Fix(Robot): Stop splitting and un-splitting URLs

[bookmarks_db.git] / Robots / bkmk_rurllib.py

diff --git a/Robots/bkmk_rurllib.py b/Robots/bkmk_rurllib.py
index f71a8d2bb55d22f5f823e44fae765abc800c7e32..057c018e510c9fc6ff4de4e992e6cbd765a0ce63 100644 (file)
--- a/Robots/bkmk_rurllib.py
+++ b/Robots/bkmk_rurllib.py
@@ -5,18 +5,15 @@ This file is a part of Bookmarks database and Internet robot.
 """
 
 __author__ = "Oleg Broytman <phd@phdru.name>"
-__copyright__ = "Copyright (C) 2000-2023 PhiloSoft Design"
+__copyright__ = "Copyright (C) 2000-2024 PhiloSoft Design"
 __license__ = "GNU GPL"
 
 __all__ = ['robot_urllib']
 
 
-import os
-import sys
-import time
 import urllib
 
-from Robots.bkmk_robot_base import robot_base, get_error
+from Robots.bkmk_robot_base import robot_base, request_headers, get_error
 
 
 class RedirectException(Exception):
@@ -29,9 +26,9 @@ class RedirectException(Exception):
 class MyURLopener(urllib.URLopener):
     # Error 301 -- relocated (permanently)
     def http_error_301(self, url, fp, errcode, errmsg, headers, data=None):
-        if headers.has_key('location'):
+        if 'location' in headers:
             newurl = headers['location']
-        elif headers.has_key('uri'):
+        elif 'uri' in headers:
             newurl = headers['uri']
         else:
             newurl = "Nowhere"
@@ -43,6 +40,8 @@ class MyURLopener(urllib.URLopener):
     http_error_303 = http_error_301
     # Error 307 -- relocated (temporarily)
     http_error_307 = http_error_301
+    # Error 308 -- relocated (permanently)
+    http_error_308 = http_error_301
 
     # Error 401 -- authentication required
     def http_error_401(self, url, fp, errcode, errmsg, headers, data=None):
@@ -51,29 +50,24 @@ class MyURLopener(urllib.URLopener):
 
     def http_error_default(self, url, fp, errcode, errmsg, headers):
         if fp:
-            void = fp.read()
+            fp.read()
             fp.close()
         raise IOError(('http error', errcode, errmsg, headers))
 
 
-urllib._urlopener = MyURLopener()
+def add_headers(opener):
+    try:
+        _user_agent = request_headers.pop('User-Agent')
+    except KeyError:
+        pass
+    else:
+        opener.addheaders[0] = ('User-Agent', _user_agent)
+    for h, v in request_headers.items():
+        opener.addheader(h, v)
 
-# Fake headers to pretend this is a real browser
-_user_agent = "Mozilla/5.0 (X11; U; Linux 2.6 i686; en)"
-" Gecko/20001221 Firefox/2.0.0"
-urllib._urlopener.addheaders[0] = ('User-Agent', _user_agent)
-_x_user_agent = "bookmarks_db (Python %d.%d.%d; urllib/%s)" % (
-   sys.version_info[0], sys.version_info[1],
-   sys.version_info[2], urllib.__version__
-)
-urllib._urlopener.addheader('X-User-Agent', _x_user_agent)
-urllib._urlopener.addheader('Referer', '')
-
-urllib._urlopener.addheader('Accept', '*/*')
-urllib._urlopener.addheader('Accept-Language', 'ru,en')
-urllib._urlopener.addheader('Cache-Control', 'max-age=300')
-urllib._urlopener.addheader('Connection', 'close')
 
+urllib._urlopener = opener = MyURLopener()
+add_headers(opener)
 
 urllib_ftpwrapper = urllib.ftpwrapper
 ftpcache_key = None
@@ -93,18 +87,18 @@ class robot_urllib(robot_base):
     def get(self, bookmark, url, accept_charset=False):
         try:
             # Set fake referer to the base URL
-            urllib._urlopener.addheaders[2] = ('Referer', url)
+            opener.addheaders[2] = ('Referer', url)
 
             if accept_charset and bookmark.charset:
-                urllib._urlopener.addheader('Accept-Charset', bookmark.charset)
+                opener.addheader('Accept-Charset', bookmark.charset)
             try:
                 fname, headers = urllib.urlretrieve(url)
             finally:
                 if accept_charset and bookmark.charset:
                     # Remove Accept-Charset
-                    del urllib._urlopener.addheaders[-1]
+                    del opener.addheaders[-1]
 
-            infile = open(fname, 'rb')
+            infile = open(fname, 'rt')
             content = infile.read()
             infile.close()
 
@@ -127,7 +121,7 @@ class robot_urllib(robot_base):
 
     def get_ftp_welcome(self):
         global ftpcache_key
-        _welcome = urllib._urlopener.ftpcache[ftpcache_key].ftp.welcome
+        _welcome = opener.ftpcache[ftpcache_key].ftp.welcome
         # I am assuming there are no duplicate ftp URLs in db.
         # If there are - ftpcache_key in next line is invalid.
         ftpcache_key = None