Added Bangla language support to num2word for Bangladesh. (#589)

* Feature:bangla number to word conversion library.

* bangla number to word conversion library resgister to num2word converter class.

* Test case for bangla number to word conversion.

* refactor and rename class for library compatibility.

* fix: isort and flake8 formating issue.

* fix: add more test to increase coverage.

---------

Co-authored-by: Mehedi Hasan Khondoker <mehedi.hasan@upaybd.com>
Co-authored-by: Marlon Rodriguez Garcia <47992153+mrodriguezg1991@users.noreply.github.com>

README.rst

@@ -80,6 +80,7 @@ Besides the numerical argument, there are two main optional arguments, ``to:`` a
 * ``ar`` (Arabic)
 * ``az`` (Azerbaijani)
 * ``be`` (Belarusian)
+* ``bn`` (Bangladeshi)
 * ``ce`` (Chechen)
 * ``cy`` (Welsh)
 * ``cz`` (Czech)

num2words/__init__.py

@@ -17,21 +17,22 @@
 
 from __future__ import unicode_literals
 
-from . import (lang_AM, lang_AR, lang_AZ, lang_BE, lang_CA, lang_CE, lang_CY,
-               lang_CZ, lang_DE, lang_DK, lang_EN, lang_EN_IN, lang_EN_NG,
-               lang_EO, lang_ES, lang_ES_CO, lang_ES_CR, lang_ES_GT,
-               lang_ES_NI, lang_ES_VE, lang_FA, lang_FI, lang_FR, lang_FR_BE,
-               lang_FR_CH, lang_FR_DZ, lang_HE, lang_HU, lang_ID, lang_IS,
-               lang_IT, lang_JA, lang_KN, lang_KO, lang_KZ, lang_LT, lang_LV,
-               lang_NL, lang_NO, lang_PL, lang_PT, lang_PT_BR, lang_RO,
-               lang_RU, lang_SK, lang_SL, lang_SR, lang_SV, lang_TE, lang_TET,
-               lang_TG, lang_TH, lang_TR, lang_UK, lang_VI)
+from . import (lang_AM, lang_AR, lang_AZ, lang_BE, lang_BN, lang_CA, lang_CE,
+               lang_CY, lang_CZ, lang_DE, lang_DK, lang_EN, lang_EN_IN,
+               lang_EN_NG, lang_EO, lang_ES, lang_ES_CO, lang_ES_CR,
+               lang_ES_GT, lang_ES_NI, lang_ES_VE, lang_FA, lang_FI, lang_FR,
+               lang_FR_BE, lang_FR_CH, lang_FR_DZ, lang_HE, lang_HU, lang_ID,
+               lang_IS, lang_IT, lang_JA, lang_KN, lang_KO, lang_KZ, lang_LT,
+               lang_LV, lang_NL, lang_NO, lang_PL, lang_PT, lang_PT_BR,
+               lang_RO, lang_RU, lang_SK, lang_SL, lang_SR, lang_SV, lang_TE,
+               lang_TET, lang_TG, lang_TH, lang_TR, lang_UK, lang_VI)
 
 CONVERTER_CLASSES = {
     'am': lang_AM.Num2Word_AM(),
     'ar': lang_AR.Num2Word_AR(),
     'az': lang_AZ.Num2Word_AZ(),
     'be': lang_BE.Num2Word_BE(),
+    'bn': lang_BN.Num2Word_BN(),
     'ca': lang_CA.Num2Word_CA(),
     'ce': lang_CE.Num2Word_CE(),
     'cy': lang_CY.Num2Word_CY(),

num2words/lang_BN.py

@@ -0,0 +1,203 @@
+# -*- coding: utf-8 -*-
+# Author: Mehedi Hasan Khondoker
+# Email: mehedihasankhondoker [at] gmail.com
+# Copyright (c) 2024, Mehedi Hasan Khondoker.  All Rights Reserved.
+
+# This library is build for Bangladesh format Number to Word conversion.
+# You are welcome as contributor to the library.
+
+# This library is free software; you can redistribute it and/or
+# modify it under the terms of the GNU Lesser General Public
+# License as published by the Free Software Foundation; either
+# version 2.1 of the License, or (at your option) any later version.
+# This library is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+# Lesser General Public License for more details.
+
+
+from decimal import Decimal
+
+RANKING = ['', 'প্রথম', 'দ্বিতীয়', 'তৃতীয়', 'চতুর্থ', 'পঞ্চম', 'ষষ্ঠ',
+           'সপ্তম', 'অষ্টম', 'নবম', 'দশম']  # pragma: no cover
+
+AKOK = ['', 'এক', 'দুই', 'তিন', 'চার', 'পাঁচ', 'ছয়',
+        'সাত', 'আট', 'নয়']  # pragma: no cover
+
+DOSOK = [
+    'দশ', 'এগারো', 'বারো', 'তেরো', 'চৌদ্দ', 'পনের',
+    'ষোল', 'সতের', 'আঠারো', 'উনিশ',
+    'বিশ', 'একুশ', 'বাইশ', 'তেইশ', 'চব্বিশ', 'পঁচিশ',
+    'ছাব্বিশ', 'সাতাশ', 'আটাশ', 'উনত্রিশ',
+    'ত্রিশ', 'একত্রিশ', 'বত্রিশ', 'তেত্রিশ', 'চৌত্রিশ', 'পঁইত্রিশ',
+    'ছত্রিশ', 'সাতত্রিশ', 'আটত্রিশ', 'উনচল্লিশ', 'চল্লিশ',
+    'একচল্লিশ', 'বিয়াল্লিশ', 'তেতাল্লিশ', 'চৌচল্লিশ',
+    'পঁয়তাল্লিশ', 'ছেচল্লিশ', 'সাতচল্লিশ', 'আটচল্লিশ', 'উনপঞ্চাশ',
+    'পঞ্চাশ', 'একান্ন', 'বাহান্ন', 'তিপ্পান্ন', 'চুয়ান্ন', 'পঞ্চান্ন',
+    'ছাপ্পান্ন', 'সাতান্ন', 'আটান্ন', 'উনষাট', 'ষাট',
+    'একষট্টি', 'বাষট্টি', 'তেষট্টি', 'চৌষট্টি', 'পঁয়ষট্টি',
+    'ছিষট্টি', 'সাতষট্টি', 'আটষট্টি', 'উনসত্তর', 'সত্তর',
+    'একাত্তর ', 'বাহাত্তর', 'তিয়াত্তর', 'চুয়াত্তর', 'পঁচাত্তর',
+    'ছিয়াত্তর', 'সাতাত্তর', 'আটাত্তর', 'উনআশি', 'আশি',
+    'একাশি', 'বিরাশি', 'তিরাশি', 'চুরাশি', 'পঁচাশি',
+    'ছিয়াশি', 'সাতাশি', 'আটাশি', 'উননব্বই', 'নব্বই',
+    'একানব্বই', 'বিরানব্বই', 'তিরানব্বই', 'চুরানব্বই', 'পঁচানব্বই',
+    'ছিয়ানব্বই', 'সাতানব্বই', 'আটানব্বই', 'নিরানব্বই'
+]  # pragma: no cover
+HAZAR = ' হাজার '  # pragma: no cover
+LAKH = ' লাখ '  # pragma: no cover
+KOTI = ' কোটি '  # pragma: no cover
+MAX_NUMBER = 9999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999999  # noqa: E501 # pragma: no cover
+
+
+class NumberTooLargeError(Exception):
+    """Custom exception for numbers that are too large."""
+    pass
+
+
+class Num2Word_BN:
+
+    @staticmethod
+    def str_to_number(number):
+        return abs(Decimal(str(number)))  # pragma: no cover
+
+    @staticmethod
+    def parse_number(number: Decimal):
+        dosomik = str(number - int(number)).split('.')[1:]
+        dosomik_str = ''.join(dosomik) if dosomik else 0
+        return int(number), int(dosomik_str)
+
+    @staticmethod
+    def parse_paisa(number: Decimal):
+        # 1-99 for paisa count so two digits are valid.
+        paisa = str(number - int(number)).split('.')[1:]
+        paisa_str = ''.join(paisa) if paisa else 0
+
+        # this is need, when we parse to decimal it removes trailing 0 .
+        if paisa_str:
+            paisa_str = str(int(paisa_str) * 100)[:2]
+        return int(number), int(paisa_str)
+
+    def _is_smaller_than_max_number(self, number):
+        if MAX_NUMBER >= number:
+            return True
+        raise NumberTooLargeError(f'Too Large number maximum '
+                                  f'value={MAX_NUMBER}')
+
+    def _dosomik_to_bengali_word(self, number):
+        word = ''
+        for i in str(number):
+            word += ' ' + AKOK[int(i)]
+        return word
+
+    def _number_to_bengali_word(self, number):
+        if number == 0:
+            return 'শূন্য'
+
+        words = ''
+
+        if number >= 10 ** 7:
+            words += self._number_to_bengali_word(number // 10 ** 7) + KOTI
+            number %= 10 ** 7
+
+        if number >= 10 ** 5:
+            words += self._number_to_bengali_word(number // 10 ** 5) + LAKH
+            number %= 10 ** 5
+
+        if number >= 1000:
+            words += self._number_to_bengali_word(number // 1000) + HAZAR
+            number %= 1000
+
+        if number >= 100:
+            words += AKOK[number // 100] + 'শত '
+            number %= 100
+
+        if 10 <= number <= 99:
+            words += DOSOK[number - 10] + ' '
+            number = 0
+
+        if 0 < number < 10:
+            words += AKOK[number] + ' '
+
+        return words.strip()
+
+    def to_currency(self, val):
+        """
+        This function represent a number to word in bangla taka and paisa.
+        example:
+        1 = এক টাকা,
+        101 = একশত এক টাকা,
+        9999.15 = নয় হাজার নয়শত নিরানব্বই টাকা পনের পয়সা
+        and so on.
+        """
+
+        dosomik_word = None
+        number = self.str_to_number(val)
+        number, decimal_part = self.parse_paisa(number)
+        self._is_smaller_than_max_number(number)
+
+        if decimal_part > 0:
+            dosomik_word = f' {self._number_to_bengali_word(decimal_part)} পয়সা'  # noqa: E501
+
+        words = f'{self._number_to_bengali_word(number)} টাকা'
+
+        if dosomik_word:
+            return (words + dosomik_word).strip()
+        return words.strip()
+
+    def to_cardinal(self, number):
+        """
+        This function represent a number to word in bangla.
+        example:
+        1 = এক,
+        101 = একশত এক,
+        9999 = নয় হাজার নয়শত নিরানব্বই
+        and so on.
+        """
+
+        dosomik_word = None
+        number = self.str_to_number(number)
+        number, decimal_part = self.parse_number(number)
+        self._is_smaller_than_max_number(number)
+
+        if decimal_part > 0:
+            dosomik_word = f' দশমিক{self._dosomik_to_bengali_word(decimal_part)}'  # noqa: E501
+
+        words = self._number_to_bengali_word(number)
+
+        if dosomik_word:
+            return (words + dosomik_word).strip()
+        return words.strip()
+
+    def to_ordinal(self, number):
+        return self.to_cardinal(number)
+
+    def to_ordinal_num(self, number):
+        """
+        This function represent a number to ranking in bangla.
+        example:
+        1 = প্রথম,
+        2 = দ্বিতীয়,
+        1001 = এক হাজার একতম
+        and so on.
+        """
+        self._is_smaller_than_max_number(number)
+
+        if number in range(1, 11):
+            return RANKING[number]
+        else:
+            rank = self.to_cardinal(int(abs(number)))
+            if rank.endswith('ত'):
+                return rank + 'ম'
+            return rank + 'তম'
+
+    def to_year(self, number):
+        """
+        This function represent a number to year in bangla.
+        example:
+        2002 = দুই হাজার দুই সাল,
+        2024 = দুই হাজার চব্বিশ সাল
+        and so on.
+        """
+        self._is_smaller_than_max_number(number)
+        return self.to_cardinal(int(abs(number))) + ' সাল'