I'm asking this because after the edge version has been updated to 0.5.3 it caused my application to have problems with the observers. Meaning that, I have a queue, I'm listening to it's changes on current attribute, (works perfectly on 0.4.x (released on 29th June), can't remember the exact build version but that's the date), and now after removing the item from the Queue it says 'current' attribute of undefined. As it tries to check another item in queue that doesn't exist.. I'll try to make a ssce tomorrow and update this.
Breaking changes in 0.5.0 are listed here on github
One of those was that Ractive no longer ignores errors in observers or in the evaluation of expressions. More than likely it's in the init of observer where it doesn't matter for your use case, but now is causing an issue. Something like:
ractive.observe('foo', function(newValue, oldValue, keypath){
if(oldValue.current) {
}
})
Related
I need to upgrade an event to a new version. This entails removing an old property that is growing (due to a design decision) and is not longer needed.
Having read the documentation available at https://docs.axoniq.io/reference-guide/axon-framework/events/event-versioning and implementing the required parts
#Bean
public EventUpcasterChain eventUpcasterChain(){
return new EventUpcasterChain(
new CurrentLoanBalanceAcceptedEventUpcaster(),
new InterestCalculatedEventUpcaster()
);
}
I cannot seem to get the upcasters to fire the "doUpcast" method. My understanding is that on start up the application will find all old events and convert them to the new version of the events. Is this correct? Or does it only upcast the old events when they are replayed? I am hoping there is a way to do this without replaying all the events.
Versions:
Spring Boot: 2.4.11
Axon Framework: 4.5.4
Update:
After trying to figure out how to remove nodes from the event xml using the upcasters I failed miserably. I must point out that this issue is not an Axon issue but was a poor design decision and a lack of understanding what Axon was doing with the events. The primary lesson is to keep events as simple as possible.
I managed to unload the unnecessary xml from the domain event store using the following query:
update domainevententry
set payload = lo_from_bytea(0, decode(regexp_replace(
subquery.output,
'\<nodeToReplace\>(.*)\<\/nodeToReplace\>',
''
), 'escape'))
from (
SELECT eventidentifier, payloadtype, encode(lo_get(payload::oid), 'escape') as output
FROM domainevententry
WHERE eventidentifier in (
'<id>'
)
AND payloadtype = '<payloadType>'
) as subquery
where domainevententry.eventidentifier = subquery.eventidentifier;
every upcaster in the upcaster chain is called only when it finds an 'old' event to convert to the new one. New events that are fired are the new version of it, so the upcaster will not be used for that.
Upcasters are used when:
An aggregate is loaded (without snapshot) and an old event is encountered
A projection is replayed and an old event is encountered
In all other cases, it will be the new event, so the upcaster won't have to be fired. The upcaster is there so aggregates can keep being loaded, and projections can be replayed from the beginning of the event store.
If this is not the case, we need to take a look at the Revision parameter on the old event definition and the new event definition. For example, if the old event had no #Revision annotation, the SimpleSerializedType needs a version of null or it won't match.
Please include the code of the upcasters if we need to dig further, that would help greatly!
I have a set of events that have been refactored to another package. This works as is until I execute the event replay. Digging deeper I noticed a payloadtype in the domainevententry table and figure changing this would be sufficient but alas it seems the xml root element of the event needs to be changed as well. I am hoping there is a simple way to do this.
I cannot find any examples on upcasting to different packages or using XStream aliasing so any help would be greatly appreciated.
Thanks
As you noticed, the default payload type stored in events is the fully qualified class name. This ensures that out of the box serialization and deserialization work as intended. However, moving classes around thus means the payload type can no longer be found, requiring some adjustment to be made.
You could have used the EventTypeUpcaster, as referred to in the Reference Guide. The EventTypeUpcaster is dedicated to adjusting the payload type, and thus can also be used to deal with changing package names.
When using (the default) XStreamSerializer, aliasing the tags would indeed also work. How to set aliases can be seen here for example. AS noticed in that sample, the alias is added to the XStream instance. The XStreamSerializer uses an XStream instance to support de-/serialization from/to XML. To adjust the XStream instance, you can simply use the builder paradigm on the XStreamSerializer. The JavaDoc of the builder should be specific enough to help you out how to use it.
Went the long way round with this but it seems to work. As always, backup your database before executing large volume changes. I also restarted the service using the database on completion. Needless to say I will make sure the events are in logical packages before deploying next time :)
Database Engine: Postgres 10
Table: domainevententry
update domainevententry
set
payloadtype = '<new.package.Classname>',
payload = lo_from_bytea(0, decode(REPLACE(
subquery.output,
'<old.package.Classname>',
'<new.package.Classname>'
), 'escape'))
from (
SELECT eventidentifier, payloadtype, encode(lo_get(payload::oid), 'escape') as output FROM domainevententry
WHERE eventidentifier in (
'<event guid 1>',
'<event guid 2>'
)
AND payloadtype = '<old.package.Classname>'
) as subquery
where domainevententry.eventidentifier = subquery.eventidentifier;
Once that is completed I needed to update the OWNER of the large object:
ALTER LARGE OBJECT <LargeObjectId> OWNER TO database_role;
Probably not the most elegant solution but based on the time constraints I have, it did the job. There are probably encoding issues with this solution for the large object but it all worked out for me in the end. Feel free to share any optimizations that would make the above more suitable.
Firing off the Axon Framework replays rebuilt the projections and everything lined up.
Upgrading AngleSharp from 0.9.6 to 0.9.9 I have this line of code no longer compiling:
return configuration.With(LoaderService(new[] { requester }));
It complains that LoaderService does not exist in the current context. So what happened to LoaderService? Is there a replacement? Does it still exist but just somewhere else?
Good question. Sorry for being late to the party, but even though you may have solved your problem someone else is having a hard time figuring it out.
LoaderService was essentially just a helper to create a loader. But having a service for anything creating a little thing would be overkill and not scale much. Also AngleSharp.Core would need to define all these. So, instead a generic mechanism was introduced, which allows registering such "creator services" via Func<IBrowsingContext, TService>.
However, to solve your piece of code I guess the following line would do the trick:
return configuration.WithDefaultLoader(requesters: requester);
This registers the default loader creator services (one for documents, one for resources inside documents) with the default options (options involve some middleware etc.).
Under the hood (besides some other things) the following is happening:
// just one example, config.Filter is created based on the passed in options
return configuration.With<IDocumentLoader>(ctx => new DocumentLoader(ctx, config.Filter));
I have an observer to a value "App.selectedValue". I also have another Ember object that has a binding (App.someObj.appValueBinding) to App.selectedValue. However, when my observer is called, the binding of App.someObj is not updated.
This is illustrated in http://jsfiddle.net/Ur2Qj/8/
In the jsfiddle, you can see in the Chrome debugger or FireBug, that App.selectedValue and App.someObj.appValue have different values, even tho' the latter is bound to the former.
Seems like the binding should be updated when the observer is called. Is this expected behavior in Emberjs or is it a bug? Is there a work-around?
Thanks for looking at this!
Take a look at this: http://jsfiddle.net/ud3323/GUHCD/ (in JavaScript; I don't like CoffeeScript... sorry).
The two main things you've got wrong here is not using get() and set() properly and in your observer you need to set App.someController.content after the end of the current runloop (which means after all the other bindings have taken place). You do this by using Ember.run.next(). You could also use Ember.run.sync() there as well.
Oh and you need to use jQuery 1.7.1. Version 1.5.2 is not compatible with Ember.
Regularly during my application run, I get
TypeError: Error #1009: Cannot access a property or method of a null object reference.
at mx.managers::SystemManager/stageEventHandler()[C:\autobuild\3.4.0\frameworks\projects\framework\src\mx\managers\SystemManager.as:5649]
This is the full stack trace. Obviously, I guess there is something wrong, but I can't understand what.
Is there any way for me to find the origin of that bad behaviour ?
EDIT
Having added my SDK sources to my debugger, I can now say precisely which line it is :
private function stageEventHandler(event:Event):void
{
if (event.target is Stage)
mouseCatcher.dispatchEvent(event); // This is line 5649
}
mouseCatcher is indeed null. The current event target is indeed a Stage object, and event type contains the "deactivate" String. As event occurs at application startup (before I try to do any kind of user interaction), I guess it's a kind of initialization bug, but where ? and why ?
Look at the source code, this is always your best option. The 3.4 SDK is open source (datavisualization and the flash player itself aside) and you probably already have the source for it in your FlashBuilder/FlexBuilder install/sdks folder. Use grep or windows grep to find the file in question (or find, whatever floats your boat). Open the SystemManager file and check what's happening at that line, check for calls to the method (if it's public use grep again, if it's private you just need to look within the SystemManager). Try to understand why it gets to this point, as pointed out by some others it's likely a timing related issue where you're trying to access something before it has been assigned, in this case the SystemManager, you probably need to defer whatever action you're taking that is causing the error to a later part of the life-cycle (if you're using initialize event or pre-initialize try on creationComplete instead since that will be dispatched after the createChildren method is called).
Note: Mine is located here
C:\CleanFS\SDKs\flex\3.4.0.9271\frameworks\projects\framework\src\mx\managers
In my copy of SystemManager with the version of the SDK I have that line number doesn't make any sense since it's a block closure not an executable line so you'll have to look at your specific version.
It looks like you are using the Flex 3.4 SDK. Are you listening for the ADDED_TO_STAGE event when the application loads? Or doing anything with the Stage object on load? If so, you might be hitting a bug specific to the 3.4 SDK:
http://bugs.adobe.com/jira/browse/SDK-23332
The most obvious solution is to swap out the 3.4 SDK for a later version (3.4A, 3.5 or 3.6). You can do that here: http://opensource.adobe.com/wiki/display/flexsdk/Download+Flex+3
All of your code should be backwards compatable with the newer Flex 3 SDKs.