[CentralOH] 2014-01-17_道場_Scribbles_ 落書/惡文? Much more

[Eric Floehr]

>
> re.split("(\s|\.|\!|\?)+", "Blah. Blah!Blah! Blah Blah?Blah")
>
> Results in:
>
> ['Blah', ' ', 'Blah', '!', 'Blah', ' ', 'Blah', ' ', 'Blah', '?', 'Blah']
>
> Where as in Java:
>
> "Blah. Blah!Blah! Blah Blah?Blah".split("(\\s|\\.|\\!|\\?)+")
>
> Results in:
>
> [Blah,Blah,Blah,Blah,Blah,Blah]
>
>
> Don't blame me. I didn't break it.
>


Parens are a "capturing group" versus brackets which are a "character
class".  Python's split will "capture" matched groups in the split, whereas
Java doesn't. In other contexts, using parens indicates to the regex engine
you'd like to keep the match to refer to it later.

Perhaps Python's implementation is less broken in that regard.

To be correct in both Java and Python, this regex should be using brackets.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/centraloh/attachments/20140121/a00d6665/attachment.html>