Removed module 'string'.

[bookmarks_db.git] / Robots / bkmk_rsimple.py

diff --git a/Robots/bkmk_rsimple.py b/Robots/bkmk_rsimple.py
index 27c7dc3625c5e00554f55538bcf5c77a806d1ecd..14a3f5804799365ebd0db637f8ecb04a3b593df9 100644 (file)
--- a/Robots/bkmk_rsimple.py
+++ b/Robots/bkmk_rsimple.py
@@ -5,7 +5,7 @@
 """
 
 
-import sys, os, string
+import sys, os
 import time, urllib
 from base64 import b64encode
 from urlparse import urljoin
@@ -21,6 +21,7 @@ class RedirectException(Exception):
    }
    def __init__(self, errcode, newurl):
       Exception.__init__(self, "(%s) to %s" % (self.reloc_dict[errcode], newurl))
+      self.url = newurl
 
 
 class MyURLopener(urllib.URLopener):
@@ -44,19 +45,30 @@ class MyURLopener(urllib.URLopener):
 
 urllib._urlopener = MyURLopener()
 
-server_version = "bookmarks_db (Python-urllib/%s)" % urllib.__version__
-urllib._urlopener.addheaders[0] = ('User-agent', server_version)
+# Fake headers to pretend this is a real browser
+_version = "Links (2.1; Linux 2.6 i686; 80x25)"
+urllib._urlopener.addheaders[0] = ('User-Agent', _version)
+_version = "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', _version)
+
+urllib._urlopener.addheader('Connection', 'close')
+urllib._urlopener.addheader('Content-Length', '0')
+urllib._urlopener.addheader('Accept', '*/*')
+urllib._urlopener.addheader('Accept-Language', 'ru,en')
+urllib._urlopener.addheader('Cache-Control', 'max-age=300')
+urllib._urlopener.addheader('Referer', 'http://www.yahoo.com/')
 
 
 def get_error(msg):
-   if type(msg) == type(""):
+   if isinstance(msg, str):
       return msg
 
    else:
       s = []
       for i in msg:
-         s.append("'%s'" % string.join(string.split(str(i), "\n"), "\\n"))
-      return "(%s)" % string.join(s)
+         s.append("'%s'" % str(i).replace('\n', "\\n"))
+      return "(%s)" % ' '.join(s)
 
 
 urllib_ftpwrapper = urllib.ftpwrapper
@@ -66,7 +78,7 @@ class myftpwrapper(urllib_ftpwrapper):
    def __init__(self, user, passwd, host, port, dirs):
       urllib_ftpwrapper.__init__(self, user, passwd, host, port, dirs)
       global ftpcache_key
-      ftpcache_key = (user, host, port, string.join(dirs, '/'))
+      ftpcache_key = (user, host, port, tuple(dirs))
 
 urllib.ftpwrapper = myftpwrapper
 
@@ -144,9 +156,7 @@ class robot_simple(Robot):
                      if self.log: self.log("   no charset in Content-Type header")
                   if content_type == "text/html":
                      parser = parse_html(fname, charset, self.log)
-                     title = parser.title.replace('\r', '').replace('\n', ' ').strip()
-                     bookmark.real_title = parser.unescape(title)
-                     if self.log: self.log("   final title    : %s" % bookmark.real_title)
+                     bookmark.real_title = parser.title
                      if parser.refresh:
                         refresh = parser.refresh
                         try:
@@ -158,6 +168,10 @@ class robot_simple(Robot):
                         except (IndexError, ValueError):
                            raise RedirectException("html", "Bad redirect to %s (%s)" % (url, refresh))
                         else:
+                           try:
+                              timeout = int(refresh.split(';')[0])
+                           except ValueError:
+                              pass # float timeout
                            raise RedirectException("html", "%s (%s sec)" % (url, timeout))
 
                      # Get favicon.ico
@@ -165,9 +179,18 @@ class robot_simple(Robot):
                      if not icon:
                         icon = "/favicon.ico"
                      icon = urljoin("%s://%s%s" % (url_type, url_host, url_path), icon)
-                     if self.log: self.log("   icon           : %s" % icon)
+                     if self.log: self.log("   looking for icon at: %s" % icon)
                      try:
-                        fname, headers = urllib.urlretrieve(icon)
+                        for i in range(8):
+                           try:
+                              fname, headers = urllib.urlretrieve(icon)
+                           except RedirectException, e:
+                              icon = e.url
+                              if self.log: self.log("       redirect to : %s" % icon)
+                           else:
+                              break
+                        else:
+                           raise IOError("Too many redirects")
                      except:
                         etype, emsg, tb = sys.exc_info()
                         if self.log: self.log("   no icon        : %s %s" % (etype, emsg))
@@ -181,8 +204,9 @@ class robot_simple(Robot):
                            icon = icon_file.read()
                            icon_file.close()
                            bookmark.icon = "data:%s;base64,%s" % (content_type, b64encode(icon))
+                           if self.log: self.log("   got icon       : %s" % content_type)
                         else:
-                           if self.log: self.log("   no icon        : %s" % content_type)
+                           if self.log: self.log("   no icon        : bad content type '%s'" % content_type)
                         os.remove(fname)
 
                except KeyError: