Gerard van Enk <[EMAIL PROTECTED]> wrote:

> #6372          Open    Bug     High    1.7-rc2         Database Layer          new 
> storage
> does not create primary keys/unique indices on number field


I think it has to do with postgresql.,la nd it's quite tricky.  Snce
automatic generation of indices was always a problem, and I don't think it
is worse now than it used te be. So it needs not be a show-stopper.

> 
> #6286         Open    Bug     High    1.7     Database Layer          behaviour of 
> max to limit
> number of records is broken

This is related to the remark of eduard: old database layers are not or
badly tested.

If they don't work perhaps a remark in the release-notes would suffice, but
we can in that case just as well remove them immediately (what's the point
of waiting until 1.8, if then don't work now...)

There need to be clarity about this issue before the release. I would not be
against dropping support for the 'old' database layers now, but I do not
think that was the original intention.


> #6214          Open    Bug     High    1.7     Database Layer          storage does
> not create indices if forein keys are not supported

I think, this bug might have been fixed.


> ======== open (medium) bugs ========
> 
> #6367          Open    Bug     Medium          1.6.5           Taglibs         Poor 
> performance
> of JSP code in body of cloud tag.
> 
> Does this apply to 1.7.0 also?

I suppose so, but I do think it is actually a problem of app-servers which
do not support EVAL_BODY_INCLUDE (at least, most don't), so
EVAL_BODY_BUFFERED is used, also when that it not needed.

Perhpas it should be configurable what it uses, so that you can switch it
on if your app-server happens to support it, and gain performance. Since it
is not worse then in previous releases, I'd say it is not a show-stopper.
The said configuration proposal could perhaps be implemented after the
release.


> #6366         Open    Bug     Medium  1.7     Applications    email builder of the 
> email
> application does not override the replace method for SCAN

How well was the email-application tested, it is actually ready for release
at all?


> #6321         Open    Wish    Medium  1.7.0   Database Layer          It's a bit 
> cumbersome
> to turn on 'stores-binary-as-file' and hard to configure where

Cumbersome, but possible.  Not a show-stopper.


> #6278         Open    Bug     Medium  1.7     Database Layer          SearchQuery: 
> it is
> possible to create logical queries which don't work.

It's possible, but hard, and e.g. taglibs don't. No show-stopper.

> 
> #6259         Open    Bug     Medium  1.7.0   Logging         LoggerWrapper doesn't 
> use
> right values for %C %M %L

I would mark this 'WONTFIX' if this status would be available, which isn't
the case. (Can it be added?)


> #6241         Open    Wish    Medium  1.7     Taglibs         TreeTag: No way to 
> order or
> constrain steps.

This heavily limits the usefulness of TreeTag. But TreeTag is new
functionality, so it needs not be a show-stopper for 1.7.0. It could perhaps
be added to 1.7.1, of in a separate taglib release.

Perhaps Tree-tag should be explicitely marked as 'experimental' in the
documentaion, to show that you'd probably not want to use it as yet, because
this very important feature is missing (the feature is present in the
bridge's util TreeList, so it can't be very hard to wrap it in the taglib,
but I've simply had no time for that until now).


 Michiel

-- 
Michiel Meeuwissen 
Mediapark C101 Hilversum  
+31 (0)35 6772979
nl_NL eo_XX en_US
mihxil'
 [] ()

Reply via email to