This tutorial shows you how to write an HTTP Cloud Run functionthat returns Spanner results.

Objectives

Write, deploy, and trigger anHTTP function that accessesSpanner.

Costs

This document uses Spanner and Cloud Run,which are billable components of Google Cloud.

• For information on the cost of using Spanner, seeSpanner pricing.

• For information on the cost of using Cloud Run, including freeinvocations, seeCloud Run pricing.

Before you begin

1. This document assumes you have a Spanner instance namedtest-instance and a database namedexample-db that uses the musicapplicationschema.For instructions on creating an instance and database with the music applicationschema, seeQuickstart using the console orthe tutorials for Getting Started inNode.js,orPython.

   Note: After this step, you should have two tables,Singers andAlbums,in your database with the sample data specified in the music applicationschema.

2. Enable the Cloud Run, and Cloud Build APIs.

   Enable the APIs

3. Install and initialize the gcloud CLI.

   If you already have the gcloud CLI installed, update it by running thefollowing command:

   gcloud components update

4. Prepare your development environment:

   Node.js

   See theNode.js setup guide.

   Python

   See thePython setup guide.

Prepare the application

1. Clone the sample app repository to your local machine:

   Node.js

   gitclonehttps://github.com/GoogleCloudPlatform/nodejs-docs-samples.git

   Alternatively, you candownload the sampleas a zip file and extract it.

   Python

   gitclonehttps://github.com/GoogleCloudPlatform/python-docs-samples.git

   Alternatively, you candownload the sampleas a zip file and extract it.

2. Change to the directory that contains the Cloud Run functionssample code for accessing Spanner:

   Node.js

   cdnodejs-docs-samples/functions/spanner

   Python

   cdpython-docs-samples/functions/spanner

3. Take a look at the sample code:

   Node.js

   // Imports the Google Cloud client libraryconst{Spanner}=require('@google-cloud/spanner');// Imports the functions framework to register your HTTP functionconstfunctions=require('@google-cloud/functions-framework');// Instantiates a clientconstspanner=newSpanner();// Your Cloud Spanner instance IDconstinstanceId='test-instance';// Your Cloud Spanner database IDconstdatabaseId='example-db';/** * HTTP Cloud Function. * * @param {Object} req Cloud Function request context. * @param {Object} res Cloud Function response context. */functions.http('spannerQuickstart',async(req,res)=>{// Gets a reference to a Cloud Spanner instance and databaseconstinstance=spanner.instance(instanceId);constdatabase=instance.database(databaseId);// The query to executeconstquery={sql:'SELECT * FROM Albums',};// Execute the querytry{constresults=awaitdatabase.run(query);constrows=results[0].map(row=>row.toJSON());rows.forEach(row=>{res.write(`SingerId:${row.SingerId}, `+`AlbumId:${row.AlbumId}, `+`AlbumTitle:${row.AlbumTitle}\n`);});res.status(200).end();}catch(err){res.status(500).send(`Error querying Spanner:${err}`);}});

   Python

   importfunctions_frameworkfromgoogle.cloudimportspannerinstance_id="test-instance"database_id="example-db"client=spanner.Client()instance=client.instance(instance_id)database=instance.database(database_id)@functions_framework.httpdefspanner_read_data(request):query="SELECT * FROM Albums"outputs=[]withdatabase.snapshot()assnapshot:results=snapshot.execute_sql(query)forrowinresults:output="SingerId:{}, AlbumId:{}, AlbumTitle:{}".format(*row)outputs.append(output)return"\n".join(outputs)

   The function sends a SQL query to fetch allAlbums data from yourdatabase. The function is executed when you make an HTTP requestto the function's endpoint.

Deploy the function

To deploy the function with an HTTP trigger, run the followingcommand in thespanner directory:

gcloudrundeploynodejs-spanner-function\--source.\--regionREGION\--functionspannerQuickstart\--base-imageRUNTIME_ID\--log-http

gcloudrundeploypython-spanner-function\--source.\--regionREGION\--functionspanner_read_data\--base-imageRUNTIME_ID\--log-http

Replace:

• REGION with the name of theGoogle Cloud region where you want to deploy your function(for example,us-west1).

• RUNTIME_ID with the appropriate runtime ID (forexample,nodejs22). SeeSupported language runtimes and base images.

Function deployment might take up to two minutes.

Note theurl value returned when your function finishes deploying. You willuse it when you trigger the function.

You can view your deployed functions on theCloud Run page in theGoogle Cloud console. You can also create and edit functions on that page, and getdetails and diagnostics for your functions.

Trigger the function

Make an HTTP request to your function:

curlURL

ReplaceURL with the URL value returned when your functionfinishes deploying.

You should see output that shows the results of theSQL query, assuming you worked through aGetting Started tutorialand populated the database:

SingerId:2,AlbumId:2,AlbumTitle:ForeverHoldYourPeaceSingerId:1,AlbumId:2,AlbumTitle:Go,Go,GoSingerId:2,AlbumId:1,AlbumTitle:GreenSingerId:2,AlbumId:3,AlbumTitle:TerrifiedSingerId:1,AlbumId:1,AlbumTitle:TotalJunk

You can also visit the function's URL in your browser to see the result ofyour SQL query.

• Learn more about writing Cloud Run functions.
• Learn more about deploying Cloud Run functions.
• Learn more about triggering Cloud Run functions.