RedisGears Python 快速入门

了解如何将 RedisGears 与 Python 结合使用的快速入门。

Redis 堆栈

对于本教程，您需要：

RedisGears 基础知识

在本快速入门指南中，我们将了解如何使用 RedisGears 执行批处理和事件处理。

使用 RedisGears，批处理意味着处理已存储在 Redis 数据库中的数据。事件处理是指处理对 Redis 密钥空间的更改。

以下示例假定 Redis 数据库为空。

批处理

让我们从最简单的示例开始。从redis-cli，请执行以下命令。

redis.cloud:6379> RG.PYEXECUTE "GearsBuilder().run()"
1) (empty array)
2) (empty array)

This command doesn't do much; it simply iterates over the keyspace. Let's add a key and run it again:

redis.cloud:6379> SET message "hello world"
OK
redis.cloud:6379> RG.PYEXECUTE "GearsBuilder().run()"
1) 1) "{'event': None, 'key': 'message', 'type': 'string', 'value': 'hello world'}"
2) (empty array)

We've added a single string, and you can see that this gears function processes it, even though it does nothing with the data. Let's actually do something with the data. So first, we'll add a few more strings:

redis.cloud::6379> SET message:2 "hello galaxy"
OK
redis.cloud:6379> SET message:3 "hello universe"
OK

We now have three strings in our database. Suppose we want to perform a unique word count on these strings. We can write a RedisGears function to do this just that. So open a file called wordcount.py, and add the following code:

gb = GearsBuilder()
gb.map(lambda x: x['value'])     # map each key object to its string value
gb.flatmap(lambda x: x.split())  # split each string into a list of words
gb.countby()                     # run a count-unique on these words
gb.run()

There are two ways to load files into RedisGears. For production deployments, we recommend using the special gears-cli. However, for the purpose of this demonstration, the easiest way is to pass the filename through the redis-cli command, like so:

$ redis-cli rg.pyexecute "`cat wordcount.py`"
1) 1) "{'key': 'world', 'value': 1}"
   2) "{'key': 'galaxy', 'value': 1}"
   3) "{'key': 'hello', 'value': 3}"
   4) "{'key': 'universe', 'value': 1}"
2) (empty array)

The results here show the number of occurences of each word in all of our strings. So, we've effectively processed the data in our Redis database all at once, in a batch.

Event processing

You may have noticed that all of the RedisGears functions above end with a call to run(). This indicates that the function should be run immediately on the data in the Redis database. But what if you want to process data as it arrives in Redis? In that case, your functions will end with a call to register(), which will store the function and apply it as events occur in Redis.

Let's see how to register a function. First, suppose we're writing hashes to our database that represent users. They take the following form:

redis.cloud:6379> HSET person:3 name "Summer Smith" age 17
(integer) 2
redis.cloud:6379> HSET person:4 name "James Jameson" age 21
(integer) 2

Each hash has two fields, one containing a name and the other an age. Now, suppose we want to keep a record of the maximum age of all users. We can register a RedisGears function to do this. Open up a file called maxage.py, and add the following code:

def age(x):
  ''' Extracts the age from a person's record '''
  return int(x['value']['age'])

def compare_and_swap(x):
  ''' Checks and sets the current maximum '''
  k = 'age:maximum'
  v = execute('GET', k)   # read key's current value
  v = int(v) if v else 0  # initialize to 0 if None
  if x > v:               # if a new maximum found
    execute('SET', k, x)  # set key to new value

# Event handling function registration
gb = GearsBuilder()
gb.map(age) # Extract the 'age' field from each hash
gb.foreach(compare_and_swap) # Compare the max age to the value stored at age:maximum
gb.register(prefix='person:*') # Only process keys matching the pattern 'person:*'

You can see here that we define two methods: age() and compare_and_swap(). Even if you're not familiar with Python, you should be able to see what the methods do.

Below the method definitions is the RedisGears data flow that we're defining. Notice that at the end we call register() to register the function to listen for events.

To load this function into RedisGears, run the following:

$ redis-cli RG.PYEXECUTE "`cat maxage.py`"

Now start the redis-cli, and create a couple of hashes:

redis.cloud:6379> HSET person:5 name "Marek Michalski" age 17
(integer) 2
redis.cloud:6379> HSET person:6 name "Noya Beit" age 21
(integer) 2

To see if the RedisGears function is working, check the value of age:maximum:

redis.cloud:6379> GET age:maximum
"21"

Next steps

You should now have a basic idea of how to run RedisGears functions for batch and event processing. If you're interested in write-behind caching, see our write-behind caching overview.