Soundex functions handle empty or None case

Author: bact

pythainlp/soundex/lk82.py

@@ -28,6 +28,9 @@ def lk82(text):
     :param str text: Thai word
     :return: LK82 soundex
     """
+    if not text:
+        return ""
+
     res = []
     text = _RE_1.sub("", text)  # 4.ลบวรรณยุกต์
     text = _RE_2.sub("", text)  # 4.ลบตัวการันต์

pythainlp/soundex/metasound.py

@@ -34,6 +34,9 @@ def metasound(text, length=4):
         metasound("รักษ์")  # 'ร100'
         metasound("บูรณการ", 5))  # 'บ5515'
     """
+    if not text:
+        return ""
+
     # keep only consonants and thanthakhat
     chars = []
     for ch in text:

pythainlp/soundex/udom83.py

@@ -37,6 +37,9 @@ def udom83(text):
     :return: Udom83 soundex
     """
 
+    if not text:
+        return ""
+
     text = _RE_1.sub("ัน\\1", text)
     text = _RE_2.sub("ั\\1", text)
     text = _RE_3.sub("ัน\\1", text)

tests/__init__.py

@@ -34,7 +34,7 @@
 from pythainlp.rank import rank
 from pythainlp.romanization import romanize
 from pythainlp.sentiment import sentiment
-from pythainlp.soundex import lk82, metasound, udom83
+from pythainlp.soundex import lk82, metasound, soundex, udom83
 from pythainlp.spell import correct, spell
 from pythainlp.summarize import summarize
 from pythainlp.tag import pos_tag, pos_tag_sents
@@ -233,13 +233,24 @@ def test_sentiment(self):
     # ### pythainlp.soundex
 
     def test_soundex(self):
+        self.assertIsNotNone(soundex("a", engine="lk82"))
+        self.assertIsNotNone(soundex("a", engine="udom83"))
+        self.assertIsNotNone(soundex("a", engine="metasound"))
+        self.assertIsNotNone(soundex("a", engine="XXX"))
+
         self.assertEqual(lk82("รถ"), "ร3000")
+        self.assertIsNotNone(lk82("เกาะกูร์"))
+        self.assertEqual(lk82(""), "")
+
         self.assertEqual(udom83("รถ"), "ร800000")
+        self.assertEqual(udom83(None), "")
+
         self.assertEqual(metasound("บูรณะ"), "บ550")
         self.assertEqual(metasound("คน"), "ค500")
         self.assertEqual(metasound("คนA"), "ค500")
         self.assertEqual(metasound("ดา"), "ด000")
         self.assertEqual(metasound("รักษ์"), metasound("รัก"))
+        self.assertEqual(metasound(""), "")
 
     # ### pythainlp.spell