[PATCH] minor leak in StartTLS

Brad Fitzpatrick brad at danga.com
Wed Jun 18 05:54:01 UTC 2008

On Tue, Jun 17, 2008 at 3:51 PM, Jacob Burkhart <igotimac at gmail.com> wrote:

> Hey Brad,
> about the code duplication,  isn't a really big part of StartTLS and
> OldSSLClientIn essentially duplicated code?  I saw this and assumed there
> must be some good reason for it, so I did the same with my additions...
> should I try to tackle removing this duplication too?

No, you should never be expected to clean up after more than yourself.  The
general rule is just to not make the situation worse.  Sometimes that
involves cleaning up a _bit_ more than you're responsible for, but I
wouldn't expect you to go merge those two files.

Tests appear to be passing... is there a test that specifically tests SSL?

Not sure.  There should be, but SSL tests might be hard.  Please verify by
hand as well to be sure.

> Is there a faster way to run tests other than:
> perl Makefile.PL && make && make test

"make test" alone is usually sufficient, but yeah... that's basically it.

If a certain test is failing you can use "prove" or "prove -b" on it:

$ prove -b t/foo.t

Or essentially the same thing, but using lib instead of blib:

$ perl -Ilib t/foo.t
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.danga.com/pipermail/djabberd/attachments/20080617/de4ded5e/attachment.htm 

More information about the Djabberd mailing list