Re: Is there a more efficient way to remove data from a string?

On Fri, Jul 15, 2011 at 10:52:54AM +0900, Josh Cheek wrote:
>=20
> Depending on how the string is to be interpreted, you may need to escape =
it
> before creating it:
>=20
> str =3D "a.b"
> /#{str}/ # =3D> /a.b/
> Regexp.new str # =3D> /a.b/
> Regexp.new Regexp.escape str # =3D> /a\.b/

That's a very good point.  Things get tricky when you start stuffing
strings into regexen and so on.  If you (in general -- not Josh Cheek,
per se) are not yet familiar with the various unit testing libraries
available for Ruby, now might be a good time to familiarize yourself at
least with the basics of Test::Unit, the Ruby standard library's
equivalent to NUnit.


>=20
> Depending on how big the files are, efficiency may be very important. In
> that case, you might benchmark a few different approaches and see which is
> best. Here is another possibility:
> ARGF.each_line do |line|
>   index =3D line.index regex
>   line.slice!(index..-1) if index
>   puts line
> end

This looks to me like it is likely to be more resource-intensive than my
example, thanks to the sheer number of operations involved, and the fact
that I do not think .* would impose too much additional overhead on the
regex match -- though maybe the interpolation in my version would be
somewhat expensive.  Then again, I don't know much about the
implementation of methods that come with the language, so don't just take
my word for it.  As Josh suggests, you should at least loosely benchmark
performance of various implementations if you find your first attempt is
not performing well enough for your needs.

--=20
Chad Perrin [ original content licensed OWL: http://owl.apotheon.org ]