-- tightened up parsing of octal numbers

-- improved the SRE test harness: don't use asserts, test a few more
   things (including more boundary conditions)
This commit is contained in:
Fredrik Lundh 2000-09-02 11:03:34 +00:00
parent 412f246024
commit 143328ba63
3 changed files with 157 additions and 197 deletions

View File

@ -15,7 +15,7 @@ from sre_constants import *
MAXREPEAT = 65535 MAXREPEAT = 65535
SPECIAL_CHARS = ".\\[{()*+?^$|" SPECIAL_CHARS = ".\\[{()*+?^$|"
REPEAT_CHARS = "*+?{" REPEAT_CHARS = "*+?{"
DIGITS = tuple("0123456789") DIGITS = tuple("0123456789")
@ -259,13 +259,12 @@ def _escape(source, escape, state):
# hexadecimal escape # hexadecimal escape
while source.next in HEXDIGITS and len(escape) < 4: while source.next in HEXDIGITS and len(escape) < 4:
escape = escape + source.get() escape = escape + source.get()
escape = escape[2:] if len(escape) != 4:
if len(escape) != 2: raise ValueError
raise error, "bogus escape: %s" % repr("\\" + escape) return LITERAL, int(escape[2:], 16) & 0xff
return LITERAL, int(escape, 16) & 0xff
elif escape[1:2] == "0": elif escape[1:2] == "0":
# octal escape # octal escape
while source.next in OCTDIGITS and len(escape) < 5: while source.next in OCTDIGITS and len(escape) < 4:
escape = escape + source.get() escape = escape + source.get()
return LITERAL, int(escape[1:], 8) & 0xff return LITERAL, int(escape[1:], 8) & 0xff
elif escape[1:2] in DIGITS: elif escape[1:2] in DIGITS:
@ -273,7 +272,8 @@ def _escape(source, escape, state):
here = source.tell() here = source.tell()
if source.next in DIGITS: if source.next in DIGITS:
escape = escape + source.get() escape = escape + source.get()
if escape[2] in OCTDIGITS and source.next in OCTDIGITS: if (escape[1] in OCTDIGITS and escape[2] in OCTDIGITS and
source.next in OCTDIGITS):
# got three octal digits; this is an octal escape # got three octal digits; this is an octal escape
escape = escape + source.get() escape = escape + source.get()
return LITERAL, int(escape[1:], 8) & 0xff return LITERAL, int(escape[1:], 8) & 0xff
@ -281,7 +281,7 @@ def _escape(source, escape, state):
group = _group(escape, state.groups) group = _group(escape, state.groups)
if group: if group:
return GROUPREF, group return GROUPREF, group
raise error, "bogus escape: %s" % repr(escape) raise ValueError
if len(escape) == 2: if len(escape) == 2:
return LITERAL, ord(escape[1]) return LITERAL, ord(escape[1])
except ValueError: except ValueError:

View File

@ -1,2 +1 @@
test_sre test_sre
maximum recursion limit exceeded

View File

@ -1,4 +1,7 @@
# FIXME: this is basically test_re.py, with a few minor changes # SRE test harness for the Python regression suite
# this is based on test_re.py, but uses a test function instead
# of all those asserts
import sys import sys
sys.path=['.']+sys.path sys.path=['.']+sys.path
@ -7,227 +10,188 @@ from test_support import verbose, TestFailed
import sre import sre
import sys, os, string, traceback import sys, os, string, traceback
#
# test support
def test(expression, result, exception=None):
try:
r = eval(expression)
except:
if exception:
if not isinstance(sys.exc_value, exception):
print expression, "FAILED"
# display name, not actual value
if exception is sre.error:
print "expected", "sre.error"
else:
print "expected", exception.__name__
print "got", sys.exc_type.__name__, str(sys.exc_value)
else:
print expression, "FAILED"
traceback.print_exc(file=sys.stdout)
else:
if exception:
print expression, "FAILED"
if exception is sre.error:
print "expected", "sre.error"
else:
print "expected", exception.__name__
print "got result", repr(r)
else:
if r != result:
print expression, "FAILED"
print "expected", repr(result)
print "got result", repr(r)
if verbose:
print 'Running tests on character literals'
for i in range(0, 256):
test(r"""sre.match("\%03o" % i, chr(i)) != None""", 1)
test(r"""sre.match("\%03o0" % i, chr(i)+"0") != None""", 1)
test(r"""sre.match("\%03o8" % i, chr(i)+"8") != None""", 1)
test(r"""sre.match("\x%02x" % i, chr(i)) != None""", 1)
test(r"""sre.match("\x%02x0" % i, chr(i)+"0") != None""", 1)
test(r"""sre.match("\x%02xz" % i, chr(i)+"z") != None""", 1)
test(r"""sre.match("\911", "")""", None, sre.error)
#
# Misc tests from Tim Peters' re.doc # Misc tests from Tim Peters' re.doc
if verbose: if verbose:
print 'Running tests on sre.search and sre.match' print 'Running tests on sre.search and sre.match'
try: test(r"""sre.search('x*', 'axx').span(0)""", (0, 0))
assert sre.search('x*', 'axx').span(0) == (0, 0) test(r"""sre.search('x*', 'axx').span()""", (0, 0))
assert sre.search('x*', 'axx').span() == (0, 0) test(r"""sre.search('x+', 'axx').span(0)""", (1, 3))
assert sre.search('x+', 'axx').span(0) == (1, 3) test(r"""sre.search('x+', 'axx').span()""", (1, 3))
assert sre.search('x+', 'axx').span() == (1, 3) test(r"""sre.search('x', 'aaa')""", None)
assert sre.search('x', 'aaa') == None
except:
raise TestFailed, "sre.search"
try: test(r"""sre.match('a*', 'xxx').span(0)""", (0, 0))
assert sre.match('a*', 'xxx').span(0) == (0, 0) test(r"""sre.match('a*', 'xxx').span()""", (0, 0))
assert sre.match('a*', 'xxx').span() == (0, 0) test(r"""sre.match('x*', 'xxxa').span(0)""", (0, 3))
assert sre.match('x*', 'xxxa').span(0) == (0, 3) test(r"""sre.match('x*', 'xxxa').span()""", (0, 3))
assert sre.match('x*', 'xxxa').span() == (0, 3) test(r"""sre.match('a+', 'xxx')""", None)
assert sre.match('a+', 'xxx') == None
except:
raise TestFailed, "sre.search"
if verbose: if verbose:
print 'Running tests on sre.sub' print 'Running tests on sre.sub'
try: test(r"""sre.sub("(?i)b+", "x", "bbbb BBBB")""", 'x x')
assert sre.sub("(?i)b+", "x", "bbbb BBBB") == 'x x'
def bump_num(matchobj): def bump_num(matchobj):
int_value = int(matchobj.group(0)) int_value = int(matchobj.group(0))
return str(int_value + 1) return str(int_value + 1)
assert sre.sub(r'\d+', bump_num, '08.2 -2 23x99y') == '9.3 -3 24x100y' test(r"""sre.sub(r'\d+', bump_num, '08.2 -2 23x99y')""", '9.3 -3 24x100y')
assert sre.sub(r'\d+', bump_num, '08.2 -2 23x99y', 3) == '9.3 -3 23x99y' test(r"""sre.sub(r'\d+', bump_num, '08.2 -2 23x99y', 3)""", '9.3 -3 23x99y')
assert sre.sub('.', lambda m: r"\n", 'x') == '\\n' test(r"""sre.sub('.', lambda m: r"\n", 'x')""", '\\n')
assert sre.sub('.', r"\n", 'x') == '\n' test(r"""sre.sub('.', r"\n", 'x')""", '\n')
s = r"\1\1" s = r"\1\1"
assert sre.sub('(.)', s, 'x') == 'xx'
assert sre.sub('(.)', sre.escape(s), 'x') == s
assert sre.sub('(.)', lambda m: s, 'x') == s
assert sre.sub('(?P<a>x)', '\g<a>\g<a>', 'xx') == 'xxxx' test(r"""sre.sub('(.)', s, 'x')""", 'xx')
assert sre.sub('(?P<a>x)', '\g<a>\g<1>', 'xx') == 'xxxx' test(r"""sre.sub('(.)', sre.escape(s), 'x')""", s)
assert sre.sub('(?P<unk>x)', '\g<unk>\g<unk>', 'xx') == 'xxxx' test(r"""sre.sub('(.)', lambda m: s, 'x')""", s)
assert sre.sub('(?P<unk>x)', '\g<1>\g<1>', 'xx') == 'xxxx'
assert sre.sub('a', r'\t\n\v\r\f\a\b\B\Z\a\A\w\W\s\S\d\D', 'a') == '\t\n\v\r\f\a\b\\B\\Z\a\\A\\w\\W\\s\\S\\d\\D' test(r"""sre.sub('(?P<a>x)', '\g<a>\g<a>', 'xx')""", 'xxxx')
assert sre.sub('a', '\t\n\v\r\f\a', 'a') == '\t\n\v\r\f\a' test(r"""sre.sub('(?P<a>x)', '\g<a>\g<1>', 'xx')""", 'xxxx')
assert sre.sub('a', '\t\n\v\r\f\a', 'a') == (chr(9)+chr(10)+chr(11)+chr(13)+chr(12)+chr(7)) test(r"""sre.sub('(?P<unk>x)', '\g<unk>\g<unk>', 'xx')""", 'xxxx')
test(r"""sre.sub('(?P<unk>x)', '\g<1>\g<1>', 'xx')""", 'xxxx')
assert sre.sub('^\s*', 'X', 'test') == 'Xtest' test(r"""sre.sub('a', r'\t\n\v\r\f\a\b\B\Z\a\A\w\W\s\S\d\D', 'a')""", '\t\n\v\r\f\a\b\\B\\Z\a\\A\\w\\W\\s\\S\\d\\D')
except AssertionError: test(r"""sre.sub('a', '\t\n\v\r\f\a', 'a')""", '\t\n\v\r\f\a')
raise TestFailed, "sre.sub" test(r"""sre.sub('a', '\t\n\v\r\f\a', 'a')""", (chr(9)+chr(10)+chr(11)+chr(13)+chr(12)+chr(7)))
test(r"""sre.sub('^\s*', 'X', 'test')""", 'Xtest')
try: # qualified sub
assert sre.sub('a', 'b', 'aaaaa') == 'bbbbb' test(r"""sre.sub('a', 'b', 'aaaaa')""", 'bbbbb')
assert sre.sub('a', 'b', 'aaaaa', 1) == 'baaaa' test(r"""sre.sub('a', 'b', 'aaaaa', 1)""", 'baaaa')
except AssertionError:
raise TestFailed, "qualified sre.sub"
if verbose: if verbose:
print 'Running tests on symbolic references' print 'Running tests on symbolic references'
try: test(r"""sre.sub('(?P<a>x)', '\g<a', 'xx')""", None, sre.error)
sre.sub('(?P<a>x)', '\g<a', 'xx') test(r"""sre.sub('(?P<a>x)', '\g<', 'xx')""", None, sre.error)
except sre.error, reason: test(r"""sre.sub('(?P<a>x)', '\g', 'xx')""", None, sre.error)
pass test(r"""sre.sub('(?P<a>x)', '\g<a a>', 'xx')""", None, sre.error)
else: test(r"""sre.sub('(?P<a>x)', '\g<1a1>', 'xx')""", None, sre.error)
raise TestFailed, "symbolic reference" test(r"""sre.sub('(?P<a>x)', '\g<ab>', 'xx')""", None, IndexError)
test(r"""sre.sub('(?P<a>x)|(?P<b>y)', '\g<b>', 'xx')""", None, sre.error)
try: test(r"""sre.sub('(?P<a>x)|(?P<b>y)', '\\2', 'xx')""", None, sre.error)
sre.sub('(?P<a>x)', '\g<', 'xx')
except sre.error, reason:
pass
else:
raise TestFailed, "symbolic reference"
try:
sre.sub('(?P<a>x)', '\g', 'xx')
except sre.error, reason:
pass
else:
raise TestFailed, "symbolic reference"
try:
sre.sub('(?P<a>x)', '\g<a a>', 'xx')
except sre.error, reason:
pass
else:
raise TestFailed, "symbolic reference"
try:
sre.sub('(?P<a>x)', '\g<1a1>', 'xx')
except sre.error, reason:
pass
else:
raise TestFailed, "symbolic reference"
try:
sre.sub('(?P<a>x)', '\g<ab>', 'xx')
except IndexError, reason:
pass
else:
raise TestFailed, "symbolic reference"
try:
sre.sub('(?P<a>x)|(?P<b>y)', '\g<b>', 'xx')
except sre.error, reason:
pass
else:
raise TestFailed, "symbolic reference"
try:
sre.sub('(?P<a>x)|(?P<b>y)', '\\2', 'xx')
except sre.error, reason:
pass
else:
raise TestFailed, "symbolic reference"
if verbose: if verbose:
print 'Running tests on sre.subn' print 'Running tests on sre.subn'
try: test(r"""sre.subn("(?i)b+", "x", "bbbb BBBB")""", ('x x', 2))
assert sre.subn("(?i)b+", "x", "bbbb BBBB") == ('x x', 2) test(r"""sre.subn("b+", "x", "bbbb BBBB")""", ('x BBBB', 1))
assert sre.subn("b+", "x", "bbbb BBBB") == ('x BBBB', 1) test(r"""sre.subn("b+", "x", "xyz")""", ('xyz', 0))
assert sre.subn("b+", "x", "xyz") == ('xyz', 0) test(r"""sre.subn("b*", "x", "xyz")""", ('xxxyxzx', 4))
assert sre.subn("b*", "x", "xyz") == ('xxxyxzx', 4) test(r"""sre.subn("b*", "x", "xyz", 2)""", ('xxxyz', 2))
assert sre.subn("b*", "x", "xyz", 2) == ('xxxyz', 2)
except AssertionError:
raise TestFailed, "sre.subn"
if verbose: if verbose:
print 'Running tests on sre.split' print 'Running tests on sre.split'
try: test(r"""sre.split(":", ":a:b::c")""", ['', 'a', 'b', '', 'c'])
assert sre.split(":", ":a:b::c") == ['', 'a', 'b', '', 'c'] test(r"""sre.split(":*", ":a:b::c")""", ['', 'a', 'b', 'c'])
assert sre.split(":*", ":a:b::c") == ['', 'a', 'b', 'c'] test(r"""sre.split("(:*)", ":a:b::c")""", ['', ':', 'a', ':', 'b', '::', 'c'])
assert sre.split("(:*)", ":a:b::c") == ['', ':', 'a', ':', 'b', '::', 'c'] test(r"""sre.split("(?::*)", ":a:b::c")""", ['', 'a', 'b', 'c'])
assert sre.split("(?::*)", ":a:b::c") == ['', 'a', 'b', 'c'] test(r"""sre.split("(:)*", ":a:b::c")""", ['', ':', 'a', ':', 'b', ':', 'c'])
assert sre.split("(:)*", ":a:b::c") == ['', ':', 'a', ':', 'b', ':', 'c'] test(r"""sre.split("([b:]+)", ":a:b::c")""", ['', ':', 'a', ':b::', 'c'])
assert sre.split("([b:]+)", ":a:b::c") == ['', ':', 'a', ':b::', 'c'] test(r"""sre.split("(b)|(:+)", ":a:b::c")""",
assert sre.split("(b)|(:+)", ":a:b::c") == \ ['', None, ':', 'a', None, ':', '', 'b', None, '', None, '::', 'c'])
['', None, ':', 'a', None, ':', '', 'b', None, '', None, '::', 'c'] test(r"""sre.split("(?:b)|(?::+)", ":a:b::c")""", ['', 'a', '', '', 'c'])
assert sre.split("(?:b)|(?::+)", ":a:b::c") == ['', 'a', '', '', 'c']
except AssertionError:
raise TestFailed, "sre.split"
try: test(r"""sre.split(":", ":a:b::c", 2)""", ['', 'a', 'b::c'])
assert sre.split(":", ":a:b::c", 2) == ['', 'a', 'b::c'] test(r"""sre.split(':', 'a:b:c:d', 2)""", ['a', 'b', 'c:d'])
assert sre.split(':', 'a:b:c:d', 2) == ['a', 'b', 'c:d']
assert sre.split("(:)", ":a:b::c", 2) == ['', ':', 'a', ':', 'b::c'] test(r"""sre.split("(:)", ":a:b::c", 2)""", ['', ':', 'a', ':', 'b::c'])
assert sre.split("(:*)", ":a:b::c", 2) == ['', ':', 'a', ':', 'b::c'] test(r"""sre.split("(:*)", ":a:b::c", 2)""", ['', ':', 'a', ':', 'b::c'])
except AssertionError:
raise TestFailed, "qualified sre.split"
if verbose: if verbose:
print "Running tests on sre.findall" print "Running tests on sre.findall"
try: test(r"""sre.findall(":+", "abc")""", [])
assert sre.findall(":+", "abc") == [] test(r"""sre.findall(":+", "a:b::c:::d")""", [":", "::", ":::"])
assert sre.findall(":+", "a:b::c:::d") == [":", "::", ":::"] test(r"""sre.findall("(:+)", "a:b::c:::d")""", [":", "::", ":::"])
assert sre.findall("(:+)", "a:b::c:::d") == [":", "::", ":::"] test(r"""sre.findall("(:)(:*)", "a:b::c:::d")""",
assert sre.findall("(:)(:*)", "a:b::c:::d") == [(":", ""), [(":", ""), (":", ":"), (":", "::")])
(":", ":"), test(r"""sre.findall("(a)|(b)", "abc")""", [("a", ""), ("", "b")])
(":", "::")]
assert sre.findall("(a)|(b)", "abc") == [("a", ""), ("", "b")]
except AssertionError:
raise TestFailed, "sre.findall"
if verbose: if verbose:
print "Running tests on sre.match" print "Running tests on sre.match"
try: test(r"""sre.match('a', 'a').groups()""", ())
# No groups at all test(r"""sre.match('(a)', 'a').groups()""", ('a',))
m = sre.match('a', 'a') ; assert m.groups() == () test(r"""sre.match('(a)', 'a').group(0)""", 'a')
# A single group test(r"""sre.match('(a)', 'a').group(1)""", 'a')
m = sre.match('(a)', 'a') ; assert m.groups() == ('a',) test(r"""sre.match('(a)', 'a').group(1, 1)""", ('a', 'a'))
pat = sre.compile('((a)|(b))(c)?') pat = sre.compile('((a)|(b))(c)?')
assert pat.match('a').groups() == ('a', 'a', None, None) test(r"""pat.match('a').groups()""", ('a', 'a', None, None))
assert pat.match('b').groups() == ('b', None, 'b', None) test(r"""pat.match('b').groups()""", ('b', None, 'b', None))
assert pat.match('ac').groups() == ('a', 'a', None, 'c') test(r"""pat.match('ac').groups()""", ('a', 'a', None, 'c'))
assert pat.match('bc').groups() == ('b', None, 'b', 'c') test(r"""pat.match('bc').groups()""", ('b', None, 'b', 'c'))
assert pat.match('bc').groups("") == ('b', "", 'b', 'c') test(r"""pat.match('bc').groups("")""", ('b', "", 'b', 'c'))
except AssertionError:
raise TestFailed, "match .groups() method"
try: pat = sre.compile('(?:(?P<a1>a)|(?P<b2>b))(?P<c3>c)?')
# A single group test(r"""pat.match('a').group(1, 2, 3)""", ('a', None, None))
m = sre.match('(a)', 'a') test(r"""pat.match('b').group('a1', 'b2', 'c3')""", (None, 'b', None))
assert m.group(0) == 'a' ; assert m.group(0) == 'a' test(r"""pat.match('ac').group(1, 'b2', 3)""", ('a', None, 'c'))
assert m.group(1) == 'a' ; assert m.group(1, 1) == ('a', 'a')
pat = sre.compile('(?:(?P<a1>a)|(?P<b2>b))(?P<c3>c)?')
assert pat.match('a').group(1, 2, 3) == ('a', None, None)
assert pat.match('b').group('a1', 'b2', 'c3') == (None, 'b', None)
assert pat.match('ac').group(1, 'b2', 3) == ('a', None, 'c')
except AssertionError:
raise TestFailed, "match .group() method"
if verbose: if verbose:
print "Running tests on sre.escape" print "Running tests on sre.escape"
try: p = ""
p="" for i in range(0, 256):
for i in range(0, 256): p = p + chr(i)
p = p + chr(i) test(r"""sre.match(sre.escape(chr(i)), chr(i)) != None""", 1)
assert sre.match(sre.escape(chr(i)), chr(i)) != None test(r"""sre.match(sre.escape(chr(i)), chr(i)).span()""", (0,1))
assert sre.match(sre.escape(chr(i)), chr(i)).span() == (0,1)
pat=sre.compile( sre.escape(p) )
assert pat.match(p) != None
assert pat.match(p).span() == (0,256)
except AssertionError:
raise TestFailed, "sre.escape"
pat = sre.compile(sre.escape(p))
test(r"""pat.match(p) != None""", 1)
test(r"""pat.match(p).span()""", (0,256))
if verbose: if verbose:
print 'Pickling a SRE_Pattern instance' print 'Pickling a SRE_Pattern instance'
@ -248,16 +212,14 @@ try:
except: except:
print TestFailed, 're module cPickle' # expected print TestFailed, 're module cPickle' # expected
try: # constants
assert sre.I == sre.IGNORECASE test(r"""sre.I""", sre.IGNORECASE)
assert sre.L == sre.LOCALE test(r"""sre.L""", sre.LOCALE)
assert sre.M == sre.MULTILINE test(r"""sre.M""", sre.MULTILINE)
assert sre.S == sre.DOTALL test(r"""sre.S""", sre.DOTALL)
assert sre.X == sre.VERBOSE test(r"""sre.X""", sre.VERBOSE)
assert sre.T == sre.TEMPLATE test(r"""sre.T""", sre.TEMPLATE)
assert sre.U == sre.UNICODE test(r"""sre.U""", sre.UNICODE)
except AssertionError:
raise TestFailed, 're module constants'
for flags in [sre.I, sre.M, sre.X, sre.S, sre.L, sre.T, sre.U]: for flags in [sre.I, sre.M, sre.X, sre.S, sre.L, sre.T, sre.U]:
try: try:
@ -270,10 +232,9 @@ if verbose:
# Try nasty case that overflows the straightforward recursive # Try nasty case that overflows the straightforward recursive
# implementation of repeated groups. # implementation of repeated groups.
try: test(r"""sre.match('(x)*', 50000*'x').span()""", (0, 50000), RuntimeError)
assert sre.match('(x)*', 50000*'x').span() == (0, 50000) test(r"""sre.match('(x)*y', 50000*'x'+'y').span()""", (0, 50001), RuntimeError)
except RuntimeError, v: test(r"""sre.match('(x)*?y', 50000*'x'+'y').span()""", (0, 50001), RuntimeError)
print v
from re_tests import * from re_tests import *