Added functionality (plus tests & docs): escaped whitespace in URI contexts.

git-svn-id: http://svn.code.sf.net/p/docutils/code/trunk/docutils@7998 929543f6-e4f2-0310-98a6-ba3bd3dd1d04

1 files changed, 12 insertions, 1 deletions

diff --git a/docutils/utils/__init__.py b/docutils/utils/__init__.py
index 57ce83abb..919722943 100644
--- a/docutils/utils/__init__.py
+++ b/docutils/utils/__init__.py
@@ -13,6 +13,7 @@ import sys
 import os
 import os.path
 import re
+import itertools
 import warnings
 import unicodedata
 from docutils import ApplicationError, DataError
@@ -575,7 +576,7 @@ def escape2null(text):
         parts.append('\x00' + text[found+1:found+2])
         start = found + 2               # skip character after escape
 
-def unescape(text, restore_backslashes=False):
+def unescape(text, restore_backslashes=False, respect_whitespace=False):
     """
     Return a string with nulls removed or restored to backslashes.
     Backslash-escaped spaces are also removed.
@@ -587,6 +588,16 @@ def unescape(text, restore_backslashes=False):
             text = ''.join(text.split(sep))
         return text
 
+def split_escaped_whitespace(text):
+    """
+    Split `text` on escaped whitespace (null+space or null+newline).
+    Return a list of strings.
+    """
+    strings = text.split('\x00 ')
+    strings = [string.split('\x00\n') for string in strings]
+    # flatten list of lists of strings to list of strings:
+    return list(itertools.chain(*strings))
+
 def strip_combining_chars(text):
     if isinstance(text, str) and sys.version_info < (3,0):
         return text