Redigeeri

Jagamisviis:

Azure Data Studio extensibility

  • Artikkel

Azure Data Studio has several extensibility mechanisms to customize the user experience and make those customizations available to the entire user community. The core Azure Data Studio platform is built upon Visual Studio Code, so most of the Visual Studio Code extensibility APIs are available. Additionally, we've provided additional extensibility points for data management-specific activities.

Some of the key extensibility points are:

  • Visual Studio Code extensibility APIs
  • Azure Data Studio extension authoring tools
  • Manage Dashboard tab panel contributions
  • Insights with Actions experiences
  • Azure Data Studio extensibility APIs
  • Custom Data Provider APIs

Visual Studio Code extensibility APIs

Because the core Azure Data Studio platform is built upon Visual Studio Code, details about the Visual Studio Code extensibility APIs are found in the Extension Authoring and Extension API documentation on the Visual Studio Code website.

Note

Azure Data Studio releases are aligned with a recent version of VS Code, however the included VS Code engine may not be the current VS Code release. For example, in November 2020 the VS Code engine in Azure Data Studio is 1.48 and the current VS Code version is 1.51. The error message "Unable to install extension '<name>' as it is not compatible with VS Code <version>" when installing an extension is caused by an extension that has a later VS Code engine version defined in the package manifest (package.json). You can verify the VS Code engine version in your Azure Data Studio through the Help menu under About.

Manage Dashboard tab panel contributions

For details, see Contribution Points and Context Variables.

Azure Data Studio extensibility APIs

For details, see Extensibility APIs.

Contribution points

This section covers the various contribution points that are defined in the package.json extension manifest.

The IntelliSense is supported inside azuredatastudio.

Dashboard contribution points

Contribute a tab, container, and/or insight widget to the dashboard.

Dashboard

dashboard.tabs

Dashboard.tabs creates the tab sections inside the dashboard page. It expects an object or an array of objects.

"dashboard.tabs": [
{
	"id": "test-tab1",
	"title": "Test 1",
	"description": "The test 1 displays a list of widgets.",
	"when": "connectionProvider == 'MSSQL' && !mssql:iscloud",
	"alwaysShow": true,
	"container": {
		...
	}
}
]

dashboard.containers

Instead of specifying dashboard container inline (inside the dashboard.tab). You can register containers using dashboard.containers. It accepts an object or an array of the object.

"dashboard.containers": [
{
	"id": "innerTab1",
	"container": {
		...
	}
},
{
	"id": "innerTab2",
	"container": {
	   ...
	}
}
]

To refer to registered container, specify the id of the container

"dashboard.tabs": [
{
	...
	"container": {
		"innerTab1": {             
		}
	}
}
]

dashboard.insights

You can register insights using dashboard.insights. This is similar to Tutorial: Build a custom insight widget

"dashboard.insights": {
"id": "my-widget"
"type": {
	"count": {
		"dataDirection": "vertical",
		"dataType": "number",
		"legendPosition": "none",
           "labelFirstColumn": false,
		"columnsAsLabels": false
       }
   },
   "queryFile": "{your file folder}/activeSession.sql"
}

Dashboard container types

There are currently four supported container types:

  1. widgets-container

    Widgets container

    The list of widgets that will be displayed in the container. It's a flow layout. It accepts the list of widgets.

    "container": {
	"widgets-container": [
	{
		"widget": {
			"query-data-store-db-insight": {
			}
		}
	},
	{
		"widget": {
			"explorer-widget": {
			}
		}
	}
  ]
}

  2. webview-container

    webview container

    The webview will be displayed in the entire container. It expects webview id to be the same is tab ID

    "container": {
	"webview-container": null
}

  3. grid-container

    grid container

    The list of widgets or webviews that will be displayed in the grid layout

     "container": {
 	"grid-container": [
    {
    	"name": "widget 1",
    	"widget": {
    		"explorer-widget": {
    		}
    	},
    	"row":0,
    	"col":0
    },
    {
    	"name": "widget 2",
    	"widget": {
    		"tasks-widget": {
    			"backup", 
    			"restore",
    			"configureDashboard",
    			"newQuery"
    		}
    	},
    	"row":0,
    	"col":1
    },
    {
    	"name": "Webview 1",
    	"webview": {
    		"id": "google"
    	},
    	"row":1,
    	"col":0,
    	"colspan":2
    },
    {
    	"name": "widget 3",
    	"widget": {
    		"explorer-widget": {}
    	},
    	"row":0,
    	"col":3,
    	"rowspan":2
    },
]

  4. nav-section

    nav section

    The navigation section will be displayed in the container

    "container": {
	"nav-section": [
		{
			"id": "innerTab1",
			"title": "inner-tab1",
			"icon": {
				"light": "./icons/tab1Icon.svg",
				"dark": "./icons/tab1Icon_dark.svg"
			},
			"container": {
				...
			}
		},
		{
			"id": "innerTab2",
			"title": "inner-tab2",
			"icon": {
				"light": "./icons/tab2Icon.svg",
				"dark": "./icons/tab2Icon_dark.svg"
			},
			"container": {
				...
			}
		}
	]
}

Context variables

For general information about context in Visual Studio Code and subsequently Azure Data Studio, see Extensibility.

In Azure Data Studio, we have specific context around database connections available for extensions.

Dashboard

In dashboard, we provide the following context variables:

context variable description
connectionProvider A string of the identifier for the provider of the current connection. Ex. connectionProvider == 'MSSQL'.
serverName A string of the server name of the current connection. Ex. serverName == 'localhost'.
databaseName A string of the database name of the current connection. Ex. databaseName == 'master'.
connection The full connection profile object for the current connection (IConnectionProfile)
dashboardContext A string of the context of the page the dashboard is currently on. Either 'database' or 'server'. Ex. dashboardContext == 'database'