Handy module for quoting and unquoting

1 files changed, 64 insertions, 0 deletions

diff --git a/paste/util/quoting.py b/paste/util/quoting.py
new file mode 100644
index 0000000..db150d7
--- /dev/null
+++ b/paste/util/quoting.py
@@ -0,0 +1,64 @@
+import cgi
+import htmlentitydefs
+import urllib
+import re
+
+__all__ = ['html_quote', 'html_unquote', 'url_quote', 'url_unquote']
+
+default_encoding = 'UTF-8'
+
+def html_quote(v, encoding=None):
+    r"""
+    Quote the value (turned to a string) as HTML.  This quotes <, >,
+    and quotes:
+
+    >>> html_quote(1)
+    '1'
+    >>> html_quote(None)
+    ''
+    >>> html_quote('<hey!>')
+    '&lt;hey!&gt;'
+    >>> html_quote(u'\u1029')
+    '\xe1\x80\xa9'
+    """
+    encoding = encoding or default_encoding
+    if v is None:
+        return ''
+    elif isinstance(v, str):
+        return cgi.escape(v, 1)
+    elif isinstance(v, unicode):
+        return cgi.escape(v.encode(encoding), 1)
+    else:
+        return cgi.escape(unicode(v).encode(encoding), 1)
+
+_unquote_re = re.compile(r'&([a-zA-Z]+);')
+def _entity_subber(match, name2c=htmlentitydefs.name2codepoint):
+    code = name2c.get(match.group(1))
+    if code:
+        return unichr(code)
+    else:
+        return match.group(0)
+
+def html_unquote(s, encoding=None):
+    r"""
+    Decode the value.
+
+    >>> html_unquote('&lt;hey&nbsp;you&gt;')
+    u'<hey\xa0you>'
+    >>> html_unquote('')
+    ''
+    >>> html_unquote('&blahblah;')
+    u'&blahblah;'
+    >>> html_unquote('\xe1\x80\xa9')
+    u'\u1029'
+    """
+    if isinstance(s, str):
+        s = s.decode(encoding or default_encoding)
+    return _unquote_re.sub(_entity_subber, s)
+
+url_quote = urllib.quote
+url_unquote = urllib.unquote
+
+if __name__ == '__main__':
+    import doctest
+    doctest.testmod()