Merge pull request #956 from bact/fix-maiyamok

Use common warn_deprecation

Author: bact

pythainlp/cls/__init__.py

@@ -5,12 +5,10 @@
 pythainlp.cls
 Depreciated. Use pythainlp.classify instead.
 """
-import warnings
 
 __all__ = ["GzipModel"]
 
 from pythainlp.classify.param_free import GzipModel
+from pythainlp.tools import warn_deprecation
 
-warnings.warn(
-    "Deprecated: Use pythainlp.classify instead.", DeprecationWarning
-)
+warn_deprecation("pythainlp.cls", "pythainlp.classify", "5.1")

pythainlp/corpus/common.py

@@ -23,10 +23,10 @@
     "thai_wsd_dict",
 ]
 
-import warnings
 from typing import FrozenSet, List, Union
 
 from pythainlp.corpus import get_corpus, get_corpus_as_is, get_corpus_path
+from pythainlp.tools import warn_deprecation
 
 _THAI_COUNTRIES: FrozenSet[str] = frozenset()
 _THAI_COUNTRIES_FILENAME = "countries_th.txt"
@@ -336,7 +336,11 @@ def thai_synonyms() -> dict:
 
 
 def thai_synonym() -> dict:
-    warnings.warn("Deprecated: Use thai_synonyms() instead.", DeprecationWarning)
+    warn_deprecation(
+        "pythainlp.corpus.thai_synonym",
+        "pythainlp.corpus.thai_synonyms",
+        "5.1",
+    )
     return thai_synonyms()
 
 

pythainlp/tools/__init__.py

@@ -6,8 +6,10 @@
     "get_full_data_path",
     "get_pythainlp_data_path",
     "get_pythainlp_path",
+    "warn_deprecation",
 ]
 
+from pythainlp.tools.core import warn_deprecation
 from pythainlp.tools.path import (
     PYTHAINLP_DEFAULT_DATA_DIR,
     get_full_data_path,

pythainlp/tools/core.py

@@ -0,0 +1,30 @@
+# -*- coding: utf-8 -*-
+# SPDX-FileCopyrightText: 2016-2024 PyThaiNLP Project
+# SPDX-License-Identifier: Apache-2.0
+"""
+Generic support functions for PyThaiNLP.
+"""
+
+import warnings
+
+
+def warn_deprecation(
+    deprecated_func: str,
+    replacing_func: str = "",
+    version: str = "",
+):
+    """
+    Warn about the deprecation of a function.
+
+    :param str deprecated_func: Name of the deprecated function.
+    :param str replacing_func: Name of the function to use instead (optional).
+    :param str version: PyThaiNLP version in which the function will be deprecated (optional).
+    """
+    if version:
+        version = f"PyThaiNLP {version}"
+    else:
+        version = "a future release"
+    message = f"The '{deprecated_func}' function is deprecated and will be removed in {version}."
+    if replacing_func:
+        message += f" Please use '{replacing_func}' instead."
+    warnings.warn(message, DeprecationWarning, stacklevel=2)

pythainlp/util/__init__.py

@@ -19,6 +19,7 @@
     "display_thai_char",
     "emoji_to_thai",
     "eng_to_thai",
+    "expand_maiyamok",
     "find_keyword",
     "ipa_to_rtgs",
     "is_native_thai",
@@ -97,6 +98,7 @@
     remove_tonemark,
     remove_zw,
     reorder_vowels,
+    expand_maiyamok,
 )
 from pythainlp.util.numtoword import bahttext, num_to_thaiword
 from pythainlp.util.phoneme import ipa_to_rtgs, nectec_to_ipa, remove_tone_ipa
@@ -117,7 +119,7 @@
 from pythainlp.util.trie import Trie, dict_trie
 from pythainlp.util.wordtonum import text_to_num, thaiword_to_num, words_to_num
 
-# syllable and pronounce have to be imported last
+# sound_syllable and pronounce have to be imported last
 from pythainlp.util.syllable import (
     sound_syllable,
     syllable_length,

pythainlp/util/normalize.py

@@ -14,6 +14,7 @@
 from pythainlp import thai_lead_vowels as lead_v
 from pythainlp import thai_tonemarks as tonemarks
 from pythainlp.tokenize import word_tokenize
+from pythainlp.tools import warn_deprecation
 
 _DANGLING_CHARS = f"{above_v}{below_v}{tonemarks}\u0e3a\u0e4c\u0e4d\u0e4e"
 _RE_REMOVE_DANGLINGS = re.compile(f"^[{_DANGLING_CHARS}]+")
@@ -249,12 +250,13 @@ def normalize(text: str) -> str:
     return text
 
 
-def maiyamok(sent: Union[str, List[str]]) -> List[str]:
+def expand_maiyamok(sent: Union[str, List[str]]) -> List[str]:
     """
-    Thai MaiYaMok
+    Expand Maiyamok.
+
+    Maiyamok (ๆ) (Unicode U+0E46) is a Thai character indicating word
+    repetition. This function preprocesses Thai text by expanding Maiyamok
 
-    MaiYaMok (ๆ) is the mark of duplicate word in Thai language.
-    This function is preprocessing MaiYaMok in Thai sentence.
 
     :param Union[str, List[str]] sent: input sentence (list or str)
     :return: list of words
@@ -265,15 +267,12 @@ def maiyamok(sent: Union[str, List[str]]) -> List[str]:
 
         from pythainlp.util import maiyamok
 
-        maiyamok("เด็กๆชอบไปโรงเรียน")
-        # output: ['เด็ก', 'เด็ก', 'ชอบ', 'ไป', 'โรงเรียน']
-
-        maiyamok(["ทำไม", "คน", "ดี", " ", "ๆ", "ๆ", " ", "ถึง", "ทำ", "ไม่ได้"])
-        # output: ['ทำไม', 'คน', 'ดี', 'ดี', 'ดี', ' ', 'ถึง', 'ทำ', 'ไม่ได้']
+        maiyamok("เด็กๆกิน")
+        # output: ['เด็ก', 'เด็ก', 'กิน']
     """
     if isinstance(sent, str):
         sent = word_tokenize(sent)
-    _list_word = []
+    _list_word: list[str] = []
     i = 0
     for j, text in enumerate(sent):
         if text.isspace() and "ๆ" in sent[j + 1]:
@@ -292,3 +291,28 @@ def maiyamok(sent: Union[str, List[str]]) -> List[str]:
         _list_word.append(text)
         i += 1
     return _list_word
+
+
+def maiyamok(sent: Union[str, List[str]]) -> List[str]:
+    """
+    Expand Maiyamok.
+
+    Maiyamok (ๆ) (Unicode U+0E46) is a Thai character indicating word
+    repetition. This function preprocesses Thai text by expanding Maiyamok
+
+    :param Union[str, List[str]] sent: input sentence (list or str)
+    :return: list of words
+    :rtype: List[str]
+
+    :Example:
+    ::
+
+        from pythainlp.util import maiyamok
+
+        maiyamok("เด็กๆกิน")
+        # output: ['เด็ก', 'เด็ก', 'กิน']
+    """
+    warn_deprecation(
+        "pythainlp.util.maiyamok", "pythainlp.util.expand_maiyamok"
+    )
+    return expand_maiyamok(sent)

pythainlp/util/thaiwordcheck.py

@@ -1,18 +1,16 @@
 # -*- coding: utf-8 -*-
 # SPDX-FileCopyrightText: 2016-2024 PyThaiNLP Project
 # SPDX-License-Identifier: Apache-2.0
-import warnings
+from pythainlp.tools import warn_deprecation
 
 
 def is_native_thai(word: str) -> bool:
-    warnings.warn(
-        """
-        pythainlp.util.is_native_thai has been renamed to \
-            pythainlp.morpheme.is_native_thai.
-        This function will be removed in PyThaiNLP 5.1.
-        """,
-        DeprecationWarning,
+    warn_deprecation(
+        "pythainlp.util.is_native_thai",
+        "pythainlp.morpheme.is_native_thai",
+        "5.1",
     )
+
     from pythainlp.morpheme import is_native_thai as check
 
     return check(word)

tests/test_util.py

@@ -531,7 +531,7 @@ def test_normalize(self):
         self.assertEqual(remove_zw("\u200bกา"), "กา")
         self.assertEqual(remove_zw("กา\u200b\u200c\u200b"), "กา")
 
-        # maiyamok
+        # expand maiyamok
         self.assertEqual(
             maiyamok("เด็กๆชอบไปโรงเรียน"),
             ["เด็ก", "เด็ก", "ชอบ", "ไป", "โรงเรียน"],