Execute queries in the SDK

Completed

The Microsoft.Azure.Cosmos.Container class in the SDK has a series of built-in classes to create a query, issue the query to Azure Cosmos DB for NoSQL, set up an asynchronous stream in C#, and return items efficiently back to the client.

To start, let’s use a straightforward SQL command that returns all products:

SELECT * FROM products p

The equivalent command definition in C# would use the QueryDefinition class:

QueryDefinition query = new ("SELECT * FROM products p");

First, define a C# type to represent the type of item you query, for this example, a simple C# Product class suffices:

public class Product
{
    public string id { get; set; }

    public string name { get; set; }

    public string price { get; set; }
}

Next, use the GetItemQueryIterator generic method with the C# Product type to in the await foreach loop. The asynchronous stream structure automatically handles the looping and pagination to go to the server and get each subsequent page of results. Within the foreach loop, add your code to handle each item; in this example, each item’s id, name, and price is output to the console:

using (FeedIterator<Product> feedIterator = this.Container.GetItemQueryIterator<Product>(
    query,
    null,
    new QueryRequestOptions() { }))
{
    while (feedIterator.HasMoreResults)
    {
        foreach(var item in await feedIterator.ReadNextAsync())
        {
            Console.WriteLine($"[{item.productid}]\t{item.name,35}\t{item.price,15:C}");
        }
    }
}

The azure-cosmos Python SDK provides efficient methods for querying data using SQL-like syntax, handling pagination, and streaming results back to the client.

To start, let’s use a straightforward SQL command that returns all products:

SELECT * FROM products p

In Python, this query can be passed directly as a string to the query_items method.

Define a Python dictionary to represent the structure of your items. For example, consider a product structure like this:

product = {
    "id": "string",
    "name": "string",
    "price": "string"
}

Next, use the query_items method. The method accepts the SQL query string and options for partition key and other configurations. The results are streamed efficiently using a loop. Within the loop, add your code to handle each item; in this example, each item’s id, name, and price is output to the console:

query = "SELECT * FROM products p"
for item in container.query_items(
    query=query,
    enable_cross_partition_query=True):
    print(f"{item['id']}\t{item['name']}\t{item['price']}")

The JavaScript SDK allows you to create a query, issue the query to Azure Cosmos DB for NoSQL, set up an asynchronous stream in JavaScript, and return items efficiently back to the client.

To start, let’s use a straightforward SQL command that returns all products:

SELECT * FROM products p

To execute this query using the JavaScript SDK, create a QuerySpec object or use a query string. Here's how you define the query string in your code:

const query = "SELECT * FROM products p";

For structured handling of query results, you can define a JavaScript class representing the Product data:

class Product {
    constructor(id, name, price) {
        this.id = id;
        this.name = name;
        this.price = price;
    }
}

Next, use the query method to create an instance of the QueryIterator class. The method accepts the SQL query string and options for partition key and other configurations. The asynchronous stream structure automatically handles the looping and pagination to go to the server and get each subsequent page of results. Within the loop, add your code to handle each item; in this example, each item’s id, name, and price is output to the console:

const iterator = container.items.query(
    query,
    { enableCrossPartitionQuery: true }
);

while (iterator.hasMoreResults()) {
    const { resources } = await iterator.fetchNext();
    for (const item of resources) {
        console.log(`[${item.id}]	${item.name.padEnd(35)}	${item.price.toFixed(2)}`);
    }
}

If you prefer to use the Product class to represent your items, you can use the following code:

const iterator = container.items.query(
    query,
    { enableCrossPartitionQuery: true }
);

while (iterator.hasMoreResults()) {
    const { resources } = await iterator.fetchNext();
    // Bind the results to the Product class.
    const products = resources.map((product) => {
        return new Product(product.id, product.name, 0);
    });
}