Email addresses case insensitive (RT 2.0.11)

Hi.

I’ve come across this problem with RT, that email address are case
insensitve, eg, bob@company.com == BOB@company.com , this is causing some
problems with some of our clients.
So, how can I make RT see these two addresses as being different users?
Looking in Record.pm shows that in LoadByCols it makes everything lowercase
if RT is connected to a “CaseSensitive” database, but I assume this is just
the column names, not the actual data?

Any pointers in the right direction would be greatly appreciated

Regards
Matt Watson
Netspace Online Systems.

Matthew Watson wrote:

I’ve come across this problem with RT, that email address are case
insensitve, eg, bob@company.com == BOB@company.com , this is causing some
problems with some of our clients.
So, how can I make RT see these two addresses as being different users?
Looking in Record.pm shows that in LoadByCols it makes everything lowercase
if RT is connected to a “CaseSensitive” database, but I assume this is just
the column names, not the actual data?

Are you saying that your client has two addresses:

bob@company.com

and

BOB@company.com

and they are two different people?

Your client has a very broken email system.

I can’t find the exact quite in RFC2822 (or related) which defines
case insensitivity, but it is generally accepted (and assumed) that
email addresses (like domain names) are case insensitive. Any system
which violates this is going to be very unhappy.

-R

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

  • Robert Spier [2003-04-23 13:13]:

Are you saying that your client has two addresses:

bob@company.com

and

BOB@company.com

and they are two different people?

Technically, bob and BOB refer to mailboxes, which is not the same
thing as people.

Your client has a very broken email system.

Well…

I can’t find the exact quite in RFC2822 (or related) which defines
case insensitivity, but it is generally accepted (and assumed) that
email addresses (like domain names) are case insensitive. Any system
which violates this is going to be very unhappy.

RFC2822 specifies that the left hand side (the “local part”) is
implementation specific; specifically, see section 3.4.1. Sendmail’s
implementation (the most common, and, historically speaking, the primary
reference implementation) falls back to case-insensitive local-part
matching if a literal match is not found.

(darren)


Don’t be ashamed to say what you are not ashamed to think.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (SunOS)

iD8DBQE+ps7BzsinjrVhZaoRApo5AJ4nBqNho3/CiDeeH3QKdBkrwZrf5gCfYLl3
ELsSd20dsAINVBmGR0JP6yw=
=aeLD
-----END PGP SIGNATURE-----

somewhat…

we have the situation that

bob@company.com

and

BOB@company.com

are different customers, BOB@company.com is not really an email address but
rather an indentifier for that customer within our primary customer
database. So to link between RT and our customer Database we use the
"Requestor" field, which naturally fails horribly if two customers have the
same name but different casing.

What I would really like to do is link our database into RT using our
internal userid’s, but I don’t have alot of time for RT hacking
unfortunately.

Matt Watson
Netspace Online Systems.–On Wednesday, 23 April 2003 1:34 PM -0400 darren chamberlain darren@boston.com wrote:

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

  • Robert Spier [2003-04-23 13:13]:

Are you saying that your client has two addresses:

bob@company.com

and

BOB@company.com

and they are two different people?

Technically, bob and BOB refer to mailboxes, which is not the same
thing as people.

Your client has a very broken email system.

Well…

I can’t find the exact quite in RFC2822 (or related) which defines
case insensitivity, but it is generally accepted (and assumed) that
email addresses (like domain names) are case insensitive. Any system
which violates this is going to be very unhappy.

RFC2822 specifies that the left hand side (the “local part”) is
implementation specific; specifically, see section 3.4.1. Sendmail’s
implementation (the most common, and, historically speaking, the primary
reference implementation) falls back to case-insensitive local-part
matching if a literal match is not found.

(darren)


Don’t be ashamed to say what you are not ashamed to think.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (SunOS)

iD8DBQE+ps7BzsinjrVhZaoRApo5AJ4nBqNho3/CiDeeH3QKdBkrwZrf5gCfYLl3
ELsSd20dsAINVBmGR0JP6yw=
=aeLD
-----END PGP SIGNATURE-----


rt-users mailing list
rt-users@lists.fsck.com
http://lists.fsck.com/mailman/listinfo/rt-users

Have you read the FAQ? The RT FAQ Manager lives at http://fsck.com/rtfm