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

[bookmarks_db.git] / bkmk_objects.py

diff --git a/bkmk_objects.py b/bkmk_objects.py
index 4031e5f60d8dfa0ac1379ce819f2962a202baa93..bf5904b3287d2872dd6e871398751bc3118c6700 100644 (file)
--- a/bkmk_objects.py
+++ b/bkmk_objects.py
@@ -5,19 +5,21 @@ 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__ = ['Folder', 'Bookmark', 'Ruler', 'Walker', 'Writer', 'Robot',
            'InverseLinker', 'Linear', 'make_linear', 'make_tree', 'break_tree',
            'quote_title', 'unquote_title', 'parse_params', 'set_params',
-    ]
+           ]
 
 
-import os, urllib
+from urllib.parse import unquote
+import os
 
 BKMK_FORMAT = os.environ.get("BKMK_FORMAT", "MOZILLA")
 
+
 class Folder(list):
     isFolder = 1
     isBookmark = 0
@@ -29,7 +31,7 @@ class Folder(list):
         self.last_modified = last_modified
 
     def walk_depth(self, walker, level=0):
-        if hasattr(self, "header"): # root folder
+        if hasattr(self, "header"):  # root folder
             prune = 0
             walker.root_folder(self)
         else:
@@ -56,32 +58,6 @@ class Bookmark(object):
     def __init__(self, href, add_date, last_visit=None, last_modified=None,
                  keyword=None, comment='', icon_href=None, icon=None,
                  charset=None, parser_charset=None):
-        protocol, request = urllib.splittype(href)
-        user, password, port = None, None, None
-        host, path = urllib.splithost(request)
-        if host:
-            user, host = urllib.splituser(host)
-            if user:
-                user, password = urllib.splitpasswd(user)
-            host, port = urllib.splitport(host)
-            if port: port = int(port)
-
-        if protocol == 'place':
-            href = protocol + ":"
-        else:
-            href = protocol + "://"
-        if user:
-            href += urllib.quote(user)
-            if password:
-                href += ':' + urllib.quote(password)
-            href += '@'
-        if host:
-            href += host.decode(parser_charset or 'utf-8').encode('idna')
-            if port:
-                href += ':%d' % port
-        if path:
-            href += path
-
         self.href = href
         self.add_date = add_date
         self.last_visit = last_visit
@@ -140,7 +116,7 @@ class Robot(object):
         self.log = log
 
     def stop(self):
-        pass # Nothing to do on cleanup
+        pass  # Nothing to do on cleanup
 
 
 # Helper class to make inverese links (nodes linked to their parent)
@@ -150,7 +126,8 @@ class InverseLinker(Walker):
 
     def start_folder(self, f, level):
         f.parent = self.parent_stack[-1]
-        self.parent_stack.append(f) # Push the folder onto the stack of parents
+        # Push the folder onto the stack of parents
+        self.parent_stack.append(f)
 
     def end_folder(self, f, level):
         del self.parent_stack[-1]   # Pop off the stack
@@ -181,7 +158,8 @@ class Linear(Walker):
         self.add_object(r)
 
 
-# Helper - make linked linear represenatation of the tree, suitable to be stored in sequential storage
+# Helper - make linked linear represenatation of the tree,
+# suitable to be stored in sequential storage.
 def make_linear(root_folder):
     linker = InverseLinker()
     root_folder.walk_depth(linker)
@@ -190,7 +168,8 @@ def make_linear(root_folder):
     root_folder.walk_depth(linear)
 
 
-# Helper, opposite of make_linear - make a tree from the linked linear representation
+# Helper, opposite of make_linear -
+# make a tree from the linked linear representation.
 def make_tree(linear):
     root_folder = linear[0]
     del linear[0]
@@ -200,6 +179,7 @@ def make_tree(linear):
 
     return root_folder
 
+
 def break_tree(linear):
     del linear[0]
 
@@ -212,11 +192,18 @@ def quote_title(title):
         title = title.replace("'", "&#39;")
     return title
 
+
 def unquote_title(title):
     if BKMK_FORMAT == "MOZILLA":
-        from HTMLParser import HTMLParser
-        title = HTMLParser().unescape(title.replace("&amp;", '&').decode('utf-8'))
-        title = title.encode('utf-8').replace("&#39;", "'")
+        try:
+            from HTMLParser import HTMLParser
+        except ImportError:
+            from html import unescape
+        else:
+            unescape = HTMLParser().unescape
+        title = unescape(
+            title.replace("&amp;", '&'))
+        title = title.replace("&#39;", "'")
     return title
 
 
@@ -226,9 +213,10 @@ def parse_params(param_str):
     param_list = {}
     for param in params:
         key, value = param.split('=', 1)
-        param_list[key] = value
+        param_list[key] = unquote(value)
     return main_param, param_list
 
+
 def set_params(obj, params):
     if hasattr(params, "items"):
         params = params.items()