cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sylvain Wallez <sylvain.wal...@anyware-tech.com>
Subject Re: [VOTE] allow redirects in handle-errors
Date Wed, 30 Apr 2003 14:40:22 GMT
Steven Noels wrote:

> On 30/04/2003 15:35 Nicola Ken Barozzi wrote:
>
>>
>> Steven Noels wrote, On 30/04/2003 15.12:
>>
>>> On 30/04/2003 14:58 Nicola Ken Barozzi wrote:
>>>
>>>> The vote here is about allowing redirects.
>>>
>>>
>>>
>>> I hope that this nuance is appreciated by everyone.
>>
>>
>>
>> Could you please explain what you mean :-?
>
>
> I mean no harm :-)
>
> The use case as described by Björn is about more than only plain 
> redirects, but also about supporting application-level exception 
> handling in handle-errors.
>
> IIUC, you and most of us would not be opposing a simple off-base 
> redirect, but many seem to have adverse feelings towards anything more 
> than that. So I was wondering whether Sylvain, in his vote, means 
> "only redirects, but no app logic", which means we should check on 
> that in code, or "let's give Björn what he wants", which apparenly is 
> not what others would like to see happen (me neither, but I prefer to 
> refrain from making a technical judgement call until I see you - more 
> enlightened - guys come to an agreement).
>
> Since this nuance in Sylvain's vote proposal was not immediately clear 
> to myself, I thought it was worth mentioning.


Humans are (still) more clever than computers, and although we humans 
can make such a difference, having it done by the sitemap engine would 
be much more difficult, even if we can find some heuristics such as "is 
it a cocoon: or localhost?". And the same question applies to actions 
which can be used both for app-logic and to e.g. log the error or send 
it to a sysadmin.

So there was actually no nuance in the vote proposal : do we accept 
redirects in error-handlers, just as we accept other components, be they 
used correctly or not, with all the subjectivity that this "correctly" 
implies, which may seem like an arbitrary constraint to people that 
don't happen to think the "official Cocoon way".

Sylvain

-- 
Sylvain Wallez                                  Anyware Technologies
http://www.apache.org/~sylvain           http://www.anyware-tech.com
{ XML, Java, Cocoon, OpenSource }*{ Training, Consulting, Projects }



Mime
View raw message