Update word_detokenize rule

- Fixed space bug

Author: wannaphong

pythainlp/tokenize/core.py

@@ -81,7 +81,9 @@ def word_detokenize(segments: Union[List[List[str]], List[str]], output: str = "
                         _list_sents.append(" ")
                         _add_index.append(j)
                     else:
-                        pass
+                        _add_index.append(j)
+                elif w.isspace():
+                    _add_index.append(j)
                 elif j-1 in _add_index:
                     _list_sents.append(" ")
             _list_sents.append(w)

tests/test_tokenize.py

@@ -657,6 +657,12 @@ def test_word_detokenize(self):
             ),
             "ผมเลี้ยง 5 10 ตัว ๆ คนดี"
         )
+        self.assertEqual(
+            word_detokenize(
+                ["ผม", "เลี้ยง", "5", "ตัว", " ", "ๆ", "คน", "ดี"]
+            ),
+            "ผมเลี้ยง 5 ตัว ๆ คนดี"
+        )
         self.assertTrue(
             isinstance(word_detokenize(["ผม", "เลี้ยง", "5", "ตัว"]), str)
         )