You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@cocoon.apache.org by Peter Velychko <v_...@ukr.net> on 2004/07/01 17:44:31 UTC
Caching full pages from flowscript (was Re: cocoon and hibernate sessions)
Hello Upayavira,
Thursday, July 1, 2004, 4:55:27 PM, you wrote:
> Jeremy Quinn wrote:
>>
>> On 1 Jul 2004, at 14:30, Peter Velychko wrote:
>>
>>> Hello Upayavira,
>>>
>>> Thank you for your reply.
>>>
>>>> That could work, but it would be working round, rather than working
>>>> with, Cocoon. You would also be buffering your output, which is best
>>>> avoided.
>>>
>>> Yes, this is workaround. But why whole DOM tree caching is so bad idea?
>>
>>
>> off the top of my head :
>>
>> 1) it makes lazy-initialisation useless, because I assume everything
>> would always get read from the DB regardless of whether you need it in
>> the view-layer or not
>>
>> 2) I (for one) have data models that if they are not lazy initialised
>> would actually load every record in the database, because everything
>> is ultimately linked.
>>
>> 3) you could end up with circular dependencies from (2) that make an
>> infinitely large DOM
>>
>> 4) there are much better ways of doing it IMHO ;)
>>
>> So it would depend on your data model as to how well this worked.
>>
> I don't think it does make lazy initialisation useless - as it produces
> the page, with processPipelineTo. Which is okay. It would work. It is
> more a question of storing the whole page in memory not being very
> advisible. Best avoided. It avoids the pipeline's SAX streaming benefit.
It's also possible to use caching system to store response content.
I think it may be used for pages that have high frequency of requests
to them (a la "pseudostatic" pages).
> Upayavira
--
Best regards,
Peter Velychko
v_peter@ukr.net
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@cocoon.apache.org
For additional commands, e-mail: users-help@cocoon.apache.org