Am 02.12.2008 um 08:04 schrieb Anjo Krank:
If it doesn't, rewrite the method so it re-orders the nsarray so
that it still has all items from _expectedLanguages, but the ones
from nsarray come first and in order and post a patch.
I checked in an (untested) interim patch. See if the problem
Am 02.12.2008 um 07:46 schrieb phil:
er.extensions.ERXApplication.fixCachingEnabled=false
In your properties file and see if that doesn't get the server to
respect the browser defaults. I say a partial fix, because setting
the language in the session does not change the component
return
Hi Ramsey,
Setting the er.extensions.ERXApplication.fixCachingEnabled property
solved the problem. Thank you so much for the time spent tracking the bug.
Now I've a lot of mods to do in my apps but I think I'll spend some time
to set up UI tests and try to keep up to date with versions of Won
On Dec 1, 2008, at 9:41 PM, Mike Schrag wrote:
I dunno, think you are up to it?
I can't reproduce this, btw ... I vote that you're lying to me.
Remember null vs context().page()? :-P
Chuck
--
Chuck Hill Senior Consultant / VP Development
Practical WebObjects - for devel
Am 02.12.2008 um 00:50 schrieb Chuck Hill:
On Dec 1, 2008, at 3:38 PM, Andrew Lindesay wrote:
Hello Mike;
I guess that's only really possible through a slip-up in the top-
level exception handling... hmmm... thanks.
Or sleep() or terminate() not calling super or throwing an
exception.
I dunno, think you are up to it?
I can't reproduce this, btw ... I vote that you're lying to me.
ms
___
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing list (Webobjects-dev@lists.apple.com)
Help/Unsubscribe/Up
On 02/12/2008, at 12:33 AM, Ted Archibald wrote:
Begrudgingly I put together a cloning method in my entities, you guys
are no fun!
LOL
with regards,
--
Lachlan Deck
___
Do not post admin requests to the list. They will be ignored.
Webobjects-dev m
On Dec 1, 2008, at 6:01 PM, Mike Schrag wrote:
Why doesn't terminate check in the session?
It does check in the session, but because it has terminated, the
request that is blocked on the session checkout never gets unblocked.
If it checks in the session, it should .notify() the TimeoutEntry
Why doesn't terminate check in the session?
It does check in the session, but because it has terminated, the
request that is blocked on the session checkout never gets unblocked.
If it checks in the session, it should .notify() the TimeoutEntry (the
code looks relatively straightforward)
Won
On Dec 1, 2008, at 5:13 PM, Mike Schrag wrote:
This bug happens when multiple requests for a session come in
before the first request completes (Ajax on the page can cause
this) and the first request calls terminate() on the session (e.g.
from a log out link or in response to an exception
Hello Mike;
I know I get this from time to time and have to address it, but I
think I do as you do and end up with non-mutables getting bound to
"list".
cheers.
I thought the WORepetition looked like it copies the NSArray before
using it so that this can't happen? Maybe not with mutable
This bug happens when multiple requests for a session come in before
the first request completes (Ajax on the page can cause this) and
the first request calls terminate() on the session (e.g. from a log
out link or in response to an exception during request processing).
The session terminat
On Dec 1, 2008, at 5:01 PM, Mike Schrag wrote:
I guess that's only really possible through a slip-up in the top-
level exception handling... hmmm... thanks.
yeah
Or sleep()
yeah
or terminate() not calling super or throwing an exception
yeah
. Or two requests for a session coming very
I guess that's only really possible through a slip-up in the top-
level exception handling... hmmm... thanks.
yeah
Or sleep()
yeah
or terminate() not calling super or throwing an exception
yeah
. Or two requests for a session coming very close together and the
first calling terminate(
Hello Mike;
I thought the WORepetition looked like it copies the NSArray before
using it so that this can't happen? Maybe not with mutable's I
wonder?
I was checking to see if this is a bug we introduced in
ERXWORepetition (which is actually what's being used here), but as far
as I can t
On Dec 1, 2008, at 3:38 PM, Andrew Lindesay wrote:
Hello Mike;
I guess that's only really possible through a slip-up in the top-
level exception handling... hmmm... thanks.
Or sleep() or terminate() not calling super or throwing an exception.
Or two requests for a session coming very clo
Hello Mike;
I thought the WORepetition looked like it copies the NSArray before
using it so that this can't happen? Maybe not with mutable's I wonder?
cheers.
In the long running debate over return null vs return page, I
officially concede to Chuck :) I have long maintained that
return
Hello Mike;
I guess that's only really possible through a slip-up in the top-level
exception handling... hmmm... thanks.
cheers.
Does anybody know what the monitor object for this synchronised-
block might be for?
[1] java.lang.Object.wait (native method)
[2] java.lang.Object.wait (Object
Phil,
I just built a small localized app and I can verify this problem on WO
5.4.3 and latest Wonder. Setting the language on the session
explicitly has no effect. Between English and Japanese, it always
uses the English component. Even setting the default language in the
properties fi
That means if you delete item #3 from a 3 item long list and return
null from your delete method
Since I'm pointing out a technical idiosyncrasy, exactness probably
matters -- Change this to "if you delete item #1 from a 3 item long
list" ... If you deleted #3, it would actually still work be
In the long running debate over return null vs return page, I
officially concede to Chuck :) I have long maintained that returning
null really doesn't matter that much, but I just found a case where it
does. If you have a repetition over an array and you have a delete
method that deletes
Does anybody know what the monitor object for this synchronised-
block might be for?
[1] java.lang.Object.wait (native method)
[2] java.lang.Object.wait (Object.java:474)
[3] com.webobjects.appserver.WOSessionStore.checkOutSessionWithID
(WOSessionStore.java:191)
It means a session didn't ge
Hello;
Does anybody know what the monitor object for this synchronised-block
might be for?
[1] java.lang.Object.wait (native method)
[2] java.lang.Object.wait (Object.java:474)
[3] com.webobjects.appserver.WOSessionStore.checkOutSessionWithID
(WOSessionStore.java:191)
I have no othe
Do you have project Wonder installed? Do you have an old version of
Project Wonder? This class was added recently.
Chuck
On Dec 1, 2008, at 12:47 PM, Francesco Romano wrote:
ERXRedirect could be the class I was looking for...
Only a ... little... problem.. Why eclipse does not see the cla
ERXRedirect could be the class I was looking for...
Only a ... little... problem.. Why eclipse does not see the class? I
can't find it.
On 01/dic/08, at 21:35, Chuck Hill wrote:
On Nov 28, 2008, at 11:34 PM, Francesco Romano wrote:
Thanks for the reply...
I tried something like:
reload
On Nov 28, 2008, at 11:34 PM, Francesco Romano wrote:
Thanks for the reply...
I tried something like:
reloadPage :WOJavaScript {
scriptString = session.getJSRefresh;
hideInComment = YES;
}
with:
public String getJSRefresh () {
if (_mainPageNeedRefresh) {
I thought Wonder extends the classes of WO... But.. this is not a
problem.. I removed the "complex" PKs...
Wonder does extend WO, but the Migrations API is an entirely API in
Wonder. It sits on top of the EOF schema sync API's, which DO support
composite PK's/FK's, but migrations expose a mu
mmm
This is the Project Wonder Migrations API you're referring to here,
not EOF (or Entity Modeler). EOF fully supports composite foreign
keys, I just didn't want to write an API in migrations to support it
because I don't use them (and nobody has ever asked). It's relatively
straightfo
Hi Phil,
did you check that you get the localized components in the correct
spots in the built application? Not that one localization overrides
the other or so?!
cug
On 01.12.2008, at 07:33, phil wrote:
Hi,
To Ramsey:
Yes I'm sure I'm using session, cause I'm NSLogging from my
Sessio
It's seems that EOM can't model FK with more than one attributes..
Yes it can ... I don't know about "propagates primary key" with
composite PK's, though.
ms
___
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing li
It's seems that EOM can't model FK with more than one attributes..
So the previous question (if create two-attributes PKs) is answered...
I'll create very simple PK with one attribute of type id (sigh..)
On 30/nov/08, at 08:56, Francesco Romano wrote:
Ok.. maybe it's a bit better...
this a
Hi,
To Ramsey:
Yes I'm sure I'm using session, cause I'm NSLogging from my
Session.java, extension of ERXSession...
I tried to create a new Application, following the Hello World tutorial.
As long as my only localizations are Strings in Localizable.strings,
this works, browser preferencies
You have chosen
wisely.
On Dec 1, 2008, at 8:33 AM, Ted Archibald wrote:
Begrudgingly I put together a cloning method in my entities, you guys
are no fun!
___
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing
Begrudgingly I put together a cloning method in my entities, you guys
are no fun!
___
Do not post admin requests to the list. They will be ignored.
Webobjects-dev mailing list (Webobjects-dev@lists.apple.com)
Help/Unsubscribe/Update your Subscriptio
Maybe the Strategy design pattern could be used instead of inheritance
and the Strategy behaviour becomes a selectable "behaviour" attribute
of entity "Batch" instead of using subclass types. If you are unsure
what the Strategy design pattern is, see Head First Design Patterns
book pages
On 30.11.2008, at 22:19, Ted Archibald wrote:
This app will have a small user base (<6), and will be in a
controlled environment. That said I don't intend on switching
objects that often, but approximately one in every 1000 objects
created would have to be changed due to data entry errors.
36 matches
Mail list logo
