Fixed windows fragment offset to read from the description

klembord/__init__.py

@@ -16,9 +16,11 @@
 """
 
 import sys
-from collections import OrderedDict, Mapping, ByteString, Sequence
+from collections import OrderedDict
+from collections.abc import Mapping, Sequence
 if sys.platform.startswith('win32'):
 	from .winclipboard import WinClipboard
+	from .winclipboard import parse_description
 	WINDOWS = True
 	LINUX = False
 else:
@@ -199,7 +201,7 @@ def set_with_rich_text(self, text, html):
 		else:
 			raise TypeError('text or html is not str')
 
-	def get_with_rich_text(self):
+	def get_with_rich_text(self, include_win_description=False):
 		"""Get the contents of the selection in plaintext and HTML formats.
 
 		Returns:
@@ -209,11 +211,13 @@ def get_with_rich_text(self):
 		"""
 
 		if WINDOWS:
+			description = OrderedDict()
 			content = self.get((W_HTML, W_UNICODE))
 			html = content[W_HTML]
 			if html:
 				# Strip HTML Format additions and return just the fragment.
-				html = html.decode(UTF8)[131:-38]
+				description = parse_description(html.decode(UTF8))
+				html = html[ description['StartFragment']: description['EndFragment']].decode(UTF8) # Must decode after slicing
 			text = content[W_UNICODE]
 			if text:
 				text = text.decode(UTF16)
@@ -235,7 +239,11 @@ def get_with_rich_text(self):
 						html = html.decode(UTF16)
 					except UnicodeDecodeError:
 						html = html.decode(UTF8, 'ignore')
-		return (text, html)
+
+		if WINDOWS and include_win_description:
+			return (text, html, description)
+		else:
+			return (text, html)
 
 	def clear(self):
 		"""Empty selection.
@@ -365,7 +373,7 @@ def set_with_rich_text(text, html):
 	SELECTION.set_with_rich_text(text, html)
 
 
-def get_with_rich_text():
+def get_with_rich_text(include_win_description=False):
 	"""Get the contents of the selection in plaintext and HTML formats.
 
 	Returns:
@@ -377,7 +385,7 @@ def get_with_rich_text():
 	global SELECTION
 	if SELECTION is None:
 		SELECTION = Selection()
-	return SELECTION.get_with_rich_text()
+	return SELECTION.get_with_rich_text(include_win_description)
 
 
 def clear():

klembord/winclipboard.py

@@ -1,8 +1,10 @@
 #!/usr/bin/env python3
 
-from collections import OrderedDict, ByteString
+from collections import OrderedDict
+from collections.abc import ByteString
 from ctypes import windll, create_unicode_buffer, memmove, c_uint, c_wchar
 from ctypes import c_void_p, c_bool, c_int, c_byte
+import re
 
 
 UNSUPPORTED = {
@@ -230,3 +232,31 @@ def wrap_html(self, fragment_str):
 
 
 		return wrapped
+
+
+start_html_RE = re.compile(r'(StartHTML):(-?\d+)')
+start_fragment_RE = re.compile(r'(StartFragment):(\d+)')
+end_fragment_RE = re.compile(r'(EndFragment):(\d+)')
+
+def parse_description(text):
+	''' Returns a dictionary of the description info.
+	Only StartFragment and EndFragment are guaranteed to exist and be integers.
+	'''
+	start_html = start_html_RE.search(text)
+	start_fragment = start_fragment_RE.search(text)
+	end_fragment = end_fragment_RE.search(text)
+
+	description = OrderedDict()
+
+	if start_html:
+		start_index = int(start_html.group(2))
+
+		if start_index > 0:
+			description_blob = text[:start_index]
+
+			description = OrderedDict( line.split(':', 1) for line in description_blob.splitlines() )
+
+	description[ start_fragment.group(1) ] = int(start_fragment.group(2))
+	description[ end_fragment.group(1) ] = int(end_fragment.group(2))
+
+	return description

klembord/xclipboard.py

@@ -4,7 +4,7 @@
 import sys
 from threading import Thread
 from queue import Queue, Empty
-from collections import ByteString
+from collections.abc import ByteString
 from traceback import print_exception
 from Xlib import X, display, Xatom
 from Xlib.protocol import event