Issue 24454: Improve the usability of the re match object named group API

This commit is contained in:
Eric V. Smith 2016-09-11 08:55:43 -04:00
parent a3c1728bb6
commit 605bdae078
4 changed files with 82 additions and 1 deletions

View File

@ -1015,6 +1015,22 @@ Match objects support the following methods and attributes:
'c3'
.. method:: match.__getitem__(g)
This is identical to ``m.group(g)``. This allows easier access to
an individual group from a match:
>>> m = re.match(r"(\w+) (\w+)", "Isaac Newton, physicist")
>>> m[0] # The entire match
'Isaac Newton'
>>> m[1] # The first parenthesized subgroup.
'Isaac'
>>> m[2] # The second parenthesized subgroup.
'Newton'
.. versionadded:: 3.6
.. method:: match.groups(default=None)
Return a tuple containing all the subgroups of the match, from 1 up to however

View File

@ -441,6 +441,50 @@ class ReTests(unittest.TestCase):
self.assertEqual(m.group(2, 1), ('b', 'a'))
self.assertEqual(m.group(Index(2), Index(1)), ('b', 'a'))
def test_match_getitem(self):
pat = re.compile('(?:(?P<a1>a)|(?P<b2>b))(?P<c3>c)?')
m = pat.match('a')
self.assertEqual(m['a1'], 'a')
self.assertEqual(m['b2'], None)
self.assertEqual(m['c3'], None)
self.assertEqual('a1={a1} b2={b2} c3={c3}'.format_map(m), 'a1=a b2=None c3=None')
self.assertEqual(m[0], 'a')
self.assertEqual(m[1], 'a')
self.assertEqual(m[2], None)
self.assertEqual(m[3], None)
with self.assertRaisesRegex(IndexError, 'no such group'):
m['X']
with self.assertRaisesRegex(IndexError, 'no such group'):
m[-1]
with self.assertRaisesRegex(IndexError, 'no such group'):
m[4]
with self.assertRaisesRegex(IndexError, 'no such group'):
m[0, 1]
with self.assertRaisesRegex(IndexError, 'no such group'):
m[(0,)]
with self.assertRaisesRegex(IndexError, 'no such group'):
m[(0, 1)]
with self.assertRaisesRegex(KeyError, 'a2'):
'a1={a2}'.format_map(m)
m = pat.match('ac')
self.assertEqual(m['a1'], 'a')
self.assertEqual(m['b2'], None)
self.assertEqual(m['c3'], 'c')
self.assertEqual('a1={a1} b2={b2} c3={c3}'.format_map(m), 'a1=a b2=None c3=c')
self.assertEqual(m[0], 'ac')
self.assertEqual(m[1], 'a')
self.assertEqual(m[2], None)
self.assertEqual(m[3], 'c')
# Cannot assign.
with self.assertRaises(TypeError):
m[0] = 1
# No len().
self.assertRaises(TypeError, len, m)
def test_re_fullmatch(self):
# Issue 16203: Proposal: add re.fullmatch() method.
self.assertEqual(re.fullmatch(r"a", "a").span(), (0, 1))

View File

@ -143,6 +143,10 @@ Core and Builtins
Library
-------
- Issue #24454: Regular expression match object groups are now
accessible using __getitem__. "mo[x]" is equivalent to
"mo.group(x)".
- Issue #10740: sqlite3 no longer implicitly commit an open transaction
before DDL statements.

View File

@ -2121,6 +2121,12 @@ match_group(MatchObject* self, PyObject* args)
return result;
}
static PyObject*
match_getitem(MatchObject* self, PyObject* name)
{
return match_getslice(self, name, Py_None);
}
/*[clinic input]
_sre.SRE_Match.groups
@ -2416,6 +2422,9 @@ PyDoc_STRVAR(match_group_doc,
Return subgroup(s) of the match by indices or names.\n\
For 0 returns the entire match.");
PyDoc_STRVAR(match_getitem_doc,
"__getitem__(name) <==> group(name).\n");
static PyObject *
match_lastindex_get(MatchObject *self)
{
@ -2706,6 +2715,13 @@ static PyTypeObject Pattern_Type = {
pattern_getset, /* tp_getset */
};
/* Match objects do not support length or assignment, but do support
__getitem__. */
static PyMappingMethods match_as_mapping = {
NULL,
(binaryfunc)match_getitem,
NULL
};
static PyMethodDef match_methods[] = {
{"group", (PyCFunction) match_group, METH_VARARGS, match_group_doc},
@ -2717,6 +2733,7 @@ static PyMethodDef match_methods[] = {
_SRE_SRE_MATCH_EXPAND_METHODDEF
_SRE_SRE_MATCH___COPY___METHODDEF
_SRE_SRE_MATCH___DEEPCOPY___METHODDEF
{"__getitem__", (PyCFunction)match_getitem, METH_O|METH_COEXIST, match_getitem_doc},
{NULL, NULL}
};
@ -2751,7 +2768,7 @@ static PyTypeObject Match_Type = {
(reprfunc)match_repr, /* tp_repr */
0, /* tp_as_number */
0, /* tp_as_sequence */
0, /* tp_as_mapping */
&match_as_mapping, /* tp_as_mapping */
0, /* tp_hash */
0, /* tp_call */
0, /* tp_str */