Question

我试图在沙捞越时 with,但必须满足任何特性,避免连续三个或三个以上的 com子或半殖民地。换言之,只允许连续两度 com或半殖民地。

因此,这是我目前拥有的:

^(,|;){,2}([^,;]+(,|;){,2})*$

似乎按预期开展工作:

>>> r.match(  )
<_sre.SRE_Match object at 0x7f23af8407e8>
>>> r.match( foo, )
<_sre.SRE_Match object at 0x7f23af840750>
>>> r.match( foo, a )
<_sre.SRE_Match object at 0x7f23af8407e8>
>>> r.match( foo, , )
<_sre.SRE_Match object at 0x7f23af840750>
>>> r.match( foo, ,,a )
<_sre.SRE_Match object at 0x7f23af8407e8>
>>> r.match( foo, ,,, )
>>> r.match( foo, ,,,; )
>>> r.match( foo, ,, ;; )
<_sre.SRE_Match object at 0x7f23af840750>

但是,随着我开始增加投入案文的篇幅,似乎需要更多时间来作出答复。

>>> r.match( foo, bar, baz,, foo )
<_sre.SRE_Match object at 0x7f23af8407e8>
>>> r.match( foo, bar, baz,, fooooo, baaaaar )
<_sre.SRE_Match object at 0x7f23af840750>
>>> r.match( foo, bar, baz,, fooooo, baaaaar, )
<_sre.SRE_Match object at 0x7f23af8407e8>
>>> r.match( foo, bar, baz,, fooooo, baaaaar,, )
<_sre.SRE_Match object at 0x7f23af840750>
>>> r.match( foo, bar, baz,, fooooo, baaaaar,,, )
>>> r.match( foo, bar, baz,, fooooo, baaaaar,,,, )
>>> r.match( foo, bar, baz,, fooooo, baaaaar, baaaaaaz,,,, )

And finally it gets completely stuck at this stage and the CPU usage goes up to 100%.

I m not sure if the regexp could be optimized or there s something else involved, any help appreciated.

Answer 1

You re running into catastrophic backtracking.

The reason for this is that you have made the separators optional, and therefore the [^,;]+ part (which is itself in a repeating group) of your regex will try loads of permutations (of baaaaaaaz) before finally having to admit failure when confronted with more than two commas.

RegexBuddy aborts the match attempt after 1.000.000 steps of the regex engine with your last test string. Python will keep trying.

Imagine the string baaz,,,:

在座各位,游艇发动机必须检查所有这些:

baaz,,<failure>
baa + z,,<failure>
ba + az,,<failure>
ba + a + z,,<failure>
b + aaz,,<failure>
b + aa + z,,<failure>
b + a + az,,<failure>
b + a + a +z,,<failure>

在宣布全面失败之前。参看这种增长如何与每一种额外性成倍增长?

如同这种放弃一样,可以避免拥有qu子或原子集团,它们都不幸得不到目前 Python雷发动机的支持。但你可以很容易地进行反常检查:

if ",,," in mystring or ";;;" in mystring:
    fail()

without needing a regex at all. If ,;, and the likes could also occur and should be excluded, then use Andrew s solution.

Answer 2

我认为,你想要做的是:

^(?!.*[,;]{3})

This will fail if the string contains three or more , or ; in a row. If you actually want it to match a character add a . at the end.

This utilizes negative lookahead, which will cause the entire match to fail if the regex .*[,;]{3} would match.

Answer 3

Try this regular expression:

^([^,;]|,($|[^,]|,[^,])|;($|[^;]|;[^;]))*$

It matches repetitively:

one single character that is neither , nor ;, or
a , that is either not followed by another , or a ,, that is not followed by another ,, or
a ; that is either not followed by another ; or a ;; that is not followed by another ;

until the end is reached. It is very efficient as it fails early without doing much backtracking.

Answer 4

How about this idea match the ones that have the pattern you don t want ".+,,," In Python just keep those that do not match. Should be fast

友情链接