Microsoft 365 Community Power Platform Connectors, 8 Mar 2023

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Overview and Applications of the Dataverse Connector

Description

In this half hour educational session, Matt Townsend (Senior Product Manager, Power Automate) teaches about Dataverse as a scalable data solution, and demos how to leverage the native Power Platform connector In Power Apps and Power Automate demonstrating practical business use case scenarios. This video was taken from the March 8, 2023 Power Platform Integrations / Connectors LABs Call.

Matt Townsend (Microsoft) | linkedin.com/in/matt-townsend-0033017a/

Supporting materials
https://learn.microsoft.com/en-us/connectors/commondataserviceforapps/

A

Matt Townsend with the overview and applications of the dataverse connector. Thank you. So much for being here, Matt I will stop sharing.

B

Awesome, hey folks, conversation.

A

B

Let me share my science here.

A

And Matt feel free to take up as much of the time as you would like for this latter half.

B

Sure, thanks your thing, yeah, it looks like we have about 30 minutes left um yeah nice to meet everyone uh thanks. um Thanks for all your interest in the platform, and these awesome questions here and Jocelyn I have to say that I almost fell off my chair with that demo. That was so awesome to see the the dolly pieces and everything uh tied into an awesome uh canvas app, so um love that demo. Thank.

A

B

um Try out some of those custom connectors and pieces um more myself, um yeah, so jumping in um I'm Matt Thompson uh from Atlanta Georgia um sharing some content about the database connector today. um This has been as a quick overview and I'll feel free to chime in the chat as I go along. um If this uh yeah, with any questions on this content, um this has been as a quick primer for anyone less familiar with the dataverse connector. uh Definitely, for you know other other, more advanced questions or other.

B

You know: Advanced Alm type scenarios um uh we do want to put out more tutorials and things around that. So, if you have any questions for those types of things, uh please uh please let us know what you'd be most interested in scene and with that.

B

Let me jump right into our content, um so just just to share um uh dataverse fitting in uh with our low code uh platform, um so you're used to seeing our other uh Power automate power, apps power, bi power, virtual agents and now power pages and other pieces right as a set of low code tools that span um all of Microsoft right, Azure, Dynamics, 365, M365 and other pieces right. These tools make it possible to build uh things that connect these simple, apps and complex business.

B

Apps that you know meet your workflows, um combine data from across all these different sources um yeah and help make things more efficient, underpinning Dynamics, 365 and all of our low code. Apps is Microsoft dataverse as a way to support, not just these apps, but your extensions on these apps. So you can store data, build custom apps with different um You, Know, Rich, business logic and Integrations right within this app.

B

It is more than just a data store, it's helping you, um you know build views, build Rich interfaces around these data, this data as well, um and uh today we're going to talk about how you can interact with this data source and these uh other operations uh from cloudflows and from uh Power automate.

B

um So there's two major scenarios that we'll show um just to kind of get a picture of how these pieces go together and I'll also cover, if you don't have dataverse how to get started, how to try out some of these um these pieces.

B

um The major two areas here are, um you know you can include Flows In, Dynamics, 365, apps and model driven powerapps. We have a flow menu where it's easy to publish Cloud flows that it can be triggered from any one of these contexts, making it easy to embed cloudflows within any workflows.

B

You've created from these apps, whether it's a Dynamics customer service sales, other pieces you're using or if you've just built a standalone model, driven power, app or other pieces, and you want to easily trigger flows um uh directly from that app, uh adding other shortcuts or things um uh just using cloudflows right from that context, the other set of scenarios are using the connector to automate your back office processes and and set up Integrations that you know, do work on your behalf behind the scenes using cloudflows.

B

So all of this is built around our natively integrated uh Power, automate, dataverse connector, which is built a little bit different than other connectors.

B

It's directly connected to dataverse, so it runs faster and makes it easier to um you'll see some of the um the ways that you can um not just use your connection when you're running the flow but users running the flow can share their connection and when the flow is triggered, you can have it operate in the context of folks other users changing those uh those rows and tables and dataverse have this connector use their identity for other steps in the flow and kind of yeah maintain those pieces.

B

um It's easy to respond to any changes in your data sources, so dataverse. Of course you can use the built-in tables and dataverse, you can add, custom any custom entities if you're tracking you know custom products, types of inventory, custom processes or any of those pieces. You can add custom entities and dataverse and manage those events.

B

You may have seen the last stops call where Nathan helgren and our database team was sharing more about virtual virtual uh tables and dataverse, and some of the new ways that those can be set up so that a dataverse can actually, um you know, reflect any data source in SQL um directly as a data version entity. So if you're building cloudflows building apps that all interact with dataverse, you don't need to have connections to those other sources directly handle those all those mappings and things in multiple places.

B

You can do it all from dataverse where you've brought in you know multiple other tables, multiple other things to the same place, and you can, uh you know automate all from that. One set of tables, um and the last last piece we'll see in the the number here is around querying: creating updating data in flows.

B

um It's easy to interact with dataverse and call other other actions there. From that context, um yeah, let me jump to some demo content any questions before I jump over to showing some of the interface and getting started pieces.

B

Okay switch my screen sharing here.

B

Okay, can everyone see this window.

A

The Microsoft learn documentation. Yes,.

B

Awesome awesome, so uh just jumping right in I did want to share this document up front for anyone who is uh diving into you know whether you're at the beginning of your journey or diving into more advanced pieces. um uh This documentation in this Power Platform adoption best practices area, has a lot of great guidance around.

B

um You know, spinning things up and scaling things up, um so I wanted to call out this piece around environment strategy just because um yeah understanding which dataverse environments you're using um and you know which context you're using the database connector from, is really important um as you get started so uh just to share um uh just a quick summary um for anyone less uh less familiar in dataverse.

B

um There's a default environment that's available um by default within your um your organization's tenant. You can also create custom environments, and you may also have access to other environments that others have created or that are used as part of your apps. If your organization uses different Dynamics, apps has different, uh you know they may have one or more other environments tied to those other apps that they've already built.

B

uh So when we talk about the dataverse connector today um today, this dataverse connector only operates within one of those environments, we're working on adding the ability for it to connect to other environments, making it easier to span those other pieces. So you can see here I'm working in my default environment and if I were to go to the uh Power Platform admin Center, that's admin.powerplatform.microsoft.com,.

B

I can confirm which environments that I have access to manage, which ones that I can access, and you know by clicking into this one- you can see um this underlying environment URL for this organization which, as you'll see, can be handy to refer back to sometimes as you're building different Integrations with this specific environment. So for anyone getting started with dataverse, you can use this Power Platform admin Center to check uh you know which environments you have access to manage your user access. You need by default.

B

You need an administrator system customizer um if you're doing more admin type tasks you need within this. You can make sure that your user or other users you're working with have uh the environment maker role to make sure that they can create and run cloudflows in this environment. Just to show that really quick. This is a demo user right now. This user has the basic user security role as well as environment maker, so I know this user would be able to work with cloudflows use the database connector all these other pieces.

B

If you do not yet have a data risk environment, um you can start a new trial for different pieces. You can start a power automate trial and get access to these pieces. If you have Microsoft teams in your organization, you have access to power, automate within context of teams and if you're, using a dataverse for teams, environment it'll show up as type dataverse for teams. Here.

B

The dataverse connector is always available from teams for that usage. For other environments, you do need a premium license to use the database connector, which means one of the per user power, automate, powerapps or other licenses. But if you don't have one of those you can start a 90-day trial just from power automate.com or through one of the other Dynamics trial. Experiences uh that I mentioned um yeah, so, let's say um you've um you've successfully um signed in here create an environment.

B

You have a context, um you're you're, ready to get started um here, I'm showing the powerapps maker experience, um um but I also want to make sure folks have seen the automate URL itself. um This is changed recently to be make.power automate.com. So this is your hub for everything, power automate. You can also reach it through the powerapps pages, but sometimes this embedding um doesn't load as fast. So sometimes it can be helpful to go directly here to to power automain um so yeah. Let's look at a sample app here.

B

I've created a very basic model, driven power up and just to show the steps to get here if you're following along and the recording would be from I have too many um app tabs open here, um creating a new blank uh blank app based on dataverse- and you see here it says, model driven app. This is creating a new model driven power, app that uses these same view, components and things that are common across other Dynamics, 365, apps and other um a date of respect, apps, um so I created.

B

One of these I gave it a name and I selected a table for this demo, app I selected the contact table type for this demo. So here you can see um you know, I can manage these views, the sidebars everything you can customize within uh power-ups, but by default.

B

um So sorry, from from this context, um I want to show two ways of interacting with your power automate flows from um you know from this app and kind of extending this app to have more automation.

B

um So we have a built-in flow menu from this experience. So for any selected records, uh this flow menu, um oops, sorry I've lost my assignment context. I'll show that over here.

B

All right from this little minute, you can see any flows that you have that are related to this contact table entity, so here um I've already created one flow. That has this context, but if you, if you have a model driven app and you want to easily um publish a flow um that anyone with access to this app um can trigger, um let me uh show the steps. um So if you um from this experience, if you click, create a flow, it brings you to a template.

B

um Using today. This trigger is only available in the Microsoft dataverse Legacy connector, we're working on bringing this to the current database connector later this year, but you'll click through this template.

B

um You'll select the environment they're using right. Now it's just default as the the current environment being used here. You would select the contacts table and add other steps that you want to run whenever anyone you know receives this context from the flow. So from these other steps, you would have all of the details of this contact entity available for use in other steps, and you can add other dataverse actions, other connectors to run from that point.

B

So, just to show an example here um this export contacts flow whenever one of these rows is selected, it's getting the forecast and just doing a basic list operation in dataverse. This could be extended to you know: email, a spreadsheet, take other actions from dataverse, and so on. um So let's say: you've shared this app. You want to make sure that other folks um can run this flow as well. This isn't just something you've built for you. You want it to be available uh to your team.

B

So if you find the cloudplay you've created, you see have a couple different sharing options um here.

B

um If you share as co-owner that's for folks who want to collaborate on the flow with, but if you're simply trying to share this flow in a way that other folks can run and use the same process, you want to go to run only users. Add those people or you can add a a group.

B

um I believe this can be either an AED group or a group team in dataverse to ensure that those users users have access anyone you add here will have the same experience if, if that trigger is being used, I'll have the same experience here where, just by using that trigger and having access to, you know View and run that flow they'll be able to trigger the flow from this context, um just as you um as the owner of the flow um can um and just to show here.

B

If you have dataverse steps in the flow that you want to run with the connection of that other user right, rather than your connection as the user that set up the flow. This is also the area where you can configure that.

B

um So you can say um you know for any dataverse actions in the flow uh any other connectors. uh When that user clicks Run for the float, you want um it'll, prompt them for their connections um and there here I'm, seeing my connections being used, but based on those other options. I was showing from the flow management page they'll, see, um they'll, be able to provide their own connections and their connections will be used for any of the steps in the flow.

B

um The flip side of this was for any other automatic process where you want to take more automatic steps uh in the background. So maybe you have a you know: guest check-in process or you're, managing volunteers or things, and when, when one of these is added, you want to take other steps in the background the dataverse trigger um when a row is added, modified or deleted. uh Lets you easily respond to any changes to any table in dataverse.

B

um Let me pull up the definition for that one, just to show an example, um so here this this is um uh again a simple example just to show these pieces, but here I'm showing anytime, you add a contact, and this is saying anyone in the organization I could also say any if I select user. This is any contact that I myself have added uh the parent-child business unit or business unit are scoping that to specific parts of my organization that I may have set up within dataverse. An organization is anybody at all.

B

Any rows that trigger that are added um can trigger this flow.

B

um I can say, um I can add additional filters here if there's contacts that only meet certain names or you know, if you want to um you, know if you're processing, uh new feedback items- and you want to add a certain query here um to say just ones with a certain name or other pieces, um you can limit all of that from the trigger um here and I'll share a cheat sheet with some great examples of all the different syntax to use here and making that easy to uh to filter uh in the links um right after this.

B

um So you can see whenever a contact is added here, I'm saying any contact uh that anyone else has added in this organization, um but then I've selected this option here for run ads so run, as is saying um for the dataverse actions in this flow. Just like we're. Looking at this from the manually triggered I can say myself as the flow and Connection owner um the connection being used for this trigger I want um the other database actions in this flow. To use that context. I can say the user who modified that context.

B

If it's, you know the change type modified um or modifying users, the user who added or modified on those pieces, I could also say the row owner. If uh you know, if there's a specific user and dataverse that owns that account owns that other uh data and if other operations should be used in their context.

B

um So here, if I select, modifying user and I select, this use invokers connection option. This is only for the dataverse connector to let you choose which other data for steps you want to be tied to this run as option. Then the other steps in this flow will use this user's identity. So if you had other steps, sending email through dataverse, adding other records and databers we'd all be using their their identity.

B

So here just to show when a new contact is added, um maybe they're um uh doing field work or some other pieces, and it's important for this contact entity to always have a snapshot of the current weather as they're working. So we can say, get the current forecast and then we're going to add that forecast as a note to that record in dataverse, um so just to show um what that looks like you can see for uh this John Doe record that I created a moment ago.

B

um You can see it just attaches this note that says current weather sunny skies high of uh 65 in Atlanta um going through.

B

So this is a simple example with the weather, but you can easily see how um you know if you're receiving an email receiving other context um you can take in the context of that item coming in perform other actions in dataverse, uh based on that, um uh there I've seen some great examples around using the new chat, TPT and open API endpoints to summarize uh different emails things and coming in so you know when you're, adding these other pieces or sending other pieces.

B

You can summarize that content um you know, put that in a notification or other um other records that you're linking from here um yeah, so that um so that was showing mainly tricking triggering flow. How to share that, with your organization, we just saw um triggering a background flow using the winter OAS added, modified or deleted trigger, and tying that to other dataverse steps and making sure you're still using the right connections that you intend to you for that.

B

um There's a lot of other parts of the database connector that we can get into around different actions, but just in the interest of time, I will leave um it's excellent, cheat sheet from Donnie uh kahil as a reference, so this walks through every parameter of the database connector and has great um great examples. So you know, for example, all the different filter options as you're working with you know all the different data types and dataverse um and your filtering growth coming in or listing rows. um This syntax is needed currently um uh to send.

B

You know pass these filters on to dataverse, and this cheat sheet has a bunch of great examples that make that um you know really easy to refer to anytime. You are working with files, rows relating other records. Those other pieces um and I'll include the links as well. To this will level up uh browser extension, so this makes it really easy. Anytime, you are looking at a contact record or any other just viewing the form for this row and dataverse.

B

It shows you the internal names, so a job title has a display name and then a dataverse. If you were.

B

um If you're managing um the schema for this also show um within the powerapps experience the tables View for contact, you can see how I can manage um the schema for this entity, and this is the same place you go if you are creating or modifying other other custom entities. I can see a preview of the data here. I can modify the forms. That's where I just was to see um that one. uh The view of one contact, The View, was seeing um the list page and those those layouts for multiple contacts.

B

All that can be managed here and you can see, especially when you're working with flows.

B

um These shortcuts here for copying the schema name, copying The Logical name have been added recently to make it really easy to preview the raw metadata for this table. So you can confirm exactly which names as you're building different Integrations need to be used both for the table definition, the column, names and other pieces, and for pulling sample data for this table. So anytime, you're working with new custom Integrations in dataverse.

B

Sometimes it can be helpful to check um you know, make sure you're working with the right internal properties in those pieces from this View from the powerapps tables table browser um yeah. So I mentioned the level up extension, uh the dataverse cheat sheet and xrm toolbox, another tool that makes it easy um more than just seeing the Syntax for things. It has more Tools around browsing, browsing your tables, creating customizing attributes and things um in different ways that are um yeah some anti shortcuts that aren't always available directly from the interface today.

B

um So sorry, I know, I know that was a very general overview for some of these uh pieces of dataverse, but I hope these resources are are new and useful. To um uh to some of you, um Justin I have two more slides I can cover. We can jump to any questions from the group.

C

Matt there was a question from Karsten that seems to be related to the thing that you were just showing the copy X capabilities. Karsten. Do you want to jump off mutant.

D

Yeah, it's certainly something which, which makes it uh more reliable to Rene, to remember the schema name, for instance, and do stuff like this, but it certainly would be even more easier not to go into this experience and leave the other experience. So it would be great to see an improvement in the data verse itself and in terms of Select columns to not really selecting columns in terms of a comma separated, make it really a drop down. Mark and and drill into the table schema.

B

D

B

Roadmap perspective, so um we are looking at um making it possible to directly select columns so have a more advanced um Expression Builder, where you don't need to use the odata syntax. You can use the same. You know when you're customizing a view in dataverse using a very similar interface to you know, just add the attributes you want to filter on and save that into the flow.

B

um The major improvements we're working on right now are making it possible to connect to other environments from this connector, as well as bringing in this winter row is selected trigger to the latest connector, for you know, consistency with all these other pieces, um as well as some of the tools to make it easy to migrate from the older versions of the dataverse connector. um For folks that have you know older flows or um you know other Integrations um running with those other connectors.

B

um But yes, these usability pieces around selecting, columns, filter rows. These are all pieces that we want to improve, um and uh today the um you know open, API integration uh from this. This part of our admit is a separate uh set of actions, um but there's other places where we um you know we want to improve the co-pilot experience uh directly within dataverse and other places um as well to save um save effort and save the need to go uh copy and parse.

B

The schema um for um here as well so um I guess um yeah I would encourage everyone to take. Take this as a base for working with dataverse.

B

um We're definitely looking to improve a lot of these areas further, but yeah between uh these tools um and these capabilities that are here- and uh you know these new, um the open, API, independent publisher, connector and the other Integrations uh coming up um looking forward to see what you all um create and hear your feedback on other areas that this can be uh can be streamlined.

B

um Sorry I can't see the chat right now. So if, um if there's any other questions on that or other uh questions, please feel free to uh touch on manner.

A

A lot of people asking for the link to the cheat sheet, but then they found it online. Oh.

B

Yeah sorry I should have um pasted those in Direction.

A

No worries I added the links to the documentation that you were referring to, as you were referring to it as well. Let me just keep seeing and.

C

Then Karsten had another question: Kirsten, do you want to come off and ask that uh so when running flows in the context of the of dataverse, do you want to ask that.

D

Yeah, it's uh surely it's it's something around the uh running as a service principle um and when, when you're dealing with these kind of things running into multiplexing issues, so um I know that we have a kind of a um undocumented area here, and it would be great to find more public documents which are talking about the fact when it is best to use which kind of connectivities should I use it in context of the user.

D

Should I use it in context of the service principle and if so, when do I need to make sure not to run into any kind of multiplexing Licensing issues.

B

Definitely definitely um so. Yes, um the current current guidance right multiplexing is very much around ensuring that you know each of the users uh managing or benefiting in you know, involved in the cloudflow.

B

um Do have a you know: individual per user license if they're using you know these Premium capabilities for advanced, connectors or sorry uh premium connectors, including the database connector um in their flows, um but definitely um we're working on a guide, for um you know, scaling scaling, dataverse usage, um you know, anytime you're, working with higher volume, amount of data uh and that'll include some of the steps around service principle um interactions as well. As um you know, patterns for making uh batch or bulk calls or working with some of those other pieces.

B

So um yeah we'll have that um added the public documentation soon. uh Definitely I want to make sure that those steps are um are clear for folks um working with larger data sets and that kind of thing as well.

E

Yeah I I've got another question about a weirdly named thing um when you're doing run only and you're setting a flow up for run. Only you end up having to create a mail-enable security group in exchange. Then you have to go into the data verse and then you got to go to the thing called teams and then you've got to make a pretend team up that references. The other thing that you just enabled in the mail enabled Security Group.

E

It does work and it means, then you do all your admin at the exchange end of the system.

E

Is there any way of making it slightly more seamless so that you don't actually have to create a team, and you can actually just recognize that it's a mail-enabled security group, I I know this is one of those weird licensing things that you, you you're putting it a team inside the dataverse which gets confusing when you talk about teams and but the question we've got on this is that it does work, but it just seems a little bit counter-intuitive that you have to do it in two places to get something else to work.

B

Definitely um I know: there's there's a number of Parts where we want to make sure that things sync back to ad when possible, um but yeah right now the steps to create a group team in dataverse, where that's not automatically done, um does require managing in both uh in both places, um see I, don't I, don't know of any short-term Improvement that the changes that, but that's definitely a pain point that we recognize as well with you know, keeping AED and dataverse um in sync and making sure that you know when you share and when you're managing these flows right, you don't need to go, create a bunch of other identities.

B

Just to you know, do basic management pieces right and um yeah the last overhead there, um the better.

E

Okay, that's great.

B

uh Chris I'm not sure if you have any other comments on that on that piece. Around uh service principle user like where are the group teams or things are being used with the.

C

B

Okay, um awesome um uh any other questions on the database, connector or other pieces, um I, I, guess um or I, guess any other. um Any requests that you'd like to see any any like. We talked about the usability pieces and other pieces around making easier to manage um service principles and uh group teams with dataverse um curious. If folks have other um other pieces, um they'd like to see improved.

E

The only one I'd like to see improved is the the work around for getting trigger lock into work.

E

You put you put your when a row was changed and if you don't work out how to do trigger lock and you're going to get an infinite looping, so you want the flow not to work on if something, if another column is in a set state, but you then have to put in a filter array action after that step, just to build the logic to go back into the settings trigger in the actual dataverse step, it's the same in SharePoint, but that could be made a little bit easier, because that's that until somebody shows you is a whole big bunch of Blackmagic uh it just you just don't know what it does and whether it requires values or it just requires an actual event of things.

B

Yeah, so I can speak to one one area we want to improve. That is making it easier to use um in dataverse it's called the pre-image of the change um that comes through. So um you know, when you have a record changing, uh we want to make it easy to respond, like uh you know, query on meaningful changes from uh what the record was uh to what it's changed to, which can make it uh possible to have more more scope triggers uh avoid infinite Loops between states.

B

uh You don't intend uh so I think that's the main, uh the main piece we have around if it loops around more scoped triggers that may help with that um as well, um but definitely um yeah right now we have a warning: that'll appear if you have a table referring to itself later in the flow, but that's I know that's very general, it's just a hint that that may happen, but what you're looking for is just making sure that you can can scope those pieces right, yeah.

E

The the reason this came up was we. We recently moved a lot of stuff from SharePoint over the data version.

E

We still find the same issue is that you modularize your flow, so instead of having one super complicated flow, you've got free flows that do three different things when the data is in a different state in the same table, but because we're all looking at the same table, you end up trigger locking on an admin column that you've put in that says all right when, when the first action happens, change that column to new all right and then the second flow sits and looks at that table and goes. Has that rule changed?

E

Oh, it's changed and doesn't even start because it goes I'm waiting for it to say updated, which happens at the end of another flow. We say and that's that's where the trigger locking comes in really useful, but it was just to make it easier to find the values as the point where second.

B

Definitely definitely yeah we'll definitely look into how we can make that a little bit smoother, um avoiding um yeah. We know again, whenever you have many many flows triggering on the same pieces uh as well. um You know kind of hard to see the you know the whole layout of everything uh tied to that data tied to those same tables and things as well. So there's other improvements we're going to make that make that um that easier as well right.

C

B

Awesome awesome. Well, thank uh thank you. Everyone um if you have any other comments, um I think um Justin. You shared my LinkedIn feel free to Ping uh through there through the uh this uh community area, um I think Jocelyn and I, and the power to meet uh PMT, monitor the the power automate community. So please feel free to use the ideas and other parts of the power automate uh Community Forum um for any any other questions, or um you know, sharing things you've made um but yeah. Thank you. Everyone for joining.