URL Parser

zowtar: These snippets of code can be used to check if an URL is valid or not and with some changes it can get URL's from a text.

"In general URI's as defined by RFC 3986 (page 12) may contain any of the following characters: A-Z, a-z, 0-9, -, ., _, ~, :, /, ?, #, ,, @, !, $, &, ', (, ), *, +, ,, ; and =. Any other character needs to be encoded with the percent-encoding (%hh). Each part of the URI has further restrictions about what characters need to be represented by an percent-encoded word." (Gumbo, 2009)


#
#       Checking if an URL is valid or not...
#

set blabla {http://www.ietf.org/rfc/rfc3986.txt}
if {[regexp -- {^(https?://[a-z0-9\-]+\.[a-z0-9\-\.]+(?:/|(?:/[a-zA-Z0-9!#\$%&'\*\+,\-\.:;=\?@\[\]_~]+)*))$} $blabla match url]} {
        puts "$url is a valid url."
}

#
#       Getting an URL from a HTML code...
#

set blabla {<div class="title">Edit <a href="https://wiki.tcl-lang.org/26255">URL Parser</a></div>}
if {[regexp -- {(https?://[a-z0-9\-]+\.[a-z0-9\-\.]+(?:/|(?:/[a-zA-Z0-9!#\$%&'\*\+,\-\.:;=\?@\[\]_~]+)*))} $blabla match url]} {
        puts "$url found in the HTML code."
}

Let's test it and post the results here...

PS: It doesn't work for IPv6 yet.

AMG: You had an extra closing parenthesis at the end of the expression. Fixed. I don't know why you use capturing parentheses; you only want to capture the entire match, which (in your code) already gets stored into $match. In the first expression, where you just test if the entire string matches, $match will just get set equal to $blabla. Another thing: the second expression will start the match anywhere, so "foohttp://a.b/c " is accepted. I suggest using the \m constraint [L1 ] to anchor it at the beginning of a word. Your expression rejects hostnames with capital letters; is this really your intent? It also requires at least one dot in the hostname, even though it's perfectly valid to not have one. For example, http://localhost/test.html , which is the same as http://2130706433/test.html , if you're 1337. ;^)