geronimo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rick McGuire (JIRA)" <>
Subject [jira] Commented: (GERONIMO-4135) Too many CRLF at the end of the message
Date Mon, 23 Jun 2008 14:54:45 GMT


Rick McGuire commented on GERONIMO-4135:

I've finally managed to hunt down the source of the extra line feeds, and it appears to be
coming from the James SMTP code.  In the DataCmdHandler class, the message is getting created
using the following code:  

            mail =
                new MailImpl(session.getConfigurationData().getMailServer().getId(),
                             (MailAddress) session.getState().get(SMTPSession.SENDER),
                             new SequenceInputStream(new SequenceInputStream(headersIn, msgIn),
                                     new ReaderInputStream(new StringReader("\r\n"))));

The outer-most SequenceInputStream() on the last argument is adding an extra CRLF sequence
to the stream used to create the message.  Since the content data handler reads the from the
stream until it hits the EOF, the extra CRLF gets attached to the end.  I can't find anything
in the specifications that indicates a trailing CRLF should be discarded (and indeed, when
the message was originally created, it wasn't), nor can I seem to find any reason why this
is working with the Sun implementation.  Removing that extra SequenceInputStream makes the
problem go away, but I suspect a problem would then show up with the Sun implementation. 
At this point, I'm at a loss as to what should be fixed.  

> Too many CRLF at the end of the message
> ---------------------------------------
>                 Key: GERONIMO-4135
>                 URL:
>             Project: Geronimo
>          Issue Type: Bug
>      Security Level: public(Regular issues) 
>          Components: mail
>    Affects Versions: 2.1.1
>            Reporter: Stefano Bagnara
>            Assignee: Rick McGuire
> We have a lot of tests trying to send a message to james and reading the result: most
of the tests fail because the result contains 2 trailing CRLF more than expected (we send
one crlf at the end of the message we receive 3 crlf).
> I don't exactly know where does it happen but this does not with sun implementation.

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message