Add support for compact decimal formats (#909)

6 files changed, 105 insertions, 3 deletions

diff --git a/babel/core.py b/babel/core.py
index 9393c23..220cbaf 100644
--- a/babel/core.py
+++ b/babel/core.py
@@ -565,6 +565,18 @@ class Locale:
         return self._data['decimal_formats']
 
     @property
+    def compact_decimal_formats(self):
+        """Locale patterns for compact decimal number formatting.
+
+        .. note:: The format of the value returned may change between
+                  Babel versions.
+
+        >>> Locale('en', 'US').compact_decimal_formats["short"]["one"]["1000"]
+        <NumberPattern u'0K'>
+        """
+        return self._data['compact_decimal_formats']
+
+    @property
     def currency_formats(self):
         """Locale patterns for currency number formatting.
 
diff --git a/babel/numbers.py b/babel/numbers.py
index b8971bc..192e3ed 100644
--- a/babel/numbers.py
+++ b/babel/numbers.py
@@ -425,6 +425,63 @@ def format_decimal(
         number, locale, decimal_quantization=decimal_quantization, group_separator=group_separator)
 
 
+def format_compact_decimal(number, *, format_type="short", locale=LC_NUMERIC, fraction_digits=0):
+    u"""Return the given decimal number formatted for a specific locale in compact form.
+
+    >>> format_compact_decimal(12345, format_type="short", locale='en_US')
+    u'12K'
+    >>> format_compact_decimal(12345, format_type="long", locale='en_US')
+    u'12 thousand'
+    >>> format_compact_decimal(12345, format_type="short", locale='en_US', fraction_digits=2)
+    u'12.35K'
+    >>> format_compact_decimal(1234567, format_type="short", locale="ja_JP")
+    u'123万'
+    >>> format_compact_decimal(2345678, format_type="long", locale="mk")
+    u'2 милиони'
+    >>> format_compact_decimal(21098765, format_type="long", locale="mk")
+    u'21 милион'
+
+    :param number: the number to format
+    :param format_type: Compact format to use ("short" or "long")
+    :param locale: the `Locale` object or locale identifier
+    :param fraction_digits: Number of digits after the decimal point to use. Defaults to `0`.
+    """
+    locale = Locale.parse(locale)
+    number, format = _get_compact_format(number, format_type, locale, fraction_digits)
+    pattern = parse_pattern(format)
+    return pattern.apply(number, locale, decimal_quantization=False)
+
+
+def _get_compact_format(number, format_type, locale, fraction_digits=0):
+    """Returns the number after dividing by the unit and the format pattern to use.
+    The algorithm is described here:
+    https://www.unicode.org/reports/tr35/tr35-45/tr35-numbers.html#Compact_Number_Formats.
+    """
+    format = None
+    compact_format = locale.compact_decimal_formats[format_type]
+    for magnitude in sorted([int(m) for m in compact_format["other"]], reverse=True):
+        if abs(number) >= magnitude:
+            # check the pattern using "other" as the amount
+            format = compact_format["other"][str(magnitude)]
+            pattern = parse_pattern(format).pattern
+            # if the pattern is "0", we do not divide the number
+            if pattern == "0":
+                break
+            # otherwise, we need to divide the number by the magnitude but remove zeros
+            # equal to the number of 0's in the pattern minus 1
+            number = number / (magnitude / (10 ** (pattern.count("0") - 1)))
+            # round to the number of fraction digits requested
+            number = round(number, fraction_digits)
+            # if the remaining number is singular, use the singular format
+            plural_form = locale.plural_form(abs(number))
+            plural_form = plural_form if plural_form in compact_format else "other"
+            format = compact_format[plural_form][str(magnitude)]
+            break
+    if format is None:  # Did not find a format, fall back.
+        format = locale.decimal_formats.get(None)
+    return number, format
+
+
 class UnknownCurrencyFormatError(KeyError):
     """Exception raised when an unknown currency format is requested."""
 
diff --git a/docs/api/numbers.rst b/docs/api/numbers.rst
index f9b0833..eac5692 100644
--- a/docs/api/numbers.rst
+++ b/docs/api/numbers.rst
@@ -13,6 +13,8 @@ Number Formatting
 
 .. autofunction:: format_decimal
 
+.. autofunction:: format_compact_decimal
+
 .. autofunction:: format_currency
 
 .. autofunction:: format_percent
diff --git a/docs/numbers.rst b/docs/numbers.rst
index ed3b60f..cbe05cd 100644
--- a/docs/numbers.rst
+++ b/docs/numbers.rst
@@ -12,7 +12,7 @@ the ``babel.numbers`` module:
 
 .. code-block:: pycon
 
-    >>> from babel.numbers import format_number, format_decimal, format_percent
+    >>> from babel.numbers import format_number, format_decimal, format_compact_decimal, format_percent
 
 Examples:
 
diff --git a/scripts/import_cldr.py b/scripts/import_cldr.py
index 7fc8538..92dd272 100755
--- a/scripts/import_cldr.py
+++ b/scripts/import_cldr.py
@@ -770,8 +770,6 @@ def parse_decimal_formats(data, tree):
 
                     # These are mapped into a `compact_decimal_formats` dictionary
                     # with the format {length: {count: {multiplier: pattern}}}.
-
-                    # TODO: Add support for formatting them.
                     compact_decimal_formats = data.setdefault('compact_decimal_formats', {})
                     length_map = compact_decimal_formats.setdefault(length_type, {})
                     length_count_map = length_map.setdefault(pattern_el.attrib['count'], {})
diff --git a/tests/test_numbers.py b/tests/test_numbers.py
index 78f7797..1b955c9 100644
--- a/tests/test_numbers.py
+++ b/tests/test_numbers.py
@@ -121,6 +121,39 @@ class FormatDecimalTestCase(unittest.TestCase):
         assert numbers.format_currency(101299.98, 'EUR', locale='en_US', group_separator=True, format_type='name') == u'101,299.98 euros'
         assert numbers.format_percent(251234.1234, locale='sv_SE', group_separator=True) == u'25\xa0123\xa0412\xa0%'
 
+    def test_compact(self):
+        assert numbers.format_compact_decimal(1, locale='en_US', format_type="short") == u'1'
+        assert numbers.format_compact_decimal(999, locale='en_US', format_type="short") == u'999'
+        assert numbers.format_compact_decimal(1000, locale='en_US', format_type="short") == u'1K'
+        assert numbers.format_compact_decimal(9000, locale='en_US', format_type="short") == u'9K'
+        assert numbers.format_compact_decimal(9123, locale='en_US', format_type="short", fraction_digits=2) == u'9.12K'
+        assert numbers.format_compact_decimal(10000, locale='en_US', format_type="short") == u'10K'
+        assert numbers.format_compact_decimal(10000, locale='en_US', format_type="short", fraction_digits=2) == u'10K'
+        assert numbers.format_compact_decimal(1000000, locale='en_US', format_type="short") == u'1M'
+        assert numbers.format_compact_decimal(9000999, locale='en_US', format_type="short") == u'9M'
+        assert numbers.format_compact_decimal(9000900099, locale='en_US', format_type="short", fraction_digits=5) == u'9.0009B'
+        assert numbers.format_compact_decimal(1, locale='en_US', format_type="long") == u'1'
+        assert numbers.format_compact_decimal(999, locale='en_US', format_type="long") == u'999'
+        assert numbers.format_compact_decimal(1000, locale='en_US', format_type="long") == u'1 thousand'
+        assert numbers.format_compact_decimal(9000, locale='en_US', format_type="long") == u'9 thousand'
+        assert numbers.format_compact_decimal(9000, locale='en_US', format_type="long", fraction_digits=2) == u'9 thousand'
+        assert numbers.format_compact_decimal(10000, locale='en_US', format_type="long") == u'10 thousand'
+        assert numbers.format_compact_decimal(10000, locale='en_US', format_type="long", fraction_digits=2) == u'10 thousand'
+        assert numbers.format_compact_decimal(1000000, locale='en_US', format_type="long") == u'1 million'
+        assert numbers.format_compact_decimal(9999999, locale='en_US', format_type="long") == u'10 million'
+        assert numbers.format_compact_decimal(9999999999, locale='en_US', format_type="long", fraction_digits=5) == u'10 billion'
+        assert numbers.format_compact_decimal(1, locale='ja_JP', format_type="short") == u'1'
+        assert numbers.format_compact_decimal(999, locale='ja_JP', format_type="short") == u'999'
+        assert numbers.format_compact_decimal(1000, locale='ja_JP', format_type="short") == u'1000'
+        assert numbers.format_compact_decimal(9123, locale='ja_JP', format_type="short") == u'9123'
+        assert numbers.format_compact_decimal(10000, locale='ja_JP', format_type="short") == u'1万'
+        assert numbers.format_compact_decimal(1234567, locale='ja_JP', format_type="long") == u'123万'
+        assert numbers.format_compact_decimal(-1, locale='en_US', format_type="short") == u'-1'
+        assert numbers.format_compact_decimal(-1234, locale='en_US', format_type="short", fraction_digits=2) == u'-1.23K'
+        assert numbers.format_compact_decimal(-123456789, format_type='short', locale='en_US') == u'-123M'
+        assert numbers.format_compact_decimal(-123456789, format_type='long', locale='en_US') == u'-123 million'
+        assert numbers.format_compact_decimal(2345678, locale='mk', format_type='long') == u'2 милиони'
+        assert numbers.format_compact_decimal(21098765, locale='mk', format_type='long') == u'21 милион'
 
 class NumberParsingTestCase(unittest.TestCase):