13
13
# You should have received a copy of the GNU General Public License
14
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
17
"""Lazily compiled regex objects.
19
This module defines a class which creates proxy objects for regex
20
compilation. This allows overriding re.compile() to return lazily compiled
23
We do this rather than just providing a new interface so that it will also
24
be used by existing Python modules that create regexs.
19
This module defines a class which creates proxy objects for regex compilation.
20
This allows overriding re.compile() to return lazily compiled objects.
32
class InvalidPattern(errors.BzrError):
34
_fmt = ('Invalid pattern(s) found. %(msg)s')
36
def __init__(self, msg):
40
26
class LazyRegex(object):
41
27
"""A proxy around a real regex, which won't be compiled until accessed."""
43
30
# These are the parameters on a real _sre.SRE_Pattern object, which we
44
31
# will map to local members so that we don't have the proxy overhead.
45
32
_regex_attributes_to_copy = [
46
'__copy__', '__deepcopy__', 'findall', 'finditer', 'match',
47
'scanner', 'search', 'split', 'sub', 'subn'
33
'__copy__', '__deepcopy__', 'findall', 'finditer', 'match',
34
'scanner', 'search', 'split', 'sub', 'subn'
50
37
# We use slots to keep the overhead low. But we need a slot entry for
51
38
# all of the attributes we will copy
52
39
__slots__ = ['_real_regex', '_regex_args', '_regex_kwargs',
53
] + _regex_attributes_to_copy
40
] + _regex_attributes_to_copy
55
def __init__(self, args, kwargs):
42
def __init__(self, args=(), kwargs={}):
56
43
"""Create a new proxy object, passing in the args to pass to re.compile
58
:param args: The `*args` to pass to re.compile
59
:param kwargs: The `**kwargs` to pass to re.compile
45
:param args: The *args to pass to re.compile
46
:param kwargs: The **kwargs to pass to re.compile
61
48
self._real_regex = None
62
49
self._regex_args = args
72
59
def _real_re_compile(self, *args, **kwargs):
73
60
"""Thunk over to the original re.compile"""
75
return re.compile(*args, **kwargs)
77
# raise InvalidPattern instead of re.error as this gives a
78
# cleaner message to the user.
79
raise InvalidPattern('"' + args[0] + '" ' + str(e))
81
def __getstate__(self):
82
"""Return the state to use when pickling."""
84
"args": self._regex_args,
85
"kwargs": self._regex_kwargs,
88
def __setstate__(self, dict):
89
"""Restore from a pickled state."""
90
self._real_regex = None
91
setattr(self, "_regex_args", dict["args"])
92
setattr(self, "_regex_kwargs", dict["kwargs"])
61
return _real_re_compile(*args, **kwargs)
94
63
def __getattr__(self, attr):
95
64
"""Return a member from the proxied regex object.
109
78
:return: a LazyRegex proxy object.
111
80
return LazyRegex(args, kwargs)
83
def install_lazy_compile():
84
"""Make lazy_compile the default compile mode for regex compilation.
86
This overrides re.compile with lazy_compile. To restore the original
87
functionality, call reset_compile().
89
re.compile = lazy_compile
93
"""Restore the original function to re.compile().
95
It is safe to call reset_compile() multiple times, it will always
96
restore re.compile() to the value that existed at import time.
97
Though the first call will reset back to the original (it doesn't
100
re.compile = _real_re_compile
103
_real_re_compile = re.compile
104
assert _real_re_compile is not lazy_compile, \
105
"re.compile has already been overridden as lazy_compile, but this would" \
106
" cause infinite recursion"