On Jun 29, 2006, at 4:57 PM, Christian Neukirchen wrote:

> Eric Hodel <drbrain / segment7.net> writes:
>
>> On Jun 28, 2006, at 1:10 PM, A. S. Bradbury wrote:
>>
>>> On Wednesday 28 June 2006 20:56, Pete wrote:
>>>> just too bad the rdoc is inaccurate and says that delete takes a
>>>> string
>>>> as argument and not a regexp
>>>
>>> That's actually correct. At first glance I thought it just
>>> constructed a regex
>>> from the given string (due to its understanding of character  
>>> ranges).
>>> Instead, both String#count and String#delete take this kind of
>>> pseudo-regex.
>>> Is there a reason it wouldn't make more sense for these methods to
>>> take a
>>> regexp?
>>
>> #delete and #count are restricted to character lists, a full regular
>> expression is too much.
>
> +1 for adding these to a future Ruby.

We have those now.  They are called gsub() and scan().  ;)

James Edward Gray II