Looking for iRule to convert non-ascii character to ascii

That's what worked for me:

proc hex2ascii {sHex} {
    # latin-1 supplement, unicode u+0000 - u+00ff
    set iByte 0
    # check if string begins with '0x'
    if { [scan $sHex {0x%s} sHex] > 0} {
        # loop each character
        foreach sChar [ split [binary format H* $sHex] ""] {
            # convert to decimal
            scan $sChar %c iChar
            if {$iChar < 0x80} {
                # standard ascii
                append sAscii $sChar
            } elseif {$iChar > 0xc2} {
                set iByte [expr ({$iChar} - 0xc2) * 0x40]
            } elseif  {$iChar < 0xc2} {
                # convert to ascii
                append sAscii "[format %c [expr {$iByte} + {$iChar}]]"
                set iByte 0
            }
        }
        return $sAscii
    } else {
        return $sHex
    }
}

Calling the function with the following attribute will return ASCII

[call hex2ascii 0x434e3d....]

Hope this helps.

Are these characters actually supported by ASCII? If not, they can't be converted into ASCII surely?

This is a tough one for a newbie to be facing.

Something like this might work:

set hex_string [what_ever_you_use_to_get_the_hex_dn]
set some_variable [binary format a* [encoding convertto utf-8 [$hex_string]]
log local0. "The converted string is ${some_variable}"

The string, scan and expr commands can also be used in some fashion but without a way to test I can't confidently say what will work I'm afraid.

In 12.0.0

ID 399693:

t is now possible to use the -decode option for mcget command of a branch rule to decode a session variable before using it. When you create an agent and add a branch rule, the default value of the rule contains an mcget command to fetch the session variable. By default, the session variable is HEX encoded if it contains non-ASCII characters. You need to modify the command in advanced mode and insert the -decode option for mcget command, for example: expr { [mcget -decode {session.ad.last.attr.memberOf}] contains "non-ASCII-characters" }