python-ox/ox/net.py

170 lines
5.4 KiB
Python
Raw Normal View History

2008-04-27 16:54:37 +00:00
# -*- coding: utf-8 -*-
2008-06-19 09:21:21 +00:00
# vi:si:et:sw=4:sts=4:ts=4
2008-07-06 13:00:06 +00:00
# GPL 2008
2014-09-30 19:04:46 +00:00
from __future__ import with_statement, print_function
2008-04-28 09:55:30 +00:00
import gzip
import json
import os
import re
2012-04-30 08:40:10 +00:00
import struct
2008-04-27 16:54:37 +00:00
2016-01-14 11:39:10 +00:00
from six import BytesIO, PY2
from six.moves import urllib
from chardet.universaldetector import UniversalDetector
2008-04-27 16:54:37 +00:00
2012-08-21 06:41:25 +00:00
DEBUG = False
2008-04-27 16:54:37 +00:00
# Default headers for HTTP requests.
DEFAULT_HEADERS = {
2015-11-03 22:15:57 +00:00
'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Firefox/38.0',
2010-08-12 21:35:31 +00:00
'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
'Accept-Language': 'en-us,en;q=0.5',
2008-06-19 09:21:21 +00:00
'Accept-Encoding': 'gzip'
}
2008-04-27 16:54:37 +00:00
def status(url, data=None, headers=DEFAULT_HEADERS):
2008-06-19 09:21:21 +00:00
try:
f = open_url(url, data, headers)
2008-06-19 09:21:21 +00:00
s = f.code
2014-09-30 19:04:46 +00:00
except urllib.error.HTTPError as e:
2008-06-19 09:21:21 +00:00
s = e.code
return s
def exists(url, data=None, headers=DEFAULT_HEADERS):
2008-06-19 09:21:21 +00:00
s = status(url, data, headers)
if s >= 200 and s < 400:
return True
return False
2012-09-09 17:28:11 +00:00
def get_headers(url, data=None, headers=DEFAULT_HEADERS):
2008-06-19 09:21:21 +00:00
try:
f = open_url(url, data, headers)
2008-06-19 09:21:21 +00:00
f.headers['Status'] = "%s" % f.code
headers = f.headers
f.close()
2014-09-30 19:04:46 +00:00
except urllib.error.HTTPError as e:
2008-06-19 09:21:21 +00:00
e.headers['Status'] = "%s" % e.code
headers = e.headers
return dict(headers)
def get_json(url, data=None, headers=DEFAULT_HEADERS):
return json.loads(read_url(url, data, headers).decode('utf-8'))
def open_url(url, data=None, headers=DEFAULT_HEADERS):
2016-01-14 11:39:10 +00:00
if PY2:
2014-10-06 06:22:25 +00:00
if not isinstance(url, bytes):
url = url.encode('utf-8')
2016-01-14 11:39:10 +00:00
else:
if isinstance(url, bytes):
url = url.decode('utf-8')
2008-06-19 09:21:21 +00:00
url = url.replace(' ', '%20')
2016-01-14 11:39:10 +00:00
if data and not PY2 and not isinstance(data, bytes):
2014-10-04 19:04:55 +00:00
data = data.encode('utf-8')
2014-09-30 19:04:46 +00:00
req = urllib.request.Request(url, data, headers)
return urllib.request.urlopen(req)
2008-04-27 16:54:37 +00:00
def read_url(url, data=None, headers=DEFAULT_HEADERS, return_headers=False, unicode=False):
2012-08-21 06:41:25 +00:00
if DEBUG:
2014-09-30 19:04:46 +00:00
print('ox.net.read_url', url)
f = open_url(url, data, headers)
2012-08-17 20:20:35 +00:00
result = f.read()
2008-06-19 09:21:21 +00:00
f.close()
if f.headers.get('content-encoding', None) == 'gzip':
2014-09-30 19:04:46 +00:00
result = gzip.GzipFile(fileobj=BytesIO(result)).read()
if unicode:
2014-09-30 19:04:46 +00:00
ctype = f.headers.get('content-type', '').lower()
if 'charset' in ctype:
encoding = ctype.split('charset=')[-1]
else:
encoding = detect_encoding(result)
if not encoding:
encoding = 'latin-1'
2012-08-17 20:20:35 +00:00
result = result.decode(encoding)
if return_headers:
2008-06-19 09:21:21 +00:00
f.headers['Status'] = "%s" % f.code
2014-09-30 19:04:46 +00:00
headers = {}
for key in f.headers:
headers[key.lower()] = f.headers[key]
return headers, result
2012-08-17 20:20:35 +00:00
return result
2008-04-27 16:54:37 +00:00
def detect_encoding(data):
2014-09-30 19:04:46 +00:00
data_lower = data.lower().decode('utf-8', 'ignore')
charset = re.compile('content="text/html; charset=(.*?)"').findall(data_lower)
if not charset:
2014-09-30 19:04:46 +00:00
charset = re.compile('meta charset="(.*?)"').findall(data_lower)
if charset:
return charset[0].lower()
2008-06-19 09:21:21 +00:00
detector = UniversalDetector()
2014-09-30 19:04:46 +00:00
p = 0
l = len(data)
s = 1024
while p < l:
detector.feed(data[p:p+s])
2008-06-19 09:21:21 +00:00
if detector.done:
break
2014-09-30 19:04:46 +00:00
p += s
2008-06-19 09:21:21 +00:00
detector.close()
return detector.result['encoding']
get_url=read_url
def save_url(url, filename, overwrite=False):
2009-07-15 13:53:40 +00:00
if not os.path.exists(filename) or overwrite:
dirname = os.path.dirname(filename)
2014-12-09 12:11:09 +00:00
if dirname and not os.path.exists(dirname):
2009-07-15 13:53:40 +00:00
os.makedirs(dirname)
data = read_url(url)
2014-10-02 08:28:22 +00:00
with open(filename, 'wb') as f:
f.write(data)
2009-07-15 13:53:40 +00:00
2012-04-30 08:40:10 +00:00
def oshash(url):
def get_size(url):
2014-09-30 19:04:46 +00:00
req = urllib.request.Request(url, headers=DEFAULT_HEADERS.copy())
2012-04-30 08:40:10 +00:00
req.get_method = lambda : 'HEAD'
2014-09-30 19:04:46 +00:00
u = urllib.request.urlopen(req)
2012-04-30 08:40:10 +00:00
if u.code != 200 or not 'Content-Length' in u.headers:
raise IOError
return int(u.headers['Content-Length'])
def get_range(url, start, end):
headers = DEFAULT_HEADERS.copy()
headers['Range'] = 'bytes=%s-%s' % (start, end)
2014-09-30 19:04:46 +00:00
req = urllib.request.Request(url, headers=headers)
u = urllib.request.urlopen(req)
2012-04-30 08:40:10 +00:00
return u.read()
try:
longlongformat = 'q' # long long
bytesize = struct.calcsize(longlongformat)
filesize = get_size(url)
hash = filesize
head = get_range(url, 0, min(filesize, 65536))
if filesize > 65536:
tail = get_range(url, filesize-65536, filesize)
if filesize < 65536:
2015-03-19 13:27:50 +00:00
f = BytesIO(head)
for x in range(int(filesize/bytesize)):
buffer = f.read(bytesize)
2012-04-30 08:40:10 +00:00
(l_value,)= struct.unpack(longlongformat, buffer)
hash += l_value
hash = hash & 0xFFFFFFFFFFFFFFFF #cut off 64bit overflow
else:
for offset in range(0, 65536, bytesize):
buffer = head[offset:offset+bytesize]
(l_value,)= struct.unpack(longlongformat, buffer)
hash += l_value
hash = hash & 0xFFFFFFFFFFFFFFFF #cut of 64bit overflow
for offset in range(0, 65536, bytesize):
buffer = tail[offset:offset+bytesize]
(l_value,)= struct.unpack(longlongformat, buffer)
hash += l_value
hash = hash & 0xFFFFFFFFFFFFFFFF
returnedhash = "%016x" % hash
return returnedhash
except(IOError):
return "IOError"