[Python-checkins] cpython (3.4): backout 9fcf4008b626 (#9179) for further consideration
benjamin.peterson
python-checkins at python.org
Sun Nov 30 17:52:06 CET 2014
https://hg.python.org/cpython/rev/d1f7c3f45ffe
changeset: 93665:d1f7c3f45ffe
branch: 3.4
parent: 93663:086a21998db2
user: Benjamin Peterson <benjamin at python.org>
date: Sun Nov 30 11:49:00 2014 -0500
summary:
backout 9fcf4008b626 (#9179) for further consideration
files:
Lib/re.py | 5 +--
Lib/sre_parse.py | 33 +++++++--------------------
Lib/test/test_re.py | 38 +--------------------------------
Misc/NEWS | 3 --
4 files changed, 12 insertions(+), 67 deletions(-)
diff --git a/Lib/re.py b/Lib/re.py
--- a/Lib/re.py
+++ b/Lib/re.py
@@ -352,11 +352,10 @@
s = sre_parse.Pattern()
s.flags = flags
for phrase, action in lexicon:
- gid = s.opengroup()
p.append(sre_parse.SubPattern(s, [
- (SUBPATTERN, (gid, sre_parse.parse(phrase, flags))),
+ (SUBPATTERN, (len(p)+1, sre_parse.parse(phrase, flags))),
]))
- s.closegroup(gid, p[-1])
+ s.groups = len(p)+1
p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
self.scanner = sre_compile.compile(p)
def scan(self, string):
diff --git a/Lib/sre_parse.py b/Lib/sre_parse.py
--- a/Lib/sre_parse.py
+++ b/Lib/sre_parse.py
@@ -66,25 +66,24 @@
# master pattern object. keeps track of global attributes
def __init__(self):
self.flags = 0
+ self.open = []
+ self.groups = 1
self.groupdict = {}
- self.subpatterns = [None] # group 0
- @property
- def groups(self):
- return len(self.subpatterns)
def opengroup(self, name=None):
gid = self.groups
- self.subpatterns.append(None)
+ self.groups = gid + 1
if name is not None:
ogid = self.groupdict.get(name, None)
if ogid is not None:
raise error("redefinition of group name %s as group %d; "
"was group %d" % (repr(name), gid, ogid))
self.groupdict[name] = gid
+ self.open.append(gid)
return gid
- def closegroup(self, gid, p):
- self.subpatterns[gid] = p
+ def closegroup(self, gid):
+ self.open.remove(gid)
def checkgroup(self, gid):
- return gid < self.groups and self.subpatterns[gid] is not None
+ return gid < self.groups and gid not in self.open
class SubPattern:
# a subpattern, in intermediate form
@@ -182,21 +181,7 @@
elif op in UNITCODES:
lo = lo + 1
hi = hi + 1
- elif op is GROUPREF:
- i, j = self.pattern.subpatterns[av].getwidth()
- lo = lo + i
- hi = hi + j
- elif op is GROUPREF_EXISTS:
- i, j = av[1].getwidth()
- if av[2] is not None:
- l, h = av[2].getwidth()
- i = min(i, l)
- j = max(j, h)
- else:
- i = 0
- lo = lo + i
- hi = hi + j
- elif op is SUCCESS:
+ elif op == SUCCESS:
break
self.width = min(lo, MAXREPEAT - 1), min(hi, MAXREPEAT)
return self.width
@@ -724,7 +709,7 @@
if not sourcematch(")"):
raise error("unbalanced parenthesis")
if group is not None:
- state.closegroup(group, p)
+ state.closegroup(group)
subpatternappend((SUBPATTERN, (group, p)))
else:
while 1:
diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py
--- a/Lib/test/test_re.py
+++ b/Lib/test/test_re.py
@@ -557,7 +557,7 @@
self.assertEqual(re.match("a.*b", "a\n\nb", re.DOTALL).group(0),
"a\n\nb")
- def test_lookahead(self):
+ def test_non_consuming(self):
self.assertEqual(re.match("(a(?=\s[^a]))", "a b").group(1), "a")
self.assertEqual(re.match("(a(?=\s[^a]*))", "a b").group(1), "a")
self.assertEqual(re.match("(a(?=\s[abc]))", "a b").group(1), "a")
@@ -571,42 +571,6 @@
self.assertEqual(re.match(r"(a)(?!\s\1)", "a b").group(1), "a")
self.assertEqual(re.match(r"(a)(?!\s(abc|a))", "a b").group(1), "a")
- # Group reference.
- self.assertTrue(re.match(r'(a)b(?=\1)a', 'aba'))
- self.assertIsNone(re.match(r'(a)b(?=\1)c', 'abac'))
- # Conditional group reference.
- self.assertTrue(re.match('(?:(a)|(x))b(?=(?(2)x|c))c', 'abc'))
- self.assertIsNone(re.match('(?:(a)|(x))b(?=(?(2)c|x))c', 'abc'))
- self.assertTrue(re.match('(?:(a)|(x))b(?=(?(2)x|c))c', 'abc'))
- self.assertIsNone(re.match('(?:(a)|(x))b(?=(?(1)b|x))c', 'abc'))
- self.assertTrue(re.match('(?:(a)|(x))b(?=(?(1)c|x))c', 'abc'))
- # Group used before defined.
- self.assertTrue(re.match('(a)b(?=(?(2)x|c))(c)', 'abc'))
- self.assertIsNone(re.match('(a)b(?=(?(2)b|x))(c)', 'abc'))
- self.assertTrue(re.match('(a)b(?=(?(1)c|x))(c)', 'abc'))
-
- def test_lookbehind(self):
- self.assertTrue(re.match('ab(?<=b)c', 'abc'))
- self.assertIsNone(re.match('ab(?<=c)c', 'abc'))
- self.assertIsNone(re.match('ab(?<!b)c', 'abc'))
- self.assertTrue(re.match('ab(?<!c)c', 'abc'))
- # Group reference.
- self.assertTrue(re.match(r'(a)a(?<=\1)c', 'aac'))
- self.assertIsNone(re.match(r'(a)b(?<=\1)a', 'abaa'))
- self.assertIsNone(re.match(r'(a)a(?<!\1)c', 'aac'))
- self.assertTrue(re.match(r'(a)b(?<!\1)a', 'abaa'))
- # Conditional group reference.
- self.assertIsNone(re.match('(?:(a)|(x))b(?<=(?(2)x|c))c', 'abc'))
- self.assertIsNone(re.match('(?:(a)|(x))b(?<=(?(2)b|x))c', 'abc'))
- self.assertTrue(re.match('(?:(a)|(x))b(?<=(?(2)x|b))c', 'abc'))
- self.assertIsNone(re.match('(?:(a)|(x))b(?<=(?(1)c|x))c', 'abc'))
- self.assertTrue(re.match('(?:(a)|(x))b(?<=(?(1)b|x))c', 'abc'))
- # Group used before defined.
- self.assertIsNone(re.match('(a)b(?<=(?(2)x|c))(c)', 'abc'))
- self.assertIsNone(re.match('(a)b(?<=(?(2)b|x))(c)', 'abc'))
- self.assertIsNone(re.match('(a)b(?<=(?(1)c|x))(c)', 'abc'))
- self.assertTrue(re.match('(a)b(?<=(?(1)b|x))(c)', 'abc'))
-
def test_ignore_case(self):
self.assertEqual(re.match("abc", "ABC", re.I).group(0), "ABC")
self.assertEqual(re.match(b"abc", b"ABC", re.I).group(0), b"ABC")
diff --git a/Misc/NEWS b/Misc/NEWS
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -68,9 +68,6 @@
- Issue #22821: Fixed fcntl() with integer argument on 64-bit big-endian
platforms.
-- Issues #814253, #9179: Group references and conditional group references now
- work in lookbehind assertions in regular expressions.
-
- Issue #22406: Fixed the uu_codec codec incorrectly ported to 3.x.
Based on patch by Martin Panter.
--
Repository URL: https://hg.python.org/cpython
More information about the Python-checkins
mailing list