ksh93 parse_rfc1738_url - parse RFC 1838 URLs
Posted by Anonymous on Wed 19th Jul 2023 14:59
raw | new post


#
# parse_rfc1738_url - parse RFC 1838 URLs
#
# Output variables are named after RFC 1838 Section 5 ("BNF for
# specific URL schemes")
#
function parse_rfc1738_url
{
        set -o nounset

        typeset url="$2"
        typeset leftover
        nameref data="$1" # output compound variable

        # ~(E) is POSIX extended regular expression matching (instead
        # of shell pattern), "x" means "multiline", "l" means "left
        # anchor", "r" means "right anchor"
        leftover="${url/~(Elrx)
                (.+?)                           # scheme
                :\/\/                           # '://'
                (                               # login
                        (?:
                                (.+?)           # user (optional)
                                (?::(.+))?      # password (optional)
                                @
                        )?
                        (                       # hostport
                                (.+?)           # host
                                (?::([[:digit:]]+))? # port (optional)
                        )
                )
                (?:\/(.*?))?/X}"                # path (optional)

        # All parsed data should be captured via eregex in .sh.match - if
        # there is anything left (except the 'X') then the input string did
        # not properly match the eregex
        [[ "$leftover" == 'X' ]] ||
                { print -u2 -f $"%s: Parser error, leftover=%q\n" \
                        "$0" "$leftover" ; return 1 ; }

        data.url="${.sh.match[0]}"
        data.scheme="${.sh.match[1]}"
        data.login="${.sh.match[2]}"
        # FIXME: This should use [[ ! -v .sh.match[3] ]], but ksh93u has bugs
        [[ "${.sh.match[3]-}" != '' ]] && data.user="${.sh.match[3]}"
        [[ "${.sh.match[4]-}" != '' ]] && data.password="${.sh.match[4]}"
        data.hostport="${.sh.match[5]}"
        data.host="${.sh.match[6]}"
        [[ "${.sh.match[7]-}" != '' ]] && integer data.port="${.sh.match[7]}"
        [[ "${.sh.match[8]-}" != '' ]] && data.uripath="${.sh.match[8]}"

        return 0
}

function main
{
        compound c

        input="foo://host/path1/path2" ; printf "## input=%q\n" "$input"
        parse_rfc1738_url c "$input"
        print -v c

        input="foo://myusr@host:14/path1/path2" ; printf "## input=%q\n" "$input"
        parse_rfc1738_url c "$input"
        print -v c

        input="foo://myusr:mypasswd@host:14/path1/path2" ; printf "## input=%q\n" "$input"
        parse_rfc1738_url c "$input"
        print -v c

}

main