Add family names

PyThaiNLP · Sep 12, 2020 · 04ff603 · 04ff603
1 parent 5b8ad72
commit 04ff603
Show file tree

Hide file tree

Showing 4 changed files with 27 additions and 1 deletion.
diff --git a/docs/api/corpus.rst b/docs/api/corpus.rst
@@ -19,6 +19,7 @@ Modules
 .. autofunction:: thai_words
 .. autofunction:: thai_syllables
 .. autofunction:: thai_negations
+.. autofunction:: thai_family_names
 .. autofunction:: thai_female_names
 .. autofunction:: thai_male_names
 .. autofunction:: pythainlp.corpus.conceptnet.edges

diff --git a/pythainlp/corpus/__init__.py b/pythainlp/corpus/__init__.py
@@ -18,6 +18,7 @@
     "get_corpus_path",
     "provinces",
     "remove",
+    "thai_family_names",
     "thai_female_names",
     "thai_male_names",
     "thai_negations",
@@ -86,6 +87,7 @@ def corpus_db_path() -> str:
 from pythainlp.corpus.common import (
     countries,
     provinces,
+    thai_family_names,
     thai_female_names,
     thai_male_names,
     thai_negations,

diff --git a/pythainlp/corpus/common.py b/pythainlp/corpus/common.py
@@ -6,6 +6,7 @@
 __all__ = [
     "countries",
     "provinces",
+    "thai_family_names",
     "thai_female_names",
     "thai_male_names",
     "thai_negations",
@@ -14,9 +15,10 @@
     "thai_words",
 ]
 
-from pythainlp.corpus import get_corpus
 from typing import Union
 
+from pythainlp.corpus import get_corpus
+
 _THAI_COUNTRIES = set()
 _THAI_COUNTRIES_FILENAME = "countries_th.txt"
 
@@ -36,6 +38,8 @@
 _THAI_NEGATIONS = set()
 _THAI_NEGATIONS_FILENAME = "negations_th.txt"
 
+_THAI_FAMLIY_NAMES = set()
+_THAI_FAMLIY_NAMES_FILENAME = "family_names_th.txt"
 _THAI_FEMALE_NAMES = set()
 _THAI_FEMALE_NAMES_FILENAME = "person_names_female_th.txt"
 _THAI_MALE_NAMES = set()
@@ -167,6 +171,22 @@ def thai_negations() -> frozenset:
     return _THAI_NEGATIONS
 
 
+def thai_family_names() -> frozenset:
+    """
+    Return a frozenset of Thai family names
+    \n(See: `dev/pythainlp/corpus/family_names_th.txt\
+    <https://github.com/PyThaiNLP/pythainlp/blob/dev/pythainlp/corpus/family_names_th.txt>`_)
+
+    :return: :class:`frozenset` containing Thai family names.
+    :rtype: :class:`frozenset`
+    """
+    global _THAI_FAMLIY_NAMES
+    if not _THAI_FAMLIY_NAMES:
+        _THAI_FAMLIY_NAMES = get_corpus(_THAI_FAMLIY_NAMES_FILENAME)
+
+    return _THAI_FAMLIY_NAMES
+
+
 def thai_female_names() -> frozenset:
     """
     Return a frozenset of Thai female names

diff --git a/tests/test_corpus.py b/tests/test_corpus.py
@@ -12,6 +12,7 @@
     get_corpus_path,
     provinces,
     remove,
+    thai_family_names,
     thai_female_names,
     thai_male_names,
     thai_negations,
@@ -41,6 +42,8 @@ def test_corpus(self):
         self.assertEqual(
             len(provinces(details=False)), len(provinces(details=True))
         )
+        self.assertIsInstance(thai_family_names(), frozenset)
+        self.assertIsInstance(list(thai_family_names())[0], str)
         self.assertIsInstance(thai_female_names(), frozenset)
         self.assertIsInstance(thai_male_names(), frozenset)