Re: [Chicken-users] Syntactic sugar for regular expressions and URIs

[Zbigniew]

Arto,

How about this patch, which extends your syntax to allow arbitrary delimiters?

#;6> ,x #r/\/path\/to\/file:i
(regexp "/path/to/file" #t #f #f)

#;6> ,x #r:/path/to/file:i
(regexp "/path/to/file" #t #f #f)

#;6> ,x #r{/path/to/file}i
(regexp "/path/to/file" #t #f #f)

#;7> (string-split-fields #r:[^/]+: "/path/to/file")
("path" "to" "file")

#;10> (string-match #r/\/path\/to\/file/ "/path/to/file")
("/path/to/file")

#;8> (string-match #r:/path/(to)/file: "/path/to/file")
("/path/to/file" "to")

The drawback over Perl is that nestable delimiters don't nest, so
#r(hi(bye)) is not valid; you have to escape the parentheses with
backslash, or use a different delimiter if you want grouping.

On 6/27/07, Arto Bendiken <address@hidden> wrote:
> First, the `regex-literals' egg [2] provides precompiled regular
> expression literals of the
> form `#/[a-z0-9]+/i', the Perl-like #/.../[ixu] syntax sure to be
> instantly familiar to most everyone:

regex-delim.txt
Description: Text document