ukma_url_parser 0.1.0

/// Matches an empty character
WHITESPACE = _{ " " }

/// Parses a sequence of digit characters
number = {      
    ASCII_DIGIT+
}

/// Matches any latin letter
letter = { 'a'..'z' | 'A'..'Z' }

/// Parses the scheme/protocol of the URL. Supports only URI schemes registered with the IANA 
scheme = { 
    ^"file"
    | ^"ftp"
    | ^"http"
    | ^"https"
    | ^"imap"
    | ^"irc"
    | ^"nntp"
    | ^"acap"   
    | ^"icap"   
    | ^"mtqp"
    | ^"wss"
}

/// Parses a sequence of letters
word = { letter+ }

/// Parses a sequence of characters that can contain a letter or a number
word_with_numbers= { (letter | ASCII_DIGIT)+ }

/// Parses a URL label, which is a part of domain name separated by [.]
label = ${
    word ~ ("-" ~ word_with_numbers)*
}

///  Parses the domain name of the URL, which is the main part of it
domain_name = ${
	label ~ ("." ~ label)* 
}

/// Parses a port value. Should be a number that contains no more than 6 symbols
port = {
    ASCII_DIGIT{1, 6}
}

/// Parses the path segments after the domain and port 
paths = ${
    ("/" ~ word_with_numbers)+
}

/// Parses key-value pairs in parameters, separated by '='
key_value_pair = {
    word_with_numbers ~ "=" ~ word_with_numbers
}

/// Parses the URL parameters, using key-value structure. Can have a couple of members, separated by '&'
parameters = {
    key_value_pair ~ ("&" ~ key_value_pair)*
}

/// Parses the fragment
fragment = {
    word_with_numbers
}

/// Parses the entire URL structure
/// # Input examples
/// * http://example.com
/// * file://www.sub.example.org:8080/resource/resource2?query1=value1&query2=value2#section1
/// * ftp://sub.domain.example.com/path1/path2?param=value
url = {
    SOI ~ scheme ~ "://" ~ domain_name ~ (":" ~ port)? ~ (paths)? ~ ("?" ~ parameters)? ~ ("#" ~ fragment)? ~ EOI
}