From: Jean-Baptiste Quenot <[EMAIL PROTECTED]>
Date: Mon, 16 Oct 2006 14:59:05 +0200

* Antonio Gallardo (JIRA):
>      [ http://issues.apache.org/jira/browse/COCOON-1682?page=all ]
>
> Antonio Gallardo reopened COCOON-1682:
> --------------------------------------
>
>       Assignee: Cocoon Developers Team
>
> It is a user wish. We should not close it.

Antonio,

Do  you plan  to work  on adding  random headers  support to  this
transformer?  If you don't there is no point at reopening this bug
IMHO.

Your opinion; obviously not his (nor mine for that matter). It's an enhancement request from an end user, who may or may not be a competent java developer. Closing enhancement requests as WONTFIX because an end user can't provide a patch is like cancelling airline flights because the passengers can't fly the plane themselves!

On the other hand, if there was a good technical reason why the functionality can't be added, or if it was going to conflict with planned future changes (e.g. the block is deprecated and due to be deleted in the next version), or if there were some other way to achieve the same result, then that's another matter. But that doesn't appear to be the case here. Who's to say nobody will submit a patch at some point in the future? Now you might argue that in that case they could reopen the issue, but I know that if I had some time on my hands and wanted to contribute an occasional patch or bug fix, I wouldn't bother searching closed issues looking for something to do. I'd look through the outstanding open issues, and I don't suppose I'm alone in that. So by closing these requests you're making it even less likely that they will ever get implemented.

I could file JIRA issues for very interesting wishes like:

        Implement Stax-based pipelines in Cocoon

Closing that (or at least deferring it to Cocoon 3 or 4) would be justifiable on technical grounds since it would probably require a complete rewrite.

Or:

        Allow Cocoon pipelines to be reusable from Java Code.

Can be justifiably closed since it's already possible - set up the necessary environment & context then create & initialise a Cocoon (or was it a CocoonBean, I forget) object. Sample code can be found in the CocoonServlet or Main classes...

Do you think it makes sense to leave this issue open if noone is
likely to be working on it and no patch is attached?

Sure.  At least then there's still a possibility it'll happen.


Andrew.


Best regards,
--
     Jean-Baptiste Quenot
aka  John Banana Qwerty
http://caraldi.com/jbq/


Reply via email to