| commit | author | age | ||
| 71084d | 1 | diff --git a/Lib/sre_compile.py b/Lib/sre_compile.py |
| CS | 2 | index 7cda2b6..15d2324 100644 |
| 3 | --- a/Lib/sre_compile.py | |
| 4 | +++ b/Lib/sre_compile.py | |
| 5 | @@ -355,8 +355,6 @@ def _optimize_unicode(charset, fixup): | |
| 6 | def _simple(av): | |
| 7 | # check if av is a "simple" operator | |
| 8 | lo, hi = av[2].getwidth() | |
| 9 | - if lo == 0 and hi == MAXREPEAT: | |
| 10 | - raise error, "nothing to repeat" | |
| 11 | return lo == hi == 1 and av[2][0][0] != SUBPATTERN | |
| 12 | ||
| 13 | def _compile_info(code, pattern, flags): | |
| 14 | diff --git a/Lib/sre_parse.py b/Lib/sre_parse.py | |
| 15 | index 75f8c96..644441d 100644 | |
| 16 | --- a/Lib/sre_parse.py | |
| 17 | +++ b/Lib/sre_parse.py | |
| 18 | @@ -147,7 +147,7 @@ class SubPattern: | |
| 19 | REPEATCODES = (MIN_REPEAT, MAX_REPEAT) | |
| 20 | for op, av in self.data: | |
| 21 | if op is BRANCH: | |
| 22 | - i = sys.maxint | |
| 23 | + i = MAXREPEAT - 1 | |
| 24 | j = 0 | |
| 25 | for av in av[1]: | |
| 26 | l, h = av.getwidth() | |
| 27 | @@ -165,14 +165,14 @@ class SubPattern: | |
| 28 | hi = hi + j | |
| 29 | elif op in REPEATCODES: | |
| 30 | i, j = av[2].getwidth() | |
| 31 | - lo = lo + long(i) * av[0] | |
| 32 | - hi = hi + long(j) * av[1] | |
| 33 | + lo = lo + i * av[0] | |
| 34 | + hi = hi + j * av[1] | |
| 35 | elif op in UNITCODES: | |
| 36 | lo = lo + 1 | |
| 37 | hi = hi + 1 | |
| 38 | elif op == SUCCESS: | |
| 39 | break | |
| 40 | - self.width = int(min(lo, sys.maxint)), int(min(hi, sys.maxint)) | |
| 41 | + self.width = min(lo, MAXREPEAT - 1), min(hi, MAXREPEAT) | |
| 42 | return self.width | |
| 43 | ||
| 44 | class Tokenizer: | |
| 45 | diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py | |
| 46 | index 18a81a2..f0827d8 100644 | |
| 47 | --- a/Lib/test/test_re.py | |
| 48 | +++ b/Lib/test/test_re.py | |
| 49 | @@ -897,6 +897,17 @@ class ReTests(unittest.TestCase): | |
| 50 | with self.assertRaisesRegexp(sre_constants.error, '\?foo'): | |
| 51 | re.compile('(?P<?foo>)') | |
| 52 | ||
| 53 | + def test_bug_2537(self): | |
| 54 | + # issue 2537: empty submatches | |
| 55 | + for outer_op in ('{0,}', '*', '+', '{1,187}'): | |
| 56 | + for inner_op in ('{0,}', '*', '?'): | |
| 57 | + r = re.compile("^((x|y)%s)%s" % (inner_op, outer_op)) | |
| 58 | + m = r.match("xyyzy") | |
| 59 | + self.assertEqual(m.group(0), "xyy") | |
| 60 | + self.assertEqual(m.group(1), "") | |
| 61 | + self.assertEqual(m.group(2), "y") | |
| 62 | + | |
| 63 | + | |
| 64 | ||
| 65 | def run_re_tests(): | |
| 66 | from test.re_tests import tests, SUCCEED, FAIL, SYNTAX_ERROR | |
| 67 | diff --git a/Lib/doctest.py b/Lib/doctest.py | |
| 68 | index 90bcca1..0ee40a2 100644 | |
| 69 | --- a/Lib/doctest.py | |
| 70 | +++ b/Lib/doctest.py | |
| 71 | @@ -564,7 +564,7 @@ class DocTestParser: | |
| 72 | # Want consists of any non-blank lines that do not start with PS1. | |
| 73 | (?P<want> (?:(?![ ]*$) # Not a blank line | |
| 74 | (?![ ]*>>>) # Not a line starting with PS1 | |
| 75 | - .*$\n? # But any other line | |
| 76 | + .+$\n? # But any other line | |
| 77 | )*) | |
| 78 | ''', re.MULTILINE | re.VERBOSE) | |
| 79 | ||
| 80 | ||
