sahib-sawhney-wh/plugins/dataverse/skills/dataverse-power-platform

dataverse-power-platform

This skill provides guidance on integrating with the Microsoft Power Platform ecosystem. Use when users ask about "Power Apps", "Power Automate", "Power BI", "Dataverse connector", "canvas app", "model-driven app", "Power Automate flow", or need help with Power Platform integration.

Power Apps Integration

Canvas Apps with Dataverse

Canvas apps can connect to Dataverse tables as data sources.

Connection setup:

1. Add Dataverse connector in Power Apps
2. Select tables to use
3. Use formulas to read/write data

Common formulas:

// Read all accounts
ClearCollect(colAccounts, Accounts)

// Filter records
Filter(Accounts, Status = "Active")

// Create record
Patch(Accounts, Defaults(Accounts), {Name: "New Account"})

// Update record
Patch(Accounts, LookUp(Accounts, AccountId = varId), {Name: "Updated"})

// Delete record
Remove(Accounts, LookUp(Accounts, AccountId = varId))

Model-Driven Apps

Model-driven apps are built directly on Dataverse tables.

Key components:

• Tables - Define the data structure
• Forms - UI for viewing/editing records
• Views - Lists of records with filters
• Business rules - Validation logic
• Dashboards - Analytics and charts

Best practices:

1. Design tables before building the app
2. Use business rules for validation
3. Create multiple views for different scenarios
4. Use security roles for access control

Power Automate Integration

Dataverse Triggers

When a row is added, modified or deleted
├── Table name: Accounts
├── Scope: Organization
└── Filter rows: statecode eq 0

Dataverse Actions

// List rows
List rows
├── Table name: Accounts
├── Select columns: name,telephone1
├── Filter rows: statecode eq 0
└── Row count: 100

// Get a row by ID
Get a row by ID
├── Table name: Accounts
└── Row ID: @{triggerOutputs()?['body/accountid']}

// Add a new row
Add a new row
├── Table name: Accounts
└── Name: "New Account"

// Update a row
Update a row
├── Table name: Accounts
├── Row ID: @{variables('accountId')}
└── Name: "Updated Name"

// Delete a row
Delete a row
├── Table name: Accounts
└── Row ID: @{variables('accountId')}

Flow Patterns

Sync data from external system:

HTTP Request received
    ↓
Parse JSON
    ↓
Apply to each (items)
    ↓
    ├── List rows (check if exists)
    ├── Condition (exists?)
    │   ├── Yes → Update a row
    │   └── No → Add a new row

Send notification on record change:

When a row is added (Accounts)
    ↓
Get a row by ID (Owner)
    ↓
Send an email (to Owner)

Power BI Integration

Direct Dataverse Connection

1. Get Data > Dataverse
2. Sign in with your account
3. Select tables
4. Load or Transform data

Query Optimization

// Use query folding
let
    Source = CommonDataService.Database("https://yourorg.crm.dynamics.com"),
    accounts = Source{[Schema="dbo", Item="account"]}[Data],
    filtered = Table.SelectRows(accounts, each [statecode] = 0),
    selected = Table.SelectColumns(filtered, {"name", "revenue"})
in
    selected

Best Practices for Power BI

1. Use query folding - Let Dataverse do filtering
2. Limit columns - Only import needed fields
3. Use incremental refresh - For large datasets
4. Schedule refreshes - During off-peak hours

Python to Power Platform

Call Power Automate from Python

import requests

# HTTP trigger URL from Power Automate
flow_url = "https://prod-xx.westus.logic.azure.com/workflows/..."

# Trigger the flow
response = requests.post(flow_url, json={
    "account_name": "New Account",
    "source": "Python App"
})

Sync Python App with Dataverse

# Python app writes to Dataverse
# Power Automate flow triggers on change
# Sends notifications, updates other systems

client = get_client()
client.create("account", {"name": "From Python"})
# Power Automate flow handles the rest

References

• See references/power-apps.md for Power Apps patterns
• See references/power-automate.md for flow examples
• See references/power-bi.md for reporting patterns