X509 Subject Formatting

Hi  , I had the same requirement and it seems the change is the following:

<= TMOS v12 is CN=LASTNAME\, FIRSTNAME,OU=CONTRACTOR,OU=PKI,OU=DEPT,O=COMPANY,C=US

>= TMOS v13 is C=US, O=COMPANY, OU=DEPT, OU=PKI, OU=CONTRACTOR, CN="LASTNAME, FIRSTNAME"

So differences are:

• Reverse Ordered elements (full reverse may be better than defining the order in the irule)
• space after comma separator
• double quotes around fields with special characters (like comma)
• unescaped special character between double quotes

I wrote following procs to manage conversion from v12 to v13 and from v13 to v12 :

proc subject_v13_to_v12 {subject_openssl} {
    # Create the empty subject
    set new_subject ""
    # For each subject element, do the following actions:
    # - remove double quotes arround the value
    # - escape the comma character
    # - remove space around comma separator
    # Insert the value at the begining of the subject (to reverse order)
    foreach {type value} [ split  [regsub -all {(".*?),(.*?")} $subject_openssl "\\1--COMMA--\\2"] ",=" ] {
        set new_subject "{[string trim $type]=[string trim [string map {"--COMMA--" "\\," "+" "\\+" "\\" "\\\\" "<" "\\<" ">" "\\>" ";" "\\;" } [string trim $value {"}]]]} $new_subject"
    }
    # Return the joined values with comma
    return [join  $new_subject ","]
}

proc subject_v12_to_v13 {subject_rfc2253} {
    # Create the empty subject
    set new_subject ""
    # For each subject element, do the following actions:
    # - add double quotes arround the value
    # - unescape the comma character
    # - Add space around comma separator
    # Insert the value at the begining of the subject (to reverse order)
    foreach {type value} [ split  [string map {"\\," "--COMMA--"} $subject_rfc2253] ",=" ] {
        set new_subject "{[string trim $type]=[expr {[string match "*--COMMA--*" $value] ? "\"[string trim [string map {"--COMMA--" ","} $value]]\""  : [string trim $value] }]} $new_subject"
    }
    # Return the joined values with comma and space
    return [join  $new_subject ", "]
}

I got the following results:

 subject_v12_to_v13 {CN=F5lab\, local,OU=TEST,O=Foo,L=Bar,ST=London\, City of,C=GB}
C=GB, ST="London, City of", L=Bar, O=Foo, OU=TEST, CN="F5lab, local"
 
% subject_v13_to_v12 {C=GB, ST="London, City of", L=Bar, O=Foo, OU=TEST, CN="F5lab, local"}
CN=F5lab\, local,OU=TEST,O=Foo,L=Bar,ST=London\, City of,C=GB

Updated to work better with commas and made less greedy with the use of ? in the regsub regex

OK I understand now, the regsub -all is greedy so it matches as much as it can.

Thanks again both of you for your help with that. All is good now

 : The regsub try to match the largest pattern...

The first match string is:

​

London, City of", L=Bar, O=Foo, OU=TEST, CN="F5lab

​

The second is

​

local"

​

So you have to force stoping at the first "... replace the .* with [^"]*

The problem with this expression is it will also match this string:

f", L=Bar, O=Foo, OU=TEST, CN="

This expression seems to work with your example:

% set subject [regsub -all {([a-zA-Z]+ ?= ?"[^"]+),([^"]+")} $subject "\\1--COMMA--\\2"]

C=GB, ST="London--COMMA-- City of", L=Bar, O=Foo, OU=TEST, CN="F5lab--COMMA-- local"

It works but there's just one small issue (might be a bug?). BTW my version is 13.1.3 build 0.0.6

Here's the relevant code

set subject [regsub -all {(".*),(.*")} $subject "\\1--COMMA--\\2"]
 
log local0. "Comma replaced: $subject"

In case there are 2 fields with commas in the subject it only replaces second comma occurrence inside double-quotes. This is what I get

Nov 20 01:24:26 example.lab info tmm5[12408]: Rule /Common/irule-subject-reverse <HTTP_REQUEST>: Original Subject: C=GB, ST="London, City of", L=Bar, O=Foo, OU=TEST, CN="F5lab, local"

Nov 20 01:24:26 example.lab info tmm5[12408]: Rule /Common/irule-subject-reverse <HTTP_REQUEST>: Comma replaced: C=GB, ST="London, City of", L=Bar, O=Foo, OU=TEST, CN="F5lab--COMMA-- local"

I believe the regsub -all option should replace all the occurrences of the comma(,) in the string right?

Any ideas what might be wrong?

Thank you both, I've tried replace the comma inside "" with regsub but couldn't make it work. Now it works as expected

Hi uniqng, I have changed it so that it handles commas within quotes. Tried it with your example and it seems to work so maybe you can try it out yourself.

 it doesn't seem to work, can you explain how it suppose to work? What are the different string maps suppose to do?

Please note that the string with comma is enclosed withing double quotes and not escaped with \

i.e full string from x509 command is

C=US, OU=Example, O=Example2, CN="Example, Inc"