Git issue 473: Emoji classified as letter

The values for GC:Assigned and GC:LC were flipped.
mrabarnett · Jul 8, 2022 · 5c9b260 · 5c9b260
1 parent 8a514bc
commit 5c9b260
Show file tree

Hide file tree

Showing 5 changed files with 13 additions and 9 deletions.
diff --git a/regex_3/_regex_unicode.c b/regex_3/_regex_unicode.c
@@ -2451,9 +2451,9 @@ RE_PropertyValue re_property_values[] = {
     {1226,  6,  36}, /* SEPARATOR */
     {1455,  6,  36}, /* Z */
     {1456,  6,  36}, /* Z& */
-    { 317,  6,  37}, /* CASEDLETTER */
-    { 820,  6,  37}, /* LC */
-    { 224,  6,  38}, /* ASSIGNED */
+    { 224,  6,  37}, /* ASSIGNED */
+    { 317,  6,  38}, /* CASEDLETTER */
+    { 820,  6,  38}, /* LC */
     {1092,  7,   0}, /* OTHER */
     {1439,  7,   0}, /* XX */
     { 445,  7,   1}, /* CONTROL */

diff --git a/regex_3/regex.py b/regex_3/regex.py
@@ -241,7 +241,7 @@
   "VERSION1", "X", "VERBOSE", "W", "WORD", "error", "Regex", "__version__",
   "__doc__", "RegexFlag"]
 
-__version__ = "2.5.115"
+__version__ = "2.5.116"
 
 # --------------------------------------------------------------------
 # Public interface.

diff --git a/regex_3/test_regex.py b/regex_3/test_regex.py
@@ -4313,6 +4313,10 @@ def test_hg_bugs(self):
         self.assertEqual(regex.match(r'(?a:\w)\w', 'd\N{CYRILLIC SMALL LETTER ZHE}').span(), (0, 2))
         self.assertEqual(regex.match(r'(?a:\w)(?u:\w)', 'd\N{CYRILLIC SMALL LETTER ZHE}').span(), (0, 2))
 
+        # Git issue 473: Emoji classified as letter
+        self.assertEqual(regex.match(r'^\p{LC}+$', '\N{SMILING CAT FACE WITH OPEN MOUTH}'), None)
+        self.assertEqual(regex.match(r'^\p{So}+$', '\N{SMILING CAT FACE WITH OPEN MOUTH}').span(), (0, 1))
+
     def test_fuzzy_ext(self):
         self.assertEqual(bool(regex.fullmatch(r'(?r)(?:a){e<=1:[a-z]}', 'e')),
           True)

diff --git a/setup.py b/setup.py
@@ -7,7 +7,7 @@
 
 setup(
     name='regex',
-    version='2022.6.2',
+    version='2022.7.9',
     description='Alternative regular expression module, to replace re.',
     long_description=long_description,
     long_description_content_type='text/x-rst',

diff --git a/tools/build_regex_unicode.py b/tools/build_regex_unicode.py
@@ -1398,7 +1398,7 @@ def make_key(value):
             val_list = list(unique(property['values'].values(), key=id))
 
             for value in sorted(val_list, key=lambda val: val['id']):
-                valueset.append(tuple(value['names']))
+                valueset.append((value['id'], tuple(value['names'])))
 
             valueset_id = valueset_dict.setdefault(tuple(valueset),
               len(valueset_dict))
@@ -1477,10 +1477,10 @@ def make_key(names):
 
         for valset, valset_id in sorted(valueset_dict.items(), key=lambda pair:
           pair[1]):
-            if valset_id == gc_valset_id:
-                valset = sorted(valset, key=make_key)
+            for val_id, names in valset:
+                if valset_id == gc_valset_id:
+                    names = sorted(names, key=make_key)
 
-            for val_id, names in enumerate(valset):
                 for name in names:
                     c_file.write('''    {{{:4}, {:2}, {:3}}}, /* {} */\n'''.format(strings_dict[munge(name)],
                       valset_id, val_id, munge(name)))