summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--Doc/library/string.rst12
-rw-r--r--Lib/test/test_long.py28
-rw-r--r--Misc/NEWS3
-rw-r--r--Python/formatter_unicode.c72
4 files changed, 93 insertions, 22 deletions
diff --git a/Doc/library/string.rst b/Doc/library/string.rst
index c421c72..b5d5ed1 100644
--- a/Doc/library/string.rst
+++ b/Doc/library/string.rst
@@ -300,7 +300,7 @@ non-empty format string typically modifies the result.
The general form of a *standard format specifier* is:
.. productionlist:: sf
- format_spec: [[`fill`]`align`][`sign`][#][0][`width`][,][.`precision`][`type`]
+ format_spec: [[`fill`]`align`][`sign`][#][0][`width`][,][_][.`precision`][`type`]
fill: <any character>
align: "<" | ">" | "=" | "^"
sign: "+" | "-" | " "
@@ -378,6 +378,16 @@ instead.
.. versionchanged:: 3.1
Added the ``','`` option (see also :pep:`378`).
+The ``'_'`` option signals the use of an underscore for a thousands
+separator for floating point presentation types and for integer
+presentation type ``'d'``. For integer presentation types ``'b'``,
+``'o'``, ``'x'``, and ``'X'``, underscores will be inserted every 4
+digits. For other presentation types, specifying this option is an
+error.
+
+.. versionchanged:: 3.6
+ Added the ``'_'`` option (see also :pep:`515`).
+
*width* is a decimal integer defining the minimum field width. If not
specified, then the field width will be determined by the content.
diff --git a/Lib/test/test_long.py b/Lib/test/test_long.py
index 4cc4b05..fd15f04 100644
--- a/Lib/test/test_long.py
+++ b/Lib/test/test_long.py
@@ -621,6 +621,8 @@ class LongTest(unittest.TestCase):
def test__format__(self):
self.assertEqual(format(123456789, 'd'), '123456789')
self.assertEqual(format(123456789, 'd'), '123456789')
+ self.assertEqual(format(123456789, ','), '123,456,789')
+ self.assertEqual(format(123456789, '_'), '123_456_789')
# sign and aligning are interdependent
self.assertEqual(format(1, "-"), '1')
@@ -649,8 +651,25 @@ class LongTest(unittest.TestCase):
self.assertEqual(format(int('be', 16), "X"), "BE")
self.assertEqual(format(-int('be', 16), "x"), "-be")
self.assertEqual(format(-int('be', 16), "X"), "-BE")
+ self.assertRaises(ValueError, format, 1234567890, ',x')
+ self.assertEqual(format(1234567890, '_x'), '4996_02d2')
+ self.assertEqual(format(1234567890, '_X'), '4996_02D2')
# octal
+ self.assertEqual(format(3, "o"), "3")
+ self.assertEqual(format(-3, "o"), "-3")
+ self.assertEqual(format(1234, "o"), "2322")
+ self.assertEqual(format(-1234, "o"), "-2322")
+ self.assertEqual(format(1234, "-o"), "2322")
+ self.assertEqual(format(-1234, "-o"), "-2322")
+ self.assertEqual(format(1234, " o"), " 2322")
+ self.assertEqual(format(-1234, " o"), "-2322")
+ self.assertEqual(format(1234, "+o"), "+2322")
+ self.assertEqual(format(-1234, "+o"), "-2322")
+ self.assertRaises(ValueError, format, 1234567890, ',o')
+ self.assertEqual(format(1234567890, '_o'), '111_4540_1322')
+
+ # binary
self.assertEqual(format(3, "b"), "11")
self.assertEqual(format(-3, "b"), "-11")
self.assertEqual(format(1234, "b"), "10011010010")
@@ -661,12 +680,21 @@ class LongTest(unittest.TestCase):
self.assertEqual(format(-1234, " b"), "-10011010010")
self.assertEqual(format(1234, "+b"), "+10011010010")
self.assertEqual(format(-1234, "+b"), "-10011010010")
+ self.assertRaises(ValueError, format, 1234567890, ',b')
+ self.assertEqual(format(12345, '_b'), '11_0000_0011_1001')
# make sure these are errors
self.assertRaises(ValueError, format, 3, "1.3") # precision disallowed
+ self.assertRaises(ValueError, format, 3, "_c") # underscore,
+ self.assertRaises(ValueError, format, 3, ",c") # comma, and
self.assertRaises(ValueError, format, 3, "+c") # sign not allowed
# with 'c'
+ self.assertRaisesRegex(ValueError, 'Cannot specify both', format, 3, '_,')
+ self.assertRaisesRegex(ValueError, 'Cannot specify both', format, 3, ',_')
+ self.assertRaisesRegex(ValueError, 'Cannot specify both', format, 3, '_,d')
+ self.assertRaisesRegex(ValueError, 'Cannot specify both', format, 3, ',_d')
+
# ensure that only int and float type specifiers work
for format_spec in ([chr(x) for x in range(ord('a'), ord('z')+1)] +
[chr(x) for x in range(ord('A'), ord('Z')+1)]):
diff --git a/Misc/NEWS b/Misc/NEWS
index b8cc778..5628dc9 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -10,6 +10,9 @@ What's New in Python 3.6.0 beta 1
Core and Builtins
-----------------
+- Issue #27080: Implement formatting support for PEP 515. Initial patch
+ by Chris Angelico.
+
- Issue #27199: In tarfile, expose copyfileobj bufsize to improve throughput.
Patch by Jason Fried.
diff --git a/Python/formatter_unicode.c b/Python/formatter_unicode.c
index db9f5b8..95c507e 100644
--- a/Python/formatter_unicode.c
+++ b/Python/formatter_unicode.c
@@ -32,14 +32,20 @@ invalid_comma_type(Py_UCS4 presentation_type)
{
if (presentation_type > 32 && presentation_type < 128)
PyErr_Format(PyExc_ValueError,
- "Cannot specify ',' with '%c'.",
+ "Cannot specify ',' or '_' with '%c'.",
(char)presentation_type);
else
PyErr_Format(PyExc_ValueError,
- "Cannot specify ',' with '\\x%x'.",
+ "Cannot specify ',' or '_' with '\\x%x'.",
(unsigned int)presentation_type);
}
+static void
+invalid_comma_and_underscore()
+{
+ PyErr_Format(PyExc_ValueError, "Cannot specify both ',' and '_'.");
+}
+
/*
get_integer consumes 0 or more decimal digit characters from an
input string, updates *result with the corresponding positive
@@ -108,6 +114,12 @@ is_sign_element(Py_UCS4 c)
}
}
+/* Locale type codes. LT_NO_LOCALE must be zero. */
+#define LT_NO_LOCALE 0
+#define LT_DEFAULT_LOCALE 1
+#define LT_UNDERSCORE_LOCALE 2
+#define LT_UNDER_FOUR_LOCALE 3
+#define LT_CURRENT_LOCALE 4
typedef struct {
Py_UCS4 fill_char;
@@ -223,9 +235,22 @@ parse_internal_render_format_spec(PyObject *format_spec,
/* Comma signifies add thousands separators */
if (end-pos && READ_spec(pos) == ',') {
- format->thousands_separators = 1;
+ format->thousands_separators = LT_DEFAULT_LOCALE;
++pos;
}
+ /* Underscore signifies add thousands separators */
+ if (end-pos && READ_spec(pos) == '_') {
+ if (format->thousands_separators != 0) {
+ invalid_comma_and_underscore();
+ return 0;
+ }
+ format->thousands_separators = LT_UNDERSCORE_LOCALE;
+ ++pos;
+ }
+ if (end-pos && READ_spec(pos) == ',') {
+ invalid_comma_and_underscore();
+ return 0;
+ }
/* Parse field precision */
if (end-pos && READ_spec(pos) == '.') {
@@ -275,6 +300,16 @@ parse_internal_render_format_spec(PyObject *format_spec,
case '\0':
/* These are allowed. See PEP 378.*/
break;
+ case 'b':
+ case 'o':
+ case 'x':
+ case 'X':
+ /* Underscores are allowed in bin/oct/hex. See PEP 515. */
+ if (format->thousands_separators == LT_UNDERSCORE_LOCALE) {
+ /* Every four digits, not every three, in bin/oct/hex. */
+ format->thousands_separators = LT_UNDER_FOUR_LOCALE;
+ break;
+ }
default:
invalid_comma_type(format->type);
return 0;
@@ -351,11 +386,6 @@ fill_padding(_PyUnicodeWriter *writer,
/*********** common routines for numeric formatting *********************/
/************************************************************************/
-/* Locale type codes. */
-#define LT_CURRENT_LOCALE 0
-#define LT_DEFAULT_LOCALE 1
-#define LT_NO_LOCALE 2
-
/* Locale info needed for formatting integers and the part of floats
before and including the decimal. Note that locales only support
8-bit chars, not unicode. */
@@ -667,8 +697,8 @@ static const char no_grouping[1] = {CHAR_MAX};
/* Find the decimal point character(s?), thousands_separator(s?), and
grouping description, either for the current locale if type is
- LT_CURRENT_LOCALE, a hard-coded locale if LT_DEFAULT_LOCALE, or
- none if LT_NO_LOCALE. */
+ LT_CURRENT_LOCALE, a hard-coded locale if LT_DEFAULT_LOCALE or
+ LT_UNDERSCORE_LOCALE/LT_UNDER_FOUR_LOCALE, or none if LT_NO_LOCALE. */
static int
get_locale_info(int type, LocaleInfo *locale_info)
{
@@ -691,16 +721,22 @@ get_locale_info(int type, LocaleInfo *locale_info)
break;
}
case LT_DEFAULT_LOCALE:
+ case LT_UNDERSCORE_LOCALE:
+ case LT_UNDER_FOUR_LOCALE:
locale_info->decimal_point = PyUnicode_FromOrdinal('.');
- locale_info->thousands_sep = PyUnicode_FromOrdinal(',');
+ locale_info->thousands_sep = PyUnicode_FromOrdinal(
+ type == LT_DEFAULT_LOCALE ? ',' : '_');
if (!locale_info->decimal_point || !locale_info->thousands_sep) {
Py_XDECREF(locale_info->decimal_point);
Py_XDECREF(locale_info->thousands_sep);
return -1;
}
- locale_info->grouping = "\3"; /* Group every 3 characters. The
+ if (type != LT_UNDER_FOUR_LOCALE)
+ locale_info->grouping = "\3"; /* Group every 3 characters. The
(implicit) trailing 0 means repeat
infinitely. */
+ else
+ locale_info->grouping = "\4"; /* Bin/oct/hex group every four. */
break;
case LT_NO_LOCALE:
locale_info->decimal_point = PyUnicode_FromOrdinal('.');
@@ -952,9 +988,7 @@ format_long_internal(PyObject *value, const InternalFormatSpec *format,
/* Determine the grouping, separator, and decimal point, if any. */
if (get_locale_info(format->type == 'n' ? LT_CURRENT_LOCALE :
- (format->thousands_separators ?
- LT_DEFAULT_LOCALE :
- LT_NO_LOCALE),
+ format->thousands_separators,
&locale) == -1)
goto done;
@@ -1099,9 +1133,7 @@ format_float_internal(PyObject *value,
/* Determine the grouping, separator, and decimal point, if any. */
if (get_locale_info(format->type == 'n' ? LT_CURRENT_LOCALE :
- (format->thousands_separators ?
- LT_DEFAULT_LOCALE :
- LT_NO_LOCALE),
+ format->thousands_separators,
&locale) == -1)
goto done;
@@ -1277,9 +1309,7 @@ format_complex_internal(PyObject *value,
/* Determine the grouping, separator, and decimal point, if any. */
if (get_locale_info(format->type == 'n' ? LT_CURRENT_LOCALE :
- (format->thousands_separators ?
- LT_DEFAULT_LOCALE :
- LT_NO_LOCALE),
+ format->thousands_separators,
&locale) == -1)
goto done;