Merge pull request #961 from kempei/patch-1

fixed URIRef including native unicode characters

2 files changed, 5 insertions, 5 deletions

diff --git a/rdflib/term.py b/rdflib/term.py
index 1a75d91a..3a8ce798 100644
--- a/rdflib/term.py
+++ b/rdflib/term.py
@@ -51,6 +51,7 @@ import xml.dom.minidom
 from datetime import date, time, datetime, timedelta
 from re import sub, compile
 from collections import defaultdict
+from unicodedata import category
 
 from isodate import parse_time, parse_date, parse_datetime, Duration, parse_duration, duration_isoformat
 
@@ -74,10 +75,7 @@ _invalid_uri_chars = '<>" {}|\\^`'
 
 
 def _is_valid_uri(uri):
-    for c in _invalid_uri_chars:
-        if c in uri:
-            return False
-    return True
+    return all(map(lambda c: ord(c) > 256 or not c in _invalid_uri_chars, uri))
 
 
 _lang_tag_regex = compile('^[a-zA-Z]+(?:-[a-zA-Z0-9]+)*$')
diff --git a/rdflib/util.py b/rdflib/util.py
index f0c6207d..1789aa70 100644
--- a/rdflib/util.py
+++ b/rdflib/util.py
@@ -156,7 +156,9 @@ def from_n3(s, default=None, backend=None, nsm=None):
     if not s:
         return default
     if s.startswith('<'):
-        return URIRef(s[1:-1])
+        # Hack: this should correctly handle strings with either native unicode
+        # characters, or \u1234 unicode escapes.
+        return URIRef(s[1:-1].encode("raw-unicode-escape").decode("unicode-escape"))
     elif s.startswith('"'):
         if s.startswith('"""'):
             quotes = '"""'