Publish workspace from Author instance to public - magnolia

I got newly created Magnolia instance. I tried to create an app via bundled groovy script and publish news to public instance. I got this error
It happened because 'ebtnews' workspace is not synchronised from author to private. So the question is how to sync workspace from author to private?

What I do is every time I added a new workspace in the module definition xml for my author instance, I make sure I also added this workspace in the module definition xml for my public instance. Then need to restart both author and public instance for it to create the new workspace.

Alternatively, you can just run following via groovy console/script:
// create workspace
Components.getSingleton(RepositoryManager.class).createWorkspace(app_repository, app_workspace)
// check we registered all right
appSession = ctx.getJCRSession(app_workspace)
// register node type
nodeTypeManager = appSession.getWorkspace().getNodeTypeManager()
type = NodeTypeTemplateUtil.createSimpleNodeType(nodeTypeManager, app_node_type, Arrays.asList(NodeType.NT_HIERARCHY_NODE, NodeType.MIX_REFERENCEABLE, NodeTypes.Created.NAME, NodeTypes.Activatable.NAME, NodeTypes.LastModified.NAME, NodeTypes.Renderable.NAME))
nodeTypeManager.registerNodeType(type, true)
appSession.save()
// double check it registered all right
nodeTypeManager.getNodeType(app_node_type)
You will also want to register basic security rights for the workspace, set it under subscriber workspace mapping to enable activation and possibly include/exclude it from list of triggers for flushing cache upon update of content on public instance.
You can find code to do all that in createAppScript sample script in groovy module. Code I've pasted above is actually from the same script.
Advantage being that you can do all that at runtime w/o restart. Disadvantage, that you need to run same code on each instance.

Related

Are custom repository instances created per request?

I am trying to create an application with mikro-orm and apollo-server-express, I want to use the batch processing and the caching of the Facebook dataloader.
Normally, Facebook dataloader instances are creates per request. If mikro-orm also creates custom repository instances per request and if all calls to EntityManager.getRepository() in the same request gets the same instance, it may be the perfect place to create the dataloader instances.
Repositories are created as singletons, so only one instance exists per EntityManager instance. You should fork this EM to have one instance per request, either manually, or via RequestContext middleware:
https://b4nan.github.io/mikro-orm/identity-map/
This way, each request will have its own EntityManager, that will have its own cache of repository instances.
Keep in mind that if you use RequestContext, you should get the request specific EntityManager from it, and get the repository from there:
// beware that this will return null if the context is not yet started
const em = RequestContext.getEntityManager();
// gets request specific repository instance
const repo = em.getRepository(Book);

Google Login without google-services.json file

I have been looking for a method in order to add the google-services.json at runtime as we use same base code to configure the application for the different client, hence we want to add google-services.json at runtime.
I have tried several links but am unable to find such a way to do it.
Any help will be very much appreciated.
The google-services.json is actually never read at runtime. During the build process of your Android app, the relevant information from google-services.json is translated into your App's XML resources, and that is where it's read when the app starts. Since the destination where the build plugin puts the values is hard-wired, you can't configure it to have multiple sets of configuration data.
Instead of trying to read google-services.json at runtime, I'd recommend configuring the FirebaseApp instance in your code explicitly, like this:
// Manually configure Firebase Options
FirebaseOptions options = new FirebaseOptions.Builder()
.setApplicationId("1:27992087142:android:ce3b6448250083d1") // Required for Analytics.
.setApiKey("AIzaSyADUe90ULnQDuGShD9W23RDP0xmeDc6Mvw") // Required for Auth.
.setDatabaseUrl("https://myproject.firebaseio.com") // Required for RTDB.
.build();
Now you can determine yourself where to keep the relevant information from the google-services.json and which ones to use when the app starts.
and then:
// Initialize with secondary app.
FirebaseApp.initializeApp(this /* Context */, options, "secondary");
// Retrieve secondary app.
FirebaseApp secondary = FirebaseApp.getInstance("secondary");
// Get the database for the other app.
FirebaseDatabase secondaryDatabase = FirebaseDatabase.getInstance(secondary);
Also see:
The documentation section on using multiple build flavors (where I now copied the code snippets from)
My previous answer on a related topic (where I think the code snippets in the docs originated 😀)

How to access meteor collection through the database

I want to have my application's admin code hosted on a completely different app that shares the same database. However, that means that my collections are defined, at least in the code, in the global namespace of my main application and not my admin application. How can I access my collections, that are in the database, without having the global variables defined in a file shared between the meteor server/client? For reference, I am using this article as the idea to set up my admin tools this way. admin article
To simplify the problem, let's say you have:
two applications: A and B
one shared collection: Posts
one shared database via MONGO_URL
Quick and Dirty
There's nothing complex about this solution - just copy the collection definition from one app to the next:
A/lib/collections.js
Posts = new Mongo.Collection('posts');
B/lib/collections.js
Posts = new Mongo.Collection('posts');
This works well in cases where you just need the collection name.
More Work but Maintainable
Create a shared local package for your Posts collection. In each app: meteor add posts.
This is a little more complex because you'll need to create a pacakge, but it's better for cases where your collection has a model or other extra code that needs to be shared between the applications. Additionally, you'll get the benefits of creating a package, like testing dependency management, etc.
Each application will have its own code but will share the same mongo db. You'll need to define the same collections (or a subset or even a superset) for the admin app. You can rsync certain directories between the two apps if that makes that process either but there isn't anything in Meteor that will do this for you afaik.
Or you could share data between the two servers using DDP:
var conn = DDP.connect('http://admin-server');
Tracker.autorun(function() {
var status = conn.status();
if(status.connection) {
var messages = new Mongo.Collection('messages', {connection: conn});
conn.subscribe('messages', function() { console.log('I haz messages'); });
}
});
This creates a local collection named messages that pulls data from the "admin server" over DDP. This collection only exists in memory - nothing is created in mongo. You can do this on the server or client. Definitely not the best choice for large datasets. Limit the data transfer with publications.

How to access a "plone site object" without context?

I have a scheduled job (i'm using apscheduler.scheduler lib) that needs access to the plone site object, but I don't have the context in this case. I subscribed IProcessingStart event, but unfortunately getSite() function returns None.
Also, is there a programmatic way to obtain a specific Plone Site from Zope Server root?
Additional info:
I have a job like this:
from zope.site import hooks
sched = Scheduler()
#sched.cron_schedule(day_of_week="*", hour="9", minute="0")
def myjob():
site = hooks.getSite()
print site
print site.absolute_url()
catalogtool = getToolByName(site, "portal_catalog")
print catalogtool
The site variable is always None inside a APScheduler job. And we need informations about the site to run correctly the job.
We have avoided to execute using a public URL because an user could execute the job directly.
Build a context first with setSite(), and perhaps a request object:
from zope.app.component.hooks import setSite
from Testing.makerequest import makerequest
app = makerequest(app)
site = app[site_id]
setSite(site)
This does require that you open a ZODB connection yourself and traverse to the site object yourself.
However, it is not clear how you are accessing the Plone site from your scheduler. Instead of running a full new Zope process, consider calling a URL from your scheduling job. If you integrated APScheduler into your Zope process, you'd have to create a new ZODB connection in the job, traverse to the Plone site from the root, and use the above method to set up the site hooks (needed for a lot of local components anyway).

Problem with Unit testing of ASP.NET project (NullReferenceException when running the test)

I'm trying to create a bunch of MS visual studio unit tests for my n-tiered web app but for some reason I can't run those tests and I get the following error -
"Object reference not set to an
instance of an object"
What I'm trying to do is testing of my data access layer where I use LINQ data context class to execute a certain function and return a result,however during the debugging process I found out that all the tests fail as soon as they get to the LINQ data context class and it has something to do with the connection string but I cant figure out what is the problem.
The debugging of tests fails here(the second line):
public EICDataClassesDataContext() :
base(global::System.Configuration.ConfigurationManager.ConnectionStrings["EICDatabaseConnectionString"].ConnectionString, mappingSource)
{
OnCreated();
}
And my test is as follows:
TestMethod()]
public void OnGetCustomerIDTest()
{
FrontLineStaffDataAccess target = new FrontLineStaffDataAccess(); // TODO: Initialize to an appropriate value
string regNo = "jonh"; // TODO: Initialize to an appropriate value
int expected = 10; // TODO: Initialize to an appropriate value
int actual;
actual = target.OnGetCustomerID(regNo);
Assert.AreEqual(expected, actual);
}
The method which I call from DAL is:
public int OnGetCustomerID(string regNo)
{
using (LINQDataAccess.EICDataClassesDataContext dataContext = new LINQDataAccess.EICDataClassesDataContext())
{
IEnumerable<LINQDataAccess.GetCustomerIDResult> sProcCustomerIDResult = dataContext.GetCustomerID(regNo);
int customerID = sProcCustomerIDResult.First().CustomerID;
return customerID;
}
}
So basically everything fails after it reaches the 1st line of DA layer method and when it tries to instantiate the LINQ data access class...
I've spent around 10 hours trying to troubleshoot the problem but no result...I would really appreciate any help!
UPDATE:
Finally I've fixed this!!!!:) I dont know why but for some reasons in the app.config file the connection to my database was as follows:
AttachDbFilename=|DataDirectory|\EICDatabase.MDF
So what I did is I just changed the path and instead of |DataDirectory| I put the actual path where my MDF file sits,i.e
C:\Users\1\Documents\Visual Studio 2008\Projects\EICWebSystem\EICWebSystem\App_Data\EICDatabase.mdf
After I had done that it worked out!But still it's a bit not clear what was the problem...probably incorrect path to the database?My web.config of ASP.NET project contains the |DataDirectory|\EICDatabase.MDF path though..
Is LINQDataAccess.EICDataClassesDataContext looking to the web.config or some other outside source of data for its setup?
I can tell you for a fact that you must jump thru hoops to get web.config accessible to your test code.
Update
Ah, yes. I see that you're using ConfigurationManager on the line where your test fails... ConfigurationManager looks to web.config for configuration. This has been a sticking point for me when I write my tests.
You need to either change the code so that the class can be instantiated without web.config, or you need to make it so that your tests can access web.config.
Does your test project have it's own configuration file? This type of behavior usually means the app can't find the connection string. Test projects require their own file since they are not running in the context of the client app.
UPDATE The error you describe after adding an app.config is common when testing web applications built on SQLExpress and attaching an .mdf. SQLExpress cannot be run in more than one process at a time. So if you have previously run your web application it may still be active and will conflict with the attempt to attach the database.
You can use SQL Management Studio to attach the database permanently and then use a more traditional connection string like:
Server=myServer;Database=EICDatabase;Trusted_Connection=True;
For me it seems like your problem is the connection string, which is not set.
I assume your unit test is in a different project than the DAL.
You call the 'new' command on the datacontext constructor without a connection string. So it should usually use its default, when set. But since this setting normally is stored in the web.config of the other project there is no connection string set and you get the error.
If its right, what i assume, you have to get the settings from the DAL project into the unit-testing project. Simplest solution should be to copy web.config connection string to app.config of unit test project.
There are other possibilities for sure, but i think its not easy to get the web.config configuration into your unit-testing project.

Resources