') return render_template('scrape.html')