pictuga
/
morss
1
1
Fork 0
Code Issues Pull Requests Actions Packages Releases Activity

crawler: add cookies support 

Somehow got dropped when splitting the big handler
master
pictuga 2017-03-25 19:51:42 -10:00
parent d3bc2926fc
commit 6f0efd5802
1 changed files with 11 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
morss/crawler.py
View File

@ -12,10 +12,10 @@ import sqlite3
import time import time
try: try:
from urllib2 import BaseHandler, Request, addinfourl, parse_keqv_list, parse_http_list, build_opener from urllib2 import BaseHandler, HTTPCookieProcessor, Request, addinfourl, parse_keqv_list, parse_http_list, build_opener
import mimetools import mimetools
except ImportError: except ImportError:
from urllib.request import BaseHandler, Request, addinfourl, parse_keqv_list, parse_http_list, build_opener from urllib.request import BaseHandler, HTTPCookieProcessor, Request, addinfourl, parse_keqv_list, parse_http_list, build_opener
import email import email
try: try:
@ -36,6 +36,15 @@ DEFAULT_UA = 'Mozilla/5.0 (X11; Linux x86_64; rv:25.0) Gecko/20100101 Firefox/25
def custom_handler(accept=None, strict=False, delay=None, encoding=None, basic=False): def custom_handler(accept=None, strict=False, delay=None, encoding=None, basic=False):
handlers = [] handlers = []
# as per urllib2 source code, these Handelers are added first
# *unless* one of the custom handlers inherits from one of them
#
# [ProxyHandler, UnknownHandler, HTTPHandler,
# HTTPDefaultErrorHandler, HTTPRedirectHandler,
# FTPHandler, FileHandler, HTTPErrorProcessor]
# & HTTPSHandler
handlers.append(HTTPCookieProcessor())
handlers.append(GZIPHandler()) handlers.append(GZIPHandler())
handlers.append(HTTPEquivHandler()) handlers.append(HTTPEquivHandler())
handlers.append(HTTPRefreshHandler()) handlers.append(HTTPRefreshHandler())