Deeplink from bot to tab in Microsoft Teams - microsoft-teams

I am currently working on a bot specifically targeting Teams. Search feature in this bot returns around 200-500 results. Even though i have integrated refines to narrow down search results, i still have to show around 50-100 records under a refined label. I am trying to avoid showing 100+ records as Carousel or list view in bot. so, I thought of integrating tab with bot and share a deeplink to tab where user can see complete search result in a data table (jquery).
I am unable to figure out two things on this approach and need help.
Tab content might be hosted in a different domain and needs authorization. How do I pass authorization info to Content url without asking user to login?
How do i pass custom parameters while creating a deep link to tab and read custom data in a tab? For example, userID, accessToken.
Note: I am using AzureADV1 token with Adal.Net for Bot authentication, and storing token cache in a persistent storage.
I appreciate any help on this.

You can include a "context" parameter in your static tab deeplink, similar to the configurable tab deeplink. If you include a "subEntityId" property in this context, you can get the value inside your static tab by calling getContext(). Then you can render a filtered results based on this sub-entity id.

Related

Microsoft teams user detection

I am showing a wordpress website in iframe inside Microsoft teams Tabs. is there any way to detect the loggedIn user in my website?
If you are able to wrap your website inside a Teams app (which you can do for example very easily using App Studio, then you could surface your website as a "Tab" app. Doing this means you can reference the Teams JS library in your code, which will enable you to get a context object. You can get a lot of user information from that.
Be aware though that this context is not really 'trusted' data (it's just inserted via Javascript), so it could be spoofed into your page by a malicious actor. If that's a problem, you should look to implement Tab SSO in your app, which will give you a proper JWT token that you can validate.

MS Teams custom tab app changes Session.SessionID between requests

I'm making a custom tab app of MS Teams with ASP.NET, however, the tab doesn't seem to pass a same cookie between requests on MS Teams. So the ASP.NET app behind the tab generates a new Session.SessionID on every request.
I've checked the following question, and tried some settings according to that page, but nothing helped me. Actually my web site works nicely if I navigate it via Chrome or Edge.
ASP.NET: Session.SessionID changes between requests
How do I get a same cookie between requests on MS Teams?
I've not tested this specifically so haven't seen it, but basically the broad idea of session is to have to uniquely "remember" a user, and then restore State for them from a location (e.g. database). From your question, it seems like the out of box "Session" object is giving trouble, but at any rate you should probably avoid using it because it won't "remember" the user even across devices.
However, Teams provides you a way to achieve the same thing yourself quite easily. Remember that the Teams 'Context' object provides a userObjectId property that is unique and valid for the same user on all sessions on all devices (it's actually their Azure Active Directory id). You can simply store whatever you want in your own database, key'ed by this id, and request it on page load. It's also possible to get this from the querystring for a static (personal) tab if you want to handle the behaviour server-side (e.g. C#).

Enable / disable functionalities for users on Ms Teams custom App

I'm planning to build a custom application for Ms Teams and I wanted the app to have an 'Administration' context, reserved for some users, with charts and records management that the rest of the users wouldn't be able to see. Is there a way to accomplish this?
Is it possible to hide tabs depending on a user profile? Is there another way to accomplish this on teams?
Tks
Inside your tab you can do whatever you want - it's just a custom web app/page. The only "restriction" aside from a normal web app, is that it must implement the Teams javascript library - see here for more. However, once you've implemented that library, you can get the Teams context, which includes the user's Azure AD Id and also UPN, and you could use one or both of those to do a lookup against your database, determine the user's role, and show/hide UI elements accordingly.
However, I think you might be asking about whether the entire Tab -itself- (i.e. even the tab item inside the Teams client) can be hidden from other users. To do this, an option is to use a "private channel", just for the relevant users, and pin the tab inside there only.

Use Application Insights to view all REST API requests per user

We are developing a WebApi2 application. Users can use the REST API using their specific API key. I am investigating how to setup Application Insights to be able to see the usage of the REST API endpoints per user.
I have been experimenting with using telemetryclient.TrackPageview(endpoint-ID) and telemetryclient.TrackEvent(endpoint-ID), but I am not sure if this is the right way to go. And how should I pass the user-ID (API-key) in the tracking call? As telemetryclient.Context.User.Id? Or using properties or some other way? I would like to be able to use the user-ID to aggregate/segment the pageviews/events that were registered for that particular user. Any ideas are welcome!
I'd suggest using the built in properties, like context.user.id whenever possible, and then custom properties after that.
Currently in the portal, from the search view you can then find an item for any user, then use the "related items" to view "all telemetry for this user" to see all of the items.
Then can use the App Analytics portal to do queries (click the "analytics" button toolbar in the overview blade for your resource azure portal to go to the AA site for that resource), and that service lets you write much richer queries.
in those queries, it's easiest to group/filter on the "built in" properties, for custom properties you have to write a little bit more complicated queries to parse the values out of custom code.

Create Docs through the Drive API with specific page setups (margins, orientation)

My app is already able to create Google Docs through the Drive API, and when doing so I can choose the title, data, and metadata. But I've been looking at the documentation and it seems there is no way to set some other properties of the Docs, such as the margins and the orientation (portrait or landscape) of the Docs being created. Is there a way to do so?
After not finding an answer in the documentation, the only thing I could think of was logging in with my Google Account, going to one of the created Docs, click on "Page setup", then select "Landscape", and finally "Set as default". I thought that by setting it as the default in my Google Account, it may happen that all the Docs created with it may be created as Landscapes. But it didn't work. A possible reason, however, is that the account actually creating the documents is a Service Account, not my regular Google Account. I guess both accounts, while being linked, are not the same thing, so the defaults of one don't apply to the other. Is there some way to set "Portrait" as the default orientation for the Docs created with my Service Account? If not, would the solution be to stop creating the Docs with my Service Account, and start creating them with my regular Google Account, so that the defaults I apply to them apply too to the newly created Docs?
Drive API doesn't support Docs related features for you to retrieve/set metadata about the documents.
Service accounts are individual Google accounts and no way related with your own user account. On the other hand, your default settings will only apply to you. You cant programatically set other user's defaults.
The best way I can think of, which is really a hack, is to make your own google docs API using headless chrome and pupeteer, which could be used only in part, and in connection with, the official docs API.
This is basically a web browser that runs on your server, which can be controlled to do anything a client would normally do on a web browser, like click buttons, submit forms, etc, only at the command of a server, which can be commanded through HTTP requests etc.
So after installing puppeteer, open the chromium browser, log into your google account, then open google drive with the server, and you should already be logged into the account. If not, you may have to log into the account using the server logic, which can get complicated with 2 step verification...
but anyway once you are logged into your google account on puppeteer, on drive.google.com, simply stay on that page and wait.
If the server receives some kind of HTTP request to make a new document, then have the server just literally click the new document button on the webpage, and it should open the new document in a new tab, which you should be able to access with puppeteer.
Then, when it is made, you can either just send the URL back to the request server, and / or make your page setup changes now, based on the data received in the HTTP request earlier.
For example, if, at one point (either after the document is created, or during its creation), there is some kind of message specifying the color of the background, just manually, on the server side, click File -> page setup, and select Custom, and, programmatically "type" the hex color code etc then click the OK button, then end the POST request.
If you have any questions, let me know
var doc = DocumentApp.getActiveDocument();
var body = doc.getBody();
//turn landscape if portrait
var oldHeight = body.getPageHeight();
if (oldHeight > body.getPageWidth()){
body.setPageHeight(body.getPageWidth());
body.setPageWidth(oldHeight);
}

Resources