Merged revisions 74279 via svnmerge from

mdickinson · mdickinson · commit 8d238293cd76 · 2009-08-02T10:16:33.000Z
svn+ssh://pythondev@www.python.org/python/branches/py3k ........ r74279 | mark.dickinson | 2009-08-02 11:14:23 +0100 (Sun, 02 Aug 2009) | 1 line Issue #6595: Allow Decimal constructor to accept non-European decimal digits, as recommended by the specification. ........
diff --git a/Doc/library/decimal.rst b/Doc/library/decimal.rst
@@ -324,6 +324,11 @@ Decimal objects
       numeric-value  ::=  decimal-part [exponent-part] | infinity
       numeric-string ::=  [sign] numeric-value | [sign] nan
 
+   Other Unicode decimal digits are also permitted where ``digit``
+   appears above.  These include decimal digits from various other
+   alphabets (for example, Arabic-Indic and Devanāgarī digits) along
+   with the fullwidth digits ``'\uff10'`` through ``'\uff19'``.
+
    If *value* is a :class:`tuple`, it should have three components, a sign
    (:const:`0` for positive or :const:`1` for negative), a :class:`tuple` of
    digits, and an integer exponent. For example, ``Decimal((0, (1, 4, 1, 4), -3))``
diff --git a/Lib/decimal.py b/Lib/decimal.py
@@ -554,20 +554,16 @@ def __new__(cls, value="0", context=None):
             intpart = m.group('int')
             if intpart is not None:
                 # finite number
-                fracpart = m.group('frac')
+                fracpart = m.group('frac') or ''
                 exp = int(m.group('exp') or '0')
-                if fracpart is not None:
-                    self._int = (intpart+fracpart).lstrip('0') or '0'
-                    self._exp = exp - len(fracpart)
-                else:
-                    self._int = intpart.lstrip('0') or '0'
-                    self._exp = exp
+                self._int = str(int(intpart+fracpart))
+                self._exp = exp - len(fracpart)
                 self._is_special = False
             else:
                 diag = m.group('diag')
                 if diag is not None:
                     # NaN
-                    self._int = diag.lstrip('0')
+                    self._int = str(int(diag or '0')).lstrip('0')
                     if m.group('signal'):
                         self._exp = 'N'
                     else:
@@ -5482,26 +5478,23 @@ def _convert_other(other, raiseit=False):
 # 2. For finite numbers (not infinities and NaNs) the body of the
 # number between the optional sign and the optional exponent must have
 # at least one decimal digit, possibly after the decimal point.  The
-# lookahead expression '(?=[0-9]|\.[0-9])' checks this.
-#
-# As the flag UNICODE is not enabled here, we're explicitly avoiding any
-# other meaning for \d than the numbers [0-9].
+# lookahead expression '(?=\d|\.\d)' checks this.
 
 import re
 _parser = re.compile(r"""        # A numeric string consists of:
 #    \s*
     (?P<sign>[-+])?              # an optional sign, followed by either...
     (
-        (?=[0-9]|\.[0-9])        # ...a number (with at least one digit)
-        (?P<int>[0-9]*)          # having a (possibly empty) integer part
-        (\.(?P<frac>[0-9]*))?    # followed by an optional fractional part
-        (E(?P<exp>[-+]?[0-9]+))? # followed by an optional exponent, or...
+        (?=\d|\.\d)              # ...a number (with at least one digit)
+        (?P<int>\d*)             # having a (possibly empty) integer part
+        (\.(?P<frac>\d*))?       # followed by an optional fractional part
+        (E(?P<exp>[-+]?\d+))?    # followed by an optional exponent, or...
     |
         Inf(inity)?              # ...an infinity, or...
     |
         (?P<signal>s)?           # ...an (optionally signaling)
         NaN                      # NaN
-        (?P<diag>[0-9]*)         # with (possibly empty) diagnostic info.
+        (?P<diag>\d*)            # with (possibly empty) diagnostic info.
     )
 #    \s*
     \Z
diff --git a/Lib/test/test_decimal.py b/Lib/test/test_decimal.py
@@ -425,9 +425,6 @@ def test_explicit_from_string(self):
         self.assertEqual(str(Decimal('1.3E4 \n')), '1.3E+4')
         self.assertEqual(str(Decimal('  -7.89')), '-7.89')
 
-        #but alternate unicode digits should not
-        self.assertEqual(str(Decimal('\uff11')), 'NaN')
-
     def test_explicit_from_tuples(self):
 
         #zero
@@ -534,6 +531,15 @@ def test_explicit_context_create_decimal(self):
         d = nc.create_decimal(prevdec)
         self.assertEqual(str(d), '5.00E+8')
 
+    def test_unicode_digits(self):
+        test_values = {
+            '\uff11': '1',
+            '\u0660.\u0660\u0663\u0667\u0662e-\u0663' : '0.0000372',
+            '-nan\u0c68\u0c6a\u0c66\u0c66' : '-NaN2400',
+            }
+        for input, expected in test_values.items():
+            self.assertEqual(str(Decimal(input)), expected)
+
 
 class DecimalImplicitConstructionTest(unittest.TestCase):
     '''Unit tests for Implicit Construction cases of Decimal.'''
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -42,6 +42,10 @@ C-API
 Library
 -------
 
+- Issue #6595: The Decimal constructor now allows arbitrary Unicode
+  decimal digits in input, as recommended by the standard.  Previously
+  it was restricted to accepting [0-9].
+
 - Issues #5155, #5313, #5331: multiprocessing.Process._bootstrap was
   unconditionally calling "os.close(sys.stdin.fileno())" resulting in file
   descriptor errors
