7 messages in net.sourceforge.lists.courier-users[courier-users] Re: Accepting alterna...
FromSent OnAttachments
Rodrigo SeveroJun 16, 2003 7:48 am 
Sam VarshavchikJun 16, 2003 8:43 am 
Peter C. NortonJun 16, 2003 9:12 am 
Peter C. NortonJun 16, 2003 9:47 am 
Rodrigo SeveroJun 16, 2003 11:54 am 
Randy's OpenBSD IDJun 16, 2003 2:24 pm 
Rodrigo SeveroJun 17, 2003 4:51 am 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:[courier-users] Re: Accepting alternative ATs in email addresses during login - patchActions...
From:Rodrigo Severo (rodr@fabricadeideias.com)
Date:Jun 17, 2003 4:51:14 am
List:net.sourceforge.lists.courier-users

Randy's OpenBSD ID wrote:

This has _all_ been done before.

Haven't everything? It's just a question of looking for it.

I posted a patch to 'authmoduser.c' which, among other things' has a single (read that one (1) ) codeline ditty to do exactly what you are saying. That is, grok that an address has a ":" or a "%" instead of the correct "@" - change it to "@" and go on into the rest of the code.

Scan back through this mailing list and look for 'authmoduser.c' and you will find it. Cut out the other cool stuff (that lets you determine the TCPLOCALHOST --> DOMAIN mapping and you'll have one sweet line of code that does what you want...

Kind of... I see that the one line of code I used that actually does the "@" change came from your patch but your patch is an overkill for this problem. Why?

Because that 1 line of code won't work by itself: if I remove the rest of your patch and keep that 1 line of code, it simple won't work because authdata won't be parsed in the first place. To keep the authdata parsing just to change the "@" is an overkill in my opinion.

What I am trying to do is include this option (fixing "@") in the main Courier code. If you ask me, I think your code should get merged instead of mine, it includes an extra feature in Courier that seems really usefull and messes with one single file only but I also prefer that one of the patches get merged instead of none so...