Azure Cache for Redis input binding for Azure Functions
When a function runs, the Azure Cache for Redis input binding retrieves data from a cache and passes it to your function as an input parameter.
For information on setup and configuration details, see the overview.
Scope of availability for functions bindings
Binding Type | Azure Managed Redis | Azure Cache for Redis |
---|---|---|
Input | Yes | Yes |
Important
When using Azure Managed Redis or the Enterprise tiers of Azure Cache for Redis, use port 10000 rather than port 6380 or 6379.
Important
The Node.js v4 model for Functions isn't yet supported by the Azure Cache for Redis extension. For more details about how the v4 model works, refer to the Azure Functions Node.js developer guide. To learn more about the differences between v3 and v4, refer to the migration guide.
Important
The Python v2 model for Functions isn't yet supported by the Azure Cache for Redis extension. For more details about how the v2 model works, refer to the Azure Functions Python developer guide.
Example
A C# function can be created by using one of the following C# modes:
- Isolated worker model: Compiled C# function that runs in a worker process that's isolated from the runtime. Isolated worker process is required to support C# functions running on LTS and non-LTS versions .NET and the .NET Framework. Extensions for isolated worker process functions use
Microsoft.Azure.Functions.Worker.Extensions.*
namespaces. - In-process model: Compiled C# function that runs in the same process as the Functions runtime. In a variation of this model, Functions can be run using C# scripting, which is supported primarily for C# portal editing. Extensions for in-process functions use
Microsoft.Azure.WebJobs.Extensions.*
namespaces.
Important
For .NET functions, using the isolated worker model is recommended over the in-process model. For a comparison of the in-process and isolated worker models, see differences between the isolated worker model and the in-process model for .NET on Azure Functions.
The following code uses the key from the pub/sub trigger to obtain and log the value from an input binding using a GET
command:
using Microsoft.Extensions.Logging;
namespace Microsoft.Azure.Functions.Worker.Extensions.Redis.Samples.RedisInputBinding
{
public class SetGetter
{
private readonly ILogger<SetGetter> logger;
public SetGetter(ILogger<SetGetter> logger)
{
this.logger = logger;
}
[Function(nameof(SetGetter))]
public void Run(
[RedisPubSubTrigger(Common.connectionStringSetting, "__keyevent@0__:set")] string key,
[RedisInput(Common.connectionStringSetting, "GET {Message}")] string value)
{
logger.LogInformation($"Key '{key}' was set to value '{value}'");
}
}
}
More samples for the Azure Cache for Redis input binding are available in the GitHub repository.
The following code uses the key from the pub/sub trigger to obtain and log the value from an input binding using a GET
command:
package com.function.RedisInputBinding;
import com.microsoft.azure.functions.*;
import com.microsoft.azure.functions.annotation.*;
import com.microsoft.azure.functions.redis.annotation.*;
public class SetGetter {
@FunctionName("SetGetter")
public void run(
@RedisPubSubTrigger(
name = "key",
connection = "redisConnectionString",
channel = "__keyevent@0__:set")
String key,
@RedisInput(
name = "value",
connection = "redisConnectionString",
command = "GET {Message}")
String value,
final ExecutionContext context) {
context.getLogger().info("Key '" + key + "' was set to value '" + value + "'");
}
}
This function.json defines both a pub/sub trigger and an input binding to the GET message on an Azure Cache for Redis instance:
{
"bindings": [
{
"type": "redisPubSubTrigger",
"connection": "redisConnectionString",
"channel": "__keyevent@0__:set",
"name": "key",
"direction": "in"
},
{
"type": "redis",
"connection": "redisConnectionString",
"command": "GET {Message}",
"name": "value",
"direction": "in"
}
],
"scriptFile": "index.js"
}
This JavaScript code (from index.js) retrieves and logs the cached value related to the key provided by the pub/sub trigger.
module.exports = async function (context, key, value) {
context.log("Key '" + key + "' was set to value '" + value + "'");
}
This function.json defines both a pub/sub trigger and an input binding to the GET message on an Azure Cache for Redis instance:
{
"bindings": [
{
"type": "redisPubSubTrigger",
"connection": "redisConnectionString",
"channel": "__keyevent@0__:set",
"name": "key",
"direction": "in"
},
{
"type": "redis",
"connection": "redisConnectionString",
"command": "GET {Message}",
"name": "value",
"direction": "in"
}
],
"scriptFile": "run.ps1"
}
This PowerShell code (from run.ps1) retrieves and logs the cached value related to the key provided by the pub/sub trigger.
param($key, $value, $TriggerMetadata)
Write-Host "Key '$key' was set to value '$value'"
The following example uses a pub/sub trigger with an input binding to the GET message on an Azure Cache for Redis instance. The example depends on whether you use the v1 or v2 Python programming model.
This function.json defines both a pub/sub trigger and an input binding to the GET message on an Azure Cache for Redis instance:
{
"bindings": [
{
"type": "redisPubSubTrigger",
"connection": "redisConnectionString",
"channel": "__keyevent@0__:set",
"name": "key",
"direction": "in"
},
{
"type": "redis",
"connection": "redisConnectionString",
"command": "GET {Message}",
"name": "value",
"direction": "in"
}
]
}
This Python code (from __init__.py) retrieves and logs the cached value related to the key provided by the pub/sub trigger:
import logging
def main(key: str, value: str):
logging.info("Key '" + key + "' was set to value '" + value + "'")
The configuration section explains these properties.
Attributes
Note
Not all commands are supported for this binding. At the moment, only read commands that return a single output are supported. The full list can be found here
Attribute property | Description |
---|---|
Connection |
The name of the application setting that contains the cache connection string, such as: <cacheName>.redis.cache.windows.net:6380,password... |
Command |
The redis-cli command to be executed on the cache with all arguments separated by spaces, such as: GET key , HGET key field . |
Annotations
The RedisInput
annotation supports these properties:
Property | Description |
---|---|
name |
The name of the specific input binding. |
connection |
The name of the application setting that contains the cache connection string, such as: <cacheName>.redis.cache.windows.net:6380,password... |
command |
The redis-cli command to be executed on the cache with all arguments separated by spaces, such as: GET key or HGET key field . |
Configuration
The following table explains the binding configuration properties that you set in the function.json file.
function.json property | Description |
---|---|
connection |
The name of the application setting that contains the cache connection string, such as: <cacheName>.redis.cache.windows.net:6380,password... |
command |
The redis-cli command to be executed on the cache with all arguments separated by spaces, such as: GET key , HGET key field . |
Note
Python v2 and Node.js v4 for Functions don't use function.json to define the function. Both of these new language versions aren't currently supported by Azure Redis Cache bindings.
See the Example section for complete examples.
Usage
The input binding expects to receive a string from the cache.
When you use a custom type as the binding parameter, the extension tries to deserialize a JSON-formatted string into the custom type of this parameter.