Updates to Flow Studio App in 2023 October

There’s been a series of updates to Flow Studio App on our development build, as I’m preparing to push this next stable build to production, I thought I’d take this time to list down lots of changes we’ve done in this latest series of updates.

View fullsize

Production 1.1.51

View fullsize

Dev 1.1.58

Oh what happened to the red colour!

• Switching UI Control Set towards Fluent 2
  
  We are in the middle of switching the overall look and feel and UX experience from default Telerik component style to the new Telerik + Fluent style, which would be more inline with the experience in Power Platform and also M365.
  

• Switching primary colour from red to a “whale” colour
  
  First is switching away from the red colour - so we can actually use red to indicate serious issues happening elsewhere on the screen that demands your attention, like that "Flow Suspension” notice on the top-right.
  

• On Flows, Flow (Admin) and Flows (solution) screens, we added pagination controls, this greatly helps rendering the grid when you have hundreds of flows.
  Don’t worry, search and sort is applied prior to paging, so it won’t leave you with having to browse through several pages before finding your item.
  

• The grid menu dropdown filter had an bug fix that allows us to select the fields more accurately. Previosuly, the menu often lose focus and we aren’t able to easily select a field to filter.

• Flow Diagram had several fixes

• Approvals tab had more fixes.

• Settings tab is brought back - we will be making lots more user-configurable settings very soon.

• We tidied up the overall page styling and reduce wasted padding around the grid and window (but do give us feedback if we mess something up on your device)

The next lot of updates will begin to drop on dev branch really soon.

Power Automate API changes - v2 Admin scope now needs user_impersonation

This blog post is about two major updates to Flow Studio App and Flow Studio for Enterprise.

MSAL v2 Update

First one, we finally updated Flow Studio to MSAL v2 there’s a few reasons for this, but primarily, this is because we want to support modern browsers that are now by default disabling 3rd party cookies, which prevented previous authentication via hidden iframe method in earlier versions of MSAL and ADALjs.

Incidentally, this also means Flow Studio App now works on iPad and Safari. And should work better for many customers within enterprise that has 3rd party cookie disabled.

MSAL also supports multiple accounts so that’s an interesting scenario in the future to support multi-user or multi-tenancy? We’ll see.

Power Automate Admin API Scope

Secondly, we have a note on Power Automate API changes and how it affects us.

Power Automate /scopes/admin/v2/ supports fetching up to 250 flows per request prior to paging, by comparison, v1 only supports 50 flows. This means reading flows as admin is once again much quicker.

But we’ve also noticed that admin flow requests now need an additional user_impersonation scope.

“Access Microsoft Flow as signed in user” (nice name!)

When customers login to Flow Studio App v1.1.45 or later, you will be asked to re-consent due to this additional scope.

field notes using Power Automate with Power BI

I have been working at a client learning, testing and building Power BI reports, datasets, dataflows, datamart and dashboards for the last 8 months. With a fairly healthy dose of Power Automate thrown in. I wanted to take a short breather and write down some of the learnings so far. I also would really like to hear your feedback regarding what your Power BI best practices are.

I will be presenting my experiences and examples in the Australian Digital Workplace Conference next week in Melbourne. I hope I’ll be seeing you there.

The system components

• Call the API of a custom business application to retrieve project & product data, the system uses a NoSQL database, so we are pulling out pages of JSON via Power Automate.

• At the moment, we store these JSON files into a document library in SharePoint.

• Use Power BI dataflow to process and merge these JSON data into a staging table.

• Use Power BI dataflow to transform data from the staging dataflow to actually do work.

• Reports use the staging dataflow, and reference it if additional local transformations are needed.

• Fancy visuals with Deneb and HTML content visuals.

Calling APIs - we are using HTTP request and calculating nonce and api-key within Power Automate - this wasn’t something Power BI can call directly without some middleware. In particular, we are interested in a subset of projects from our dataset - so every evening, Power Automate calls Power BI Dataset Query, fetch a list of project codes and make API calls. It also checks if there’s been a change from the version stored in SharePoint and skips writing if the JSON has not been modified. (Unfortunately, we don’t have a last modified metadata from source).

We put JSON data into a SharePoint library and store them by months - this is because the business has a monthly reporting cycle, and we wanted data captured against last month vs this month. An alternative would be to use Azure Blob Storage for this staging area, because once we reached several hundred JSON files, SharePoint often throws too-busy error to Power BI during data refresh.

On re-use, we tried a myriad of methods, and we found using two dataflows to work the best. The first dataflow provides raw json content per file from SharePoint. This dataflow is configured for incremental refresh, so if the JSON isn’t updated, this doesn’t need to refresh in the dataflow (this solves our file too busy problem). Datamart can’t easily be used within the ETL of another dataflow (it’s more suitable for direct query).

Our second dataflow is where we do the transformation of the json data navigate the JSON structure and pull out records, lists accordingly.

We try to have very light local transformation or modification within the Power BI report. Instead, as much as possible, we want to run the transform within the dataflow.

This is a high-level overview blog post and there’s quite a bit more (smaller, bite sized) notes I want to write down in time, but this will do for now, and we’ll leave the rest for a future discussion.

Other design decisions, notes and thoughts

• What Power Automate brings to the table.

• What Power BI Dataflows brings to the table.

• Notes on using DAX and Power Query M

• Using Dataverse

• Considerations with using SharePoint or Azure Blob Storage

• Licensing costs

“Soon” Azure Data Factory and Microsoft Fabric offerings

Turning a new page

Wanted to write again, and let everyone know what I’ve been up to. I ended up taking a break through most of 2021 and 2022 simply resting, recovering, and doing light work from home.

2023 resumed with a big bang, I found motivation and drive to dive back into the many projects I’ve temporarily shelved in the last two years. I’ve also became pretty handy with a bunch of home DIY projects. It was a big change to the old me that only knew how to do digital projects but not physical projects. Perhaps more on that in a future post.

Flow Studio

We’ve had several Flow Studio fixes in the last two months

• there was an API pagination fix since the API no longer accepts 250 records at once and restricts us to only 50. (That means more pages and API call takes longer)

• API auth fix relating to Power Apps is in-progress.

• There’s a second API skip/continuation token fix.

• We’ve also tweaked the way trail is applied when anyone wants to try Flow Studio pro - you can sign up and trial will be available for two weeks - you can cancel the subscription before the trial end date to avoid being charged, if Flow Studio isn’t suitable for you.

Clarity / Flow Studio for Teams and Enterprise

We’ve had renewed interest in Flow Studio for Teams and Enterprise (Clarity) through the last few years.

• Flow Studio for Teams will be tweaked to focus on monitoring critical flows and alerting users when their business critical flows fail. This will be priced simply and does not offer governance capabilities.

• Flow Studio for Enterprise will be focused on the turnkey Power Platform Governance story, adding new features to scan more areas of the Power Platform, and integrate with CoE starter kit.

• So far this year, we’ve added BYO Azure Storage. There’s been a lot of fixes to API breakages in Power Apps area this year.

Contract Work

I started a regular part time contract work in Sydney CBD, so if you are local, hit me up for a coffee.

• I’m working with a lot of Power BI reports

• There’s a lot of Power Automate doing the heavy lifting as well.

• We are also talking about adding some Power Apps visuals to allow executive comments to be collected during a report presentation.

Community

Several of the meetups, conferences and events that I used to participate in are becoming active again. I hope to see more of the community not just virtually, but physically as well. I hope to be able to grab a coffee with you soon.

Parse CSV through Code in Power Automate Custom Connection

I was inspired reading Alex Shlega and Hiroaki Nagao ’s posts on using code with custom connections. So I set out to give it a go and work on another common problem I have: Parse CSV

First a picture showing you how it works.

View fullsize

Give it text, the action returns array of arrays.

Microsoft’s docs are here.
Write code in a custom connector | Microsoft Docs

And particularly, I need to parse CSV without using additional libraries, and using only the existing libraries available here. I noted that we do have access to System.Text.RegularExpressions, so I started my planning there.

Because parsing CSV correctly is a problem best sorted via use of a tokenizer, I went looking for a regular expression pattern that treats each line as a series of tokens. There are many patterns, but I like this one that I found on stackoverflow the best for my needs. https://stackoverflow.com/a/48806378

Code

So the code takes all the content of the body and splits by line breaks, then the regular expression is run over every line using Matches (this method returns multiple matches giving us a MatchCollection of tokens). In each match, I look for Group[2] which is the value without quotes “ and “. But if failing that match, we take Group[1] value.
We do not take the Match.Value because that would include the comma.

/end of regular expression explanation.

We cast the matches back to array via Linq and then back to JArray and return that back to Flow.

public class Script : ScriptBase
{
    public override async Task<HttpResponseMessage> ExecuteAsync()
    {
        if (this.Context.OperationId == "csv")
        {
            return await this.HandleCSV().ConfigureAwait(false);
        }

        HttpResponseMessage response = new HttpResponseMessage(HttpStatusCode.BadRequest);
        response.Content = CreateJsonContent($"Unknown operation ID '{this.Context.OperationId}'");
        return response;
    }

    private async Task<HttpResponseMessage> HandleCSV()
    {
        var contentAsString = await this.Context.Request.Content.ReadAsStringAsync().ConfigureAwait(false);

        // (?:,|\n|^)("(?:(?:"")*[^"]*)*"|[^",\n]*|(?:\n|$))
        // https://stackoverflow.com/a/48806378

        var re = new Regex("(?!$)(\"((?:(?:\"\")*[^\"]*)*)\"|[^\",\r\n]*)(?:,|$)");
        var lines = Regex.Split(contentAsString, "\r\n|\r|\n");

        var result = new JArray(lines.Select(line=>{ 
            var matches = re.Matches(line);
            return new JArray(matches.Cast<Match>().Select(match => {
                return match.Groups[2].Success ?
                match.Groups[2].Value :
                match.Groups[1].Value;
            } ).ToArray());
        }).ToArray());

        var response = new HttpResponseMessage(HttpStatusCode.OK);
        response.Content = CreateJsonContent(result.ToString());
        return response;
    }
}

View fullsize

explain the regex and match groups

Swagger

This is the custom connection swagger YAML file.

swagger: '2.0'
info: {title: CustomCode, description: Custom Code, version: '1.0'}
host: johnliu.net
basePath: /
schemes: [https]
consumes: []
produces: []
paths:
  /Csv:
    post:
      responses:
        default:
          description: default
          schema:
            type: array
            items: {}
            description: Array
            title: Array
      summary: Parse CSV
      description: Parse CSV
      operationId: csv
      parameters:
      - name: value
        in: body
        required: true
        schema: {type: string, description: Text, title: value}
      x-ms-visibility: important
definitions: {}
parameters: {}
responses: {}
securityDefinitions: {}
security: []
tags: []

I want to add more parameters over time, and that will involve a tweak to the input parameters on the Swagger definition. But that’s probably a task for another day.

Links:

Write code in a custom connector | Microsoft Docs

C# code in Power Automate: let’s sort a string array? | It Ain't Boring (itaintboring.com)

Calculate Sum & Average in Power Automate using C# code in a custom connector - MoreBeerMorePower (hatenablog.com)