So this will make my planned IMAP4 implementation easier.

I'm running on w32 and planning to go Linux (mail is my only legacy W32
platform) and then implement a native Linux IMAP4 server, but I remember
lots of work arounds for the CRLF/LF issue, so I was not exited about the
idea of implementing it.

Davide, I assume I will need to convert any existing users mail to LF format
for this migration to work?
That is, work with IMAP4 server - since Xmail will read either format.

Also when migrating from W32 to Linux - the default mail store format
changes.
Can you confirm that I will need to over ride xmail's default Linux mail
format to be compatible with any migrated mail.
 

Rob :-)
 
_________________________________________________
It might look like I'm doing nothing, but on a cellular level, I'm quite
busy.

-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
Behalf Of Davide Libenzi
Sent: Wednesday, September 03, 2008 6:22 AM
To: XMail mailing list
Subject: [xmail] 1.26-pre03

I made 1.26-pre03:

http://www.xmailserver.org/xmail-1.26-pre03.tar.gz
http://www.xmailserver.org/xmail-1.26-pre03.win32bin.zip


Changelog:

- Changed the line termination used to store messages into the mailboxes
  of Unix versions of XMail. Now messages are stored with the OS native LF
  termination, instead of the RFC CRLF that was used before.
  This allows other Unix softwares working together with XMail to not be
  confused by the extra CR present in the line termination.

- The "smtprelay" behaviour with respect to 5xx responses from one of the 
  servers in the relay list, has been changed. Now a 5xx response will 
  stop the processing, instead of iterating on the remaining servers of 
  the list.

- Avoid to crawl all the USER.TAB file during a "userlist" CTRL command.

- Fixed a bug that allowed non-RFC characters inside domain names.

- Fixed OSX Leopard build error.

- Added "timeo" option to flags execution.

- Added "NoAuth", "EnableVRFY" and "EnableETRN" settings inside IP
properties.


The biggest change is the first. Since when XMail started delivering to 
Maildir (long time ago), people started using XMail together with other 
softwares that can get confused by the extra CR. How can this happen is 
beyond my grasp, since handling both cases requires like one line of extra 
code. But anyway.
One of the reasons why I chose to save the mailbox file in CRLF form, was 
due to the fact that the message could have been delivered with fast OS 
primitives, if already in CRLF (and hence RFC) form. This avoided creating 
extra copies of the message file during the SMAIL processing.
In order to shove an extra file copy during the processing, I changed the 
way temporary file are created when it comes to SMAIL local delivery.
Before such files were created inside the system temporary directory (/tmp 
or %TEMP%), and then a copy+delete was done to move the message inside the 
mailbox. The copy+delete was needed, insated of a standard OS move/rename, 
because the mailbox mount/drive coulad have been located in another 
mount/drive WRT the temporary directory.
Now things are changed and XMail will use a "tmp" directory inside 
MAIL_ROOT.
What does it change for you? If you don't mount domain subdirectories to 
other mount/drives (like 99% of XMail users), nothing. Xmail will 
automatically create the "tmp" subdirectory and the switch to the new 
version will be transparent.
If you happen to mount domain subdirectories onto other mount/drives WRT 
MAIL_ROOT, you simply have to remember to create a ".tmp" directory at the 
root of the directory mount/drive.




- Davide


-
To unsubscribe from this list: send the line "unsubscribe xmail" in
the body of a message to [EMAIL PROTECTED]
For general help: send the line "help" in the body of a message to
[EMAIL PROTECTED]

-
To unsubscribe from this list: send the line "unsubscribe xmail" in
the body of a message to [EMAIL PROTECTED]
For general help: send the line "help" in the body of a message to
[EMAIL PROTECTED]

Reply via email to