a g@sdZddlZddlmZddlmZmZmZmZm Z m Z m Z m Z m Z mZddlmZddlmZejejejZdZdZeed d d Zdeeed ddZefeeed ddZeed ddZeeeefdddZ ee!dddZ"dS)z Check if it is Thai text N)Tuple) thai_lead_vowelsthai_follow_vowelsthai_above_vowelsthai_below_vowelsthai_consonants thai_vowelsthai_tonemarks thai_signs thai_digitsthai_punctuations) pronunciate) tone_detectorii)chreturncCs(t|}t|krtkr$nndSdS)uCheck if a character is a Thai character. :param ch: input character :type ch: str :return: True if ch is a Thai character, otherwise False. :rtype: bool :Example: :: from pythainlp.util import isthaichar isthaichar("ก") # THAI CHARACTER KO KAI # output: True isthaichar("๕") # THAI DIGIT FIVE # output: True TF)ord_TH_FIRST_CHAR_ASCII_TH_LAST_CHAR_ASCII)rZch_valr=/usr/local/lib/python3.9/dist-packages/pythainlp/util/thai.py isthaicharsr.)text ignore_charsrcCs,|sd}|D]}||vr t|s dSq dS)uCheck if every character in a string is a Thai character. :param text: input text :type text: str :param ignore_chars: characters to be ignored, defaults to "." :type ignore_chars: str, optional :return: True if every character in the input string is Thai, otherwise False. :rtype: bool :Example: :: from pythainlp.util import isthai isthai("กาลเวลา") # output: True isthai("กาลเวลา.") # output: True isthai("กาล-เวลา") # output: False isthai("กาล-เวลา +66", ignore_chars="01234567890+-.,") # output: True FT)r)rrrrrristhai7s rcCsr|rt|tsdS|sd}d}d}|D]&}||vr<|d7}q&t|r&|d7}q&t||}|dkrfdS||dS)uFind proportion of Thai characters in a given text :param text: input text :type text: str :param ignore_chars: characters to be ignored, defaults to whitespace,\ digits, and punctuation marks. :type ignore_chars: str, optional :return: proportion of Thai characters in the text (percentage) :rtype: float :Example: :: from pythainlp.util import countthai countthai("ไทยเอ็นแอลพี 3.0") # output: 100.0 countthai("PyThaiNLP 3.0") # output: 0.0 countthai("ใช้งาน PyThaiNLP 3.0") # output: 40.0 countthai("ใช้งาน PyThaiNLP 3.0", ignore_chars="") # output: 30.0 grrd) isinstancestrrlen)rrZnum_thaiZ num_ignorerZ num_countrrr countthai]s   r!cCs(|tvs|tvs|dvr d|S|SdS)u>Prefix an underscore (_) to a high-position vowel or a tone mark, to ease readability. :param ch: input character :type ch: str :return: "_" + ch :rtype: str :Example: :: from pythainlp.util import display_thai_char display_thai_char("้") # output: "_้" u ำ์ํ๎_N)rr )rrrrdisplay_thai_charsr#)wordrcCst|d}dd|DS)uo Thai tone detector for word. It uses pythainlp.transliterate.pronunciate for converting word to pronunciation. :param str word: Thai word. :return: Thai pronunciation with tones in each syllable. (l, m, h, r, f or empty if it cannot be detected) :rtype: Tuple[str, str] :Example: :: from pythainlp.util import thai_word_tone_detector print(thai_word_tone_detector("คนดี")) # output: [('คน', 'm'), ('ดี', 'm')] print(thai_word_tone_detector("มือถือ")) # output: [('มือ', 'm'), ('ถือ', 'r')] -cSs g|]}|t|ddfqS)uหฺuห)rreplace).0irrr z+thai_word_tone_detector..)r split)r$Z _pronunciaterrrthai_word_tone_detectorsr,)rrc CsBdddddddddddd }|D]}|tvr>|dd7<|tvrX|dd7<q |tvrr|dd7<q |tvr|dd7<q |tvr|dd7<q |tvr|d d7<q |tvr|d d7<q |tvr|d d7<q |tvr|d d7<q |t vr,|d d7<q |dd7<q |S)u7 Count Thai characters by type This function will give you numbers of Thai characters by type (consonants, vowels, lead_vowels, follow_vowels, above_vowels, below_vowels, tonemarks, signs, thai_digits, punctuations, non_thai) :param str text: Text :return: Dict with numbers of Thai characters by type :rtype: dict :Example: :: from pythainlp.util import count_thai_chars count_thai_chars("ทดสอบภาษาไทย") # output: { # 'vowels': 3, # 'lead_vowels': 1, # 'follow_vowels': 2, # 'above_vowels': 0, # 'below_vowels': 0, # 'consonants': 9, # 'tonemarks': 0, # 'signs': 0, # 'thai_digits': 0, # 'punctuations': 0, # 'non_thai': 0 # } r) vowels lead_vowels follow_vowels above_vowels below_vowels consonants tonemarkssignsr punctuationsnon_thair-rr.r/r0r1r2r3r4r r5r6) rrrrrrr r r r )r_dictcrrrcount_thai_charssF!   r9)r)#__doc__stringtypingrZ pythainlprrrrrrr r r r Zpythainlp.transliterater Zpythainlp.util.syllabler whitespacedigits punctuationZ_DEFAULT_IGNORE_CHARSrrrboolrrfloatr!r#r,dictr9rrrrs 0  &3