Thanks all. Some good ideas.
On 26/01/2010, at 7:12 AM, Chuck Hill wrote:
> On Jan 25, 2010, at 5:35 AM, Pascal Robert wrote:
>
>> Le 10-01-25 à 01:40, Lachlan Deck a écrit :
>>
>>> Hi all,
>>>
>>> just polling for ideas / techniques people use in detecting dead instances.
>>> Naturally when
If this has just started happening and you've done nothing else (e.g., changed
configs etc) then maybe try creating a new workspace.
On 26/01/2010, at 12:07 AM, Frank Gibau wrote:
> Hey WO-friends,
> ich have a problem with Eclipse. Running a WO-Application in Debugmode you
> normally get a ver
Yeah, I try to write an overall "how do you use it" document for the
package.html with smaller chunks in the individual class javadoc.
ms
On Jan 25, 2010, at 6:45 PM, Joe Little wrote:
> Searching for things within Confluence is still problematic, but I do
> believe its still the best place for
Once again, thank you for your ideas. Going to increase my caffeine intake
now.
Amiel
On Tue, Jan 26, 2010 at 4:24 AM, Chuck Hill wrote:
>
> On Jan 25, 2010, at 12:32 AM, Amiel Montecillo wrote:
>
> Hi List,
>>
>> I am having a dilemma on how to best implement this kind of scenario.
>>
>> I nee
Searching for things within Confluence is still problematic, but I do
believe its still the best place for showing _how_ to use things in
general. I've found better explanations there on average than what I
can glean from the java docs, but mostly that's because I'm more
example-driven than one to
the great quandry of javadocs i wonder if this should be a symlink to the
package.html for the bulk of this page? Maybe it should only include notes and
comments above and beyond what appears in the javadoc? Ray is working on a tool
that might make this process suck less.
ms
On Jan 25, 20
I was originally using the confluence docs.. and when that failed,
went to the class file docs. Both were wrong. I'll go and update the
confluence docs.
http://wiki.objectstyle.org/confluence/display/WONDER/ERTaggable
On Mon, Jan 25, 2010 at 3:38 PM, Mike Schrag wrote:
>> not sure which docs yo
Thanks.. that did it. I thought it was strange using the constructor there :)
On Mon, Jan 25, 2010 at 3:37 PM, Mike Schrag wrote:
> yeah -- docs bug in ERTaggableEntity ... Sorry about that. They got out of
> date. I think it USED to be what was in ERTaggableEntity and then changed.
>
> ms
>
> O
> not sure which docs you were looking at, but the package.html* (which i
> consider to be The Docs (TM) ) looks correct:
*
http://webobjects.mdimension.com/hudson/job/Wonder53/javadoc/er/taggable/package-summary.html
ms
___
Do not post admin requests
yeah -- docs bug in ERTaggableEntity ... Sorry about that. They got out of
date. I think it USED to be what was in ERTaggableEntity and then changed.
ms
On Jan 25, 2010, at 6:35 PM, Mike Schrag wrote:
> not sure which docs you were looking at, but the package.html (which i
> consider to be The
not sure which docs you were looking at, but the package.html (which i consider
to be The Docs (TM) ) looks correct:
public ERTaggable taggable() {
return ERTaggable.taggable(this);
}
public static ERTaggableEntity taggableEntity() {
return ERTaggableEntity.taggableEntity(Person.
Another hitch. Docs say to create a convenience method:
ERTaggableEntity provides entity-level tag management and fetching
methods. Typically you would provide a cover method from your entity
class to an instance of an ERTaggableEntity: public class Person
extends _Person { ... public static ERTag
Or, like WOnder ERJars, maybe you could make a framework just for necessary
jars and include that framework on the build path of those projects?
Tim Worman
UCLA GSE&IS
On Jan 25, 2010, at 2:57 PM, Joe Moreno wrote:
> Chuck,
>
> Thanks for the quick reply. What you're saying makes sens
Chuck,
Thanks for the quick reply. What you're saying makes sense - so I'll
try removing the jar from /Library/Java/Extensions and add it to each
project.
Many thanks,
Joe
On Jan 24, 2010, at 20:15:50, Chuck Hill wrote:
On Jan 24, 2010, at 6:29 PM, Joe Moreno wrote:
Eclipse 3.3.2
WOL
I wanted a to-many relationship for attachments on multiple entities
(journals, payments,invoices, whatnot). In the end, I added that one
line model dependency and it work. My model defines an entity called
FileAttachment, which itself is modeled as a to-many for multiple
other entities. FileAttach
Hi All,
The main question Dave was asking was how to give back to the community
his useful fix for a very important piece of this problem.
I believe the answer is to just use NSProperties. That way the option can
be given in the Properties file, from the command line, or even from your
home di
sketchy.
In your model, it has a to-one relationship to ERAttachment, right? it should
have figured that out automagically and not even required ERAttachment declared
explicitly, though it's generally a good thing to do anyway, since an explicit
declaration defines the version #.
On Jan 25, 2
no .. i just assumed you had an ERAttachmentMigration subclass in your app that
was adding the foreign key. if you just generated a migration from Entity
Modeler directly, then you won't need that subclass (because the foreign key
will be a part of your existing migration class). the way things
Never mind my last comment. Yes, I put the below modelDependencies
value in my primary migration, and it works now.
On Mon, Jan 25, 2010 at 1:50 PM, Mike Schrag wrote:
> hmmm ... out of curiosity, in your ERAttachmentMigration subclass, if you
> add the method:
>
>
>
> @Override
> public NSAr
Once you stray outside of ERXAjax*, you definitely take on dependencies ... I
just committed a fix that removes two presumptions that you are inside of an
ERXApplication. I would make sure to run your test cases if you switch over.
ms
On Jan 25, 2010, at 4:51 PM, Sacha Michel Mallais wrote:
>
So, I need to define a migration explicitly for ERAttachment,
including setting an ERAttachment.MigrationClassPrefix=blah.foo.bar?
On Mon, Jan 25, 2010 at 1:50 PM, Mike Schrag wrote:
> hmmm ... out of curiosity, in your ERAttachmentMigration subclass, if you
> add the method:
>
>
>
> @Override
On Jan 25, 2010, at 1:08 PM, Mike Schrag wrote:
If you're extending ERXApp, I seem to recall that we fix this
ourselves "the hard way," though I ran into this in 5.3 as well --
that you'd randomly get a port stuck on when you didn't want it (80
or 443).
In ERXRequest:
Thanks, Mike, I t
hmmm ... out of curiosity, in your ERAttachmentMigration subclass, if you add
the method:
@Override
public NSArray modelDependencies() {
return new NSArray(new ERXModelVersion("ERAttachment",
1));
}
does that fix it? we should probably put that in there.
ms
On Jan 25, 2010,
I'll do that. Also ran into trouble with ERAttachment. Doesn't it
auto-migrate in the table creation?
Failed to execute 'ALTER TABLE FILE_ATTACHMENT ADD CONSTRAINT
FILE_ATTACHMENT_ATTACHMENT_ID_id_FK FOREIGN KEY (ATTACHMENT_ID)
REFERENCES ERAttachment (id) DEFERRABLE INITIALLY DEFERRED'
ERROR: rel
i honestly don't know why we don't just tell you this outright rather than make
you visually diff the two ... seems like something these new-fangled
computational machines would be good at. feel free to log a jira on that.
ms
On Jan 25, 2010, at 4:16 PM, Joe Little wrote:
> doh.. didn't see th
doh.. didn't see that when checking the two.
On Mon, Jan 25, 2010 at 12:54 PM, Mike Schrag wrote:
> you can see from the two connection dicts what the diff is ...
> AffiliateEOModel explicitly declares a plugin name while ERTaggable does not.
> Just set a global override for plugin and it shoul
Thanks Chuck!
I figured it out minutes after I sent my plea for help ... and I was
using that page for guidance.
My only problem with that page was that there were some "connections"
implied but not spelled out. I will try to add a comment that would
make it clearer to a caveman like me!
On Jan 25, 2010, at 12:31 PM, Pascal Robert wrote:
Le 10-01-25 à 15:12, Chuck Hill a écrit :
On Jan 25, 2010, at 5:35 AM, Pascal Robert wrote:
Le 10-01-25 à 01:40, Lachlan Deck a écrit :
Hi all,
just polling for ideas / techniques people use in detecting dead
instances. Naturally wh
I think I called for help too fast...
I found that my segment
in the build.xml file did not have a jvmOptions key-value pair from a
default Project Wonder Application...
It looks like I have a lot of build.xml and build.properties files to
update...
Can someone please confirm if I am c
If you're extending ERXApp, I seem to recall that we fix this ourselves "the
hard way," though I ran into this in 5.3 as well -- that you'd randomly get a
port stuck on when you didn't want it (80 or 443).
In ERXRequest:
@Override
public void _completeURLPrefix(StringBuffer str
Hi Tim,
On Jan 25, 2010, at 12:51 PM, Tim Kliewer wrote:
Hi all,
I'm still somewhat new to eclipse, so I need some help...
How do I set the JVMOptions property that is used at launch time in
a deployed application?
I expect it is in the classpath file, MacOSClassPath.txt in
particular
On Jan 25, 2010, at 12:33 PM, Sacha Michel Mallais wrote:
Hi list,
Any idea why WO is appending ":443" to some URLs? They seem to
appear on URLs that WO generates (URLs generated by my code seem to
be fine), such as on a WO form action method. The page comes in as https://domain/wo/reques
you can see from the two connection dicts what the diff is ... AffiliateEOModel
explicitly declares a plugin name while ERTaggable does not. Just set a global
override for plugin and it should fix your problem.
ms
On Jan 25, 2010, at 3:44 PM, Joe Little wrote:
> I'm getting this error,
>
> A
Hi all,
I'm still somewhat new to eclipse, so I need some help...
How do I set the JVMOptions property that is used at launch time in a
deployed application?
I expect it is in the classpath file, MacOSClassPath.txt in
particular. Or is this old school and it is somewhere else now?
So, how
I'm getting this error,
A fatal exception occurred: The connection dictionaries for ERTaggable
and AffiliateEOModel have the same URL and username, but the
connection dictionaries are not equal. Check your connection
dictionaries carefully! This problem is often caused by jdbc2Info not
matching be
Hi list,
Any idea why WO is appending ":443" to some URLs? They seem to appear
on URLs that WO generates (URLs generated by my code seem to be fine),
such as on a WO form action method. The page comes in as https://domain/wo/request/url
but the generated form action method looks like http
Le 10-01-25 à 15:12, Chuck Hill a écrit :
On Jan 25, 2010, at 5:35 AM, Pascal Robert wrote:
Le 10-01-25 à 01:40, Lachlan Deck a écrit :
Hi all,
just polling for ideas / techniques people use in detecting dead
instances. Naturally when there's lots of instances running in
deployment i
On Jan 25, 2010, at 12:32 AM, Amiel Montecillo wrote:
Hi List,
I am having a dilemma on how to best implement this kind of scenario.
I need the update process of an EO by creating an new EO with the
updated values and archiving the old one. This EO has quiet a few to-
many relationships th
On Jan 25, 2010, at 8:14 AM, David Avendasora wrote:
Hi all,
Aaron Rosenzweig and I discussed the issue of how to properly
implement control over how Oracle and EOF sort nulls at this month's
WONoVA (http://groups.google.com/group/webobjects-nova).
The only time, as far as I know, where
On Jan 25, 2010, at 5:35 AM, Pascal Robert wrote:
Le 10-01-25 à 01:40, Lachlan Deck a écrit :
Hi all,
just polling for ideas / techniques people use in detecting dead
instances. Naturally when there's lots of instances running in
deployment it can be hard to know if a particular instanc
On Jan 24, 2010, at 10:40 PM, Lachlan Deck wrote:
Hi all,
just polling for ideas / techniques people use in detecting dead
instances. Naturally when there's lots of instances running in
deployment it can be hard to know if a particular instance has
locked up for some reason (and this can
Hi all,
Aaron Rosenzweig and I discussed the issue of how to properly implement control
over how Oracle and EOF sort nulls at this month's WONoVA
(http://groups.google.com/group/webobjects-nova).
The only time, as far as I know, where it becomes necessary to worry about this
is when you are us
We tried different things but in the end we used a trigger. If the archive is
for auditing I think a database trigger is best. No need to clutter up you
application with code to make an audit trail.
Message: 9
Date: Mon, 25 Jan 2010 16:32:23 +0800
From: Amiel Montecillo mailto:bosyot...@gmail
Le 10-01-25 à 01:40, Lachlan Deck a écrit :
Hi all,
just polling for ideas / techniques people use in detecting dead
instances. Naturally when there's lots of instances running in
deployment it can be hard to know if a particular instance has
locked up for some reason (and this can somet
Hey WO-friends,
ich have a problem with Eclipse. Running a WO-Application in Debugmode you
normally get a very helpful error message.
In most cases Eclipse show you directly the codeline with the error. Since a
few weeks i get error-message that a not so helpful:
Its very difficult to find the N
On 18 jan 2010, at 23:09, Johan Henselmans wrote:
> I have a site that has part of its content on one website with its own
> unique DNS-addres, and the Webobjects app resides in an iFrame on another
> website with another DNS-address.
>
> The first call from the website fills the frame with
Hi Lachlan,
Checking direct actions is a great idea. We also do a periodic
"netstat -n | grep CLOSE_WAIT", kill the instances which have too many
connections in CLOSE_WAIT state and send alert emails with additional
info (mysql process list, app's intance log tail, instance's dump,
open files via
This is exactly what we do too. We then use pingdom to keep hitting
the da on specific instance numbers so if it does detect a problem it
tells the exact app and instance that is dead.
On Monday, January 25, 2010, Chuck Hill wrote:
> What I have done is to have a direct action that uses EOF to pe
Hi List,
I am having a dilemma on how to best implement this kind of scenario.
I need the update process of an EO by creating an new EO with the updated
values and archiving the old one. This EO has quiet a few to-many
relationships that needs to be archived as well.
I am thinking of copying the
We normally implement a direct action - let's call it "checkSystemStatus" -
that performs a fetch to ensure the database operates the way we want it. If
this fetch succeeds, we just return the string "OK". Otherwise we return the
exception message and the stack trace.
On a second system, we run
It seems like these errors are finally gone! Yippee!!
The last thing I have done was to retire nested editing context and
stay in main one.
In fact, during this operation I've mentioned one improper transfer of
object between these contexts. But I decided to fulfill my plan to be
sure.
Thanks every
51 matches
Mail list logo
