Merge pull request #3007 from kontur/name-sort-encode-warning-only

Make NameRecord comparison not fail on encoding errors #3006

Lib/fontTools/ttLib/tables/_n_a_m_e.py

@@ -610,32 +610,38 @@ class NameRecord(object):
             return NotImplemented
 
         try:
-            # implemented so that list.sort() sorts according to the spec.
             selfTuple = (
                 self.platformID,
                 self.platEncID,
                 self.langID,
                 self.nameID,
-                self.toBytes(),
             )
             otherTuple = (
                 other.platformID,
                 other.platEncID,
                 other.langID,
                 other.nameID,
-                other.toBytes(),
             )
-            return selfTuple < otherTuple
+        except AttributeError:
-        except (UnicodeEncodeError, AttributeError):
             # This can only happen for
             # 1) an object that is not a NameRecord, or
             # 2) an unlikely incomplete NameRecord object which has not been
-            #    fully populated, or
+            #    fully populated
-            # 3) when all IDs are identical but the strings can't be encoded
-            #    for their platform encoding.
-            # In all cases it is best to return NotImplemented.
             return NotImplemented
 
+        try:
+            # Include the actual NameRecord string in the comparison tuples
+            selfTuple = selfTuple + (self.toBytes(),)
+            otherTuple = otherTuple + (other.toBytes(),)
+        except UnicodeEncodeError as e:
+            # toBytes caused an encoding error in either of the two, so content
+            # to sorting based on IDs only
+            log.error("NameRecord sorting failed to encode: %s" % e)
+
+        # Implemented so that list.sort() sorts according to the spec by using
+        # the order of the tuple items and their comparison
+        return selfTuple < otherTuple
+
     def __repr__(self):
         return "<NameRecord NameID=%d; PlatformID=%d; LanguageID=%d>" % (
             self.nameID,

Tests/ttLib/tables/_n_a_m_e_test.py

@@ -72,15 +72,34 @@ class NameTableTest(unittest.TestCase):
         ]
         table.compile(None)
 
-    def test_names_sort_bytes_str_encoding_error(self):
+    def test_names_sort_attributes(self):
         table = table__n_a_m_e()
+        # Create an actual invalid NameRecord object
+        broken = makeName("Test", 25, 3, 1, 0x409)
+        delattr(broken, "platformID")
         table.names = [
-            makeName("Test寬", 25, 1, 0, 0),
+            makeName("Test", 25, 3, 1, 0x409),
-            makeName("Test鬆鬆", 25, 1, 0, 0),
+            broken,
         ]
+        # Sorting these two is impossible, expect an error to be raised
         with self.assertRaises(TypeError):
             table.names.sort()
 
+    def test_names_sort_encoding(self):
+        """
+        Confirm that encoding errors in name table strings do not prevent at
+        least sorting by other IDs
+        """
+        table = table__n_a_m_e()
+        table.names = [
+            makeName("Mac Unicode 寬 encodes ok", 25, 3, 0, 0x409),
+            makeName("Win Latin 寬 fails to encode", 25, 1, 0, 0),
+        ]
+        table.names.sort()
+        # Encoding errors or not, sort based on other IDs nonetheless
+        self.assertEqual(table.names[0].platformID, 1)
+        self.assertEqual(table.names[1].platformID, 3)
+
     def test_addName(self):
         table = table__n_a_m_e()
         nameIDs = []