The official Redis documentation does a great job of explaining each command in detail. redis-py exposes two client classes that implement these commands. The StrictRedis class attempts to adhere to the official official command syntax.
There are a few exceptions:
In addition to the changes above, the Redis class, a subclass of StrictRedis, overrides several other commands to provide backwards compatibility with older versions of redis-py:
Provides backwards compatibility with older versions of redis-py that changed arguments to some commands to be more Pythonic, sane, or by accident.
Remove the first num occurrences of elements equal to value from the list stored at name.
Return a new pipeline object that can queue multiple commands for later execution. transaction indicates whether all commands should be executed atomically. Apart from making a group of operations atomic, pipelines are useful for reducing the back-and-forth overhead between the client and server.
Set the value of key name to value that expires in time seconds. time can be represented by an integer or a Python timedelta object.
NOTE: The order of arguments differs from that of the official ZADD command. For backwards compatability, this method accepts arguments in the form of name1, score1, name2, score2, while the official Redis documents expects score1, name1, score2, name2.
If you’re looking to use the standard syntax, consider using the StrictRedis class. See the API Reference section of the docs for more information.
Set any number of element-name, score pairs to the key name. Pairs can be specified in two ways:
As *args, in the form of: name1, score1, name2, score2, ... or as **kwargs, in the form of: name1=score1, name2=score2, ...
The following example would add four values to the ‘my-key’ key: redis.zadd(‘my-key’, ‘name1’, 1.1, ‘name2’, 2.2, name3=3.3, name4=4.4)
Implementation of the Redis protocol.
This abstract class provides a Python interface to all Redis commands and an implementation of the Redis protocol.
Connection and Pipeline derive from this, implementing how the commands are sent and received to the Redis server
Appends the string value to the value at key. If key doesn’t already exist, create it with a value of value. Returns the new length of the value at key.
Tell the Redis server to rewrite the AOF file from data in memory.
Tell the Redis server to save its data to disk. Unlike save(), this method is asynchronous and returns immediately.
LPOP a value off of the first non-empty list named in the keys list.
If none of the lists in keys has a value to LPOP, then block for timeout seconds, or until a value gets pushed on to one of the lists.
If timeout is 0, then block indefinitely.
RPOP a value off of the first non-empty list named in the keys list.
If none of the lists in keys has a value to LPOP, then block for timeout seconds, or until a value gets pushed on to one of the lists.
If timeout is 0, then block indefinitely.
Pop a value off the tail of src, push it on the head of dst and then return it.
This command blocks until a value is in src or until timeout seconds elapse, whichever is first. A timeout value of 0 blocks forever.
Return a dictionary of configuration based on the pattern
Set config item name with value
Returns the number of keys in the current database
Returns version specific metainformation about a give key
Decrements the value of key by amount. If no key exists, the value will be initialized as 0 - amount
Delete one or more keys specified by names
Echo the string back from the server
Execute a command and return a parsed response
Returns a boolean indicating whether key name exists
Set an expire flag on key name for time seconds. time can be represented by an integer or a Python timedelta object.
Set an expire flag on key name. when can be represented as an integer indicating unix time or a Python datetime object.
Delete all keys in all databases on the current host
Delete all keys in the current database
Return the value at key name, or None if the key doesn’t exist
Returns a boolean indicating the value of offset in name
Returns the substring of the string value stored at key, determined by the offsets start and end (both are inclusive)
Set the value at key name to value if key doesn’t exist Return the value at key name atomically
Delete keys from hash name
Returns a boolean indicating if key exists within hash name
Return the value of key within the hash name
Return a Python dict of the hash’s name/value pairs
Increment the value of key in hash name by amount
Return the list of keys within hash name
Return the number of elements in hash name
Returns a list of values ordered identically to keys
Sets each key in the mapping dict to its corresponding value in the hash name
Set key to value within hash name Returns 1 if HSET created a new field, otherwise 0
Set key to value within hash name if key does not exist. Returns 1 if HSETNX created a field, otherwise 0.
Return the list of values within hash name
Increments the value of key by amount. If no key exists, the value will be initialized as amount
Returns a dictionary containing information about the Redis server
Returns a list of keys matching pattern
Return a Python datetime object representing the last time the Redis database was saved to disk
Return the item from list name at position index
Negative indexes are supported and will return an item at the end of the list
Insert value in list name either immediately before or after [where] refvalue
Returns the new length of the list on success or -1 if refvalue is not in the list.
Return the length of the list name
Return a new Lock object using key name that mimics the behavior of threading.Lock.
If specified, timeout indicates a maximum life for the lock. By default, it will remain locked until release() is called.
sleep indicates the amount of time to sleep per loop iteration when the lock is in blocking mode and another client is currently holding the lock.
Remove and return the first item of the list name
Push values onto the head of the list name
Push value onto the head of the list name if name exists
Return a slice of the list name between position start and end
start and end can be negative numbers just like Python slicing notation
Remove the first count occurrences of elements equal to value from the list stored at name.
Set position of list name to value
Trim the list name, removing all values not within the slice between start and end
start and end can be negative numbers just like Python slicing notation
Returns a list of values ordered identically to keys
Moves the key name to a different Redis database db
Sets each key in the mapping dict to its corresponding value
Sets each key in the mapping dict to its corresponding value if none of the keys are already set
Return the encoding, idletime, or refcount about the key
Parses a response from the Redis server
Removes an expiration on name
Ping the Redis server
Return a new pipeline object that can queue multiple commands for later execution. transaction indicates whether all commands should be executed atomically. Apart from making a group of operations atomic, pipelines are useful for reducing the back-and-forth overhead between the client and server.
Publish message on channel. Returns the number of subscribers the message was delivered to.
Return a Publish/Subscribe object. With this object, you can subscribe to channels and listen for messages that get published to them.
Returns the name of a random key
Rename key src to dst
Rename key src to dst if dst doesn’t already exist
Remove and return the last item of the list name
RPOP a value off of the src list and atomically LPUSH it on to the dst list. Returns the value.
Push values onto the tail of the list name
Push value onto the tail of the list name if name exists
Add value(s) to set name
Tell the Redis server to save its data to disk, blocking until the save is complete
Return the number of elements in set name
Return the difference of sets specified by keys
Store the difference of sets specified by keys into a new set named dest. Returns the number of keys in the new set.
Set the value at key name to value
Set a custom Response Callback
Flag the offset in name as value. Returns a boolean indicating the previous value of offset.
Set the value of key name to value that expires in time seconds. time can be represented by an integer or a Python timedelta object.
Set the value of key name to value if key doesn’t exist
Overwrite bytes in the value of name starting at offset with value. If offset plus the length of value exceeds the length of the original value, the new value will be larger than before. If offset exceeds the length of the original value, null bytes will be used to pad between the end of the previous value and the start of what’s being injected.
Returns the length of the new string.
Shutdown the server
Return the intersection of sets specified by keys
Store the intersection of sets specified by keys into a new set named dest. Returns the number of keys in the new set.
Return a boolean indicating if value is a member of set name
Set the server to be a replicated slave of the instance identified by the host and port. If called without arguements, the instance is promoted to a master instead.
Return all members of the set name
Move value from set src to set dst atomically
Sort and return the list, set or sorted set at name.
start and num allow for paging through the sorted data
desc allows for reversing the sort
alpha allows for sorting lexicographically rather than numerically
Remove and return a random member of set name
Return a random member of set name
Remove values from set name
Return the number of bytes stored in the value of name
Return a substring of the string at key name. start and end are 0-based integers specifying the portion of the string to return.
Return the union of sets specifiued by keys
Store the union of sets specified by keys into a new set named dest. Returns the number of keys in the new set.
Convenience method for executing the callable func as a transaction while watching all keys specified in watches. The ‘func’ callable should expect a single arguement which is a Pipeline object.
Returns the number of seconds until the key name will expire
Returns the type of key name
Unwatches the value at key name, or None of the key doesn’t exist
Watches the values at keys names, or None if the key doesn’t exist
Set any number of score, element-name pairs to the key name. Pairs can be specified in two ways:
As *args, in the form of: score1, name1, score2, name2, ... or as **kwargs, in the form of: name1=score1, name2=score2, ...
The following example would add four values to the ‘my-key’ key: redis.zadd(‘my-key’, 1.1, ‘name1’, 2.2, ‘name2’, name3=3.3, name4=4.4)
Return the number of elements in the sorted set name
Increment the score of value in sorted set name by amount
Intersect multiple sorted sets specified by keys into a new sorted set, dest. Scores in the destination will be aggregated based on the aggregate, or SUM if none is provided.
Return a range of values from sorted set name between start and end sorted in ascending order.
start and end can be negative, indicating the end of the range.
desc a boolean indicating whether to sort the results descendingly
withscores indicates to return the scores along with the values. The return type is a list of (value, score) pairs
score_cast_func a callable used to cast the score return value
Return a range of values from the sorted set name with scores between min and max.
If start and num are specified, then return a slice of the range.
withscores indicates to return the scores along with the values. The return type is a list of (value, score) pairs
score_cast_func` a callable used to cast the score return value
Returns a 0-based value indicating the rank of value in sorted set name
Remove member values from sorted set name
Remove all elements in the sorted set name with ranks between min and max. Values are 0-based, ordered from smallest score to largest. Values can be negative indicating the highest scores. Returns the number of elements removed
Remove all elements in the sorted set name with scores between min and max. Returns the number of elements removed.
Return a range of values from sorted set name between start and num sorted in descending order.
start and num can be negative, indicating the end of the range.
withscores indicates to return the scores along with the values The return type is a list of (value, score) pairs
score_cast_func a callable used to cast the score return value
Return a range of values from the sorted set name with scores between min and max in descending order.
If start and num are specified, then return a slice of the range.
withscores indicates to return the scores along with the values. The return type is a list of (value, score) pairs
score_cast_func a callable used to cast the score return value
Returns a 0-based value indicating the descending rank of value in sorted set name
Return the score of element value in sorted set name
Union multiple sorted sets specified by keys into a new sorted set, dest. Scores in the destination will be aggregated based on the aggregate, or SUM if none is provided.
Manages TCP communication to and from a Redis server
Connects to the Redis server if not already connected
Disconnects from the Redis server
Return a bytestring representation of the value
Initialize the connection, authenticate and select a database
Pack a series of arguments into a value Redis command
Read the response from a previously sent command
Pack and send a command to the Redis server
Send an already packed command to the Redis server
Generic connection pool
Disconnects all connections in the pool
Get a connection from the pool
Create a new connection
Releases the connection back to the pool
>>> pool = redis.ConnectionPool(host='localhost', port=6379, db=0)
>>> r = redis.Redis(connection_pool=pool)
The client class uses a set of callbacks to cast Redis responses to the appropriate Python type. There are a number of these callbacks defined on the Redis client class in a dictionary called RESPONSE_CALLBACKS.
Custom callbacks can be added on a per-instance basis using the set_response_callback method. This method accepts two arguments: a command name and the callback. Callbacks added in this manner are only valid on the instance the callback is added to. If you want to define or override a callback globally, you should make a subclass of the Redis client and add your callback to its REDIS_CALLBACKS class dictionary.
Response callbacks take at least one parameter: the response from the Redis server. Keyword arguments may also be accepted in order to further control how to interpret the response. These keyword arguments are specified during the command’s call to execute_command. The ZRANGE implementation demonstrates the use of response callback keyword arguments with its “withscores” argument.
Redis client instances can safely be shared between threads. Internally, connection instances are only retrieved from the connection pool during command execution, and returned to the pool directly after. Command execution never modifies state on the client instance.
However, there is one caveat: the Redis SELECT command. The SELECT command allows you to switch the database currently in use by the connection. That database remains selected until another is selected or until the connection is closed. This creates an issue in that connections could be returned to the pool that are connected to a different database.
As a result, redis-py does not implement the SELECT command on client instances. If you use multiple Redis databases within the same application, you should create a separate client instance (and possibly a separate connection pool) for each database.
It is not safe to pass PubSub or Pipeline objects between threads.
Pipelines are a subclass of the base Redis class that provide support for buffering multiple commands to the server in a single request. They can be used to dramatically increase the performance of groups of commands by reducing the number of back-and-forth TCP packets between the client and server.
Pipelines are quite simple to use:
>>> r = redis.Redis(...)
>>> r.set('bing', 'baz')
>>> # Use the pipeline() method to create a pipeline instance
>>> pipe = r.pipeline()
>>> # The following SET commands are buffered
>>> pipe.set('foo', 'bar')
>>> pipe.get('bing')
>>> # the EXECUTE call sends all buffered commands to the server, returning
>>> # a list of responses, one for each command.
>>> pipe.execute()
[True, 'baz']
For ease of use, all commands being buffered into the pipeline return the pipeline object itself. Therefore calls can be chained like:
>>> pipe.set('foo', 'bar').sadd('faz', 'baz').incr('auto_number').execute()
[True, True, 6]
In addition, pipelines can also ensure the buffered commands are executed atomically as a group. This happens by default. If you want to disable the atomic nature of a pipeline but still want to buffer commands, you can turn off transactions.
>>> pipe = r.pipeline(transaction=False)
A common issue occurs when requiring atomic transactions but needing to retrieve values in Redis prior for use within the transaction. For instance, let’s assume that the INCR command didn’t exist and we need to build an atomic version of INCR in Python.
The completely naive implementation could GET the value, increment it in Python, and SET the new value back. However, this is not atomic because multiple clients could be doing this at the same time, each getting the same value from GET.
Enter the WATCH command. WATCH provides the ability to monitor one or more keys prior to starting a transaction. If any of those keys change prior the execution of that transaction, the entire transaction will be canceled and a WatchError will be raised. To implement our own client-side INCR command, we could do something like this:
>>> with r.pipeline() as pipe:
... while 1:
... try:
... # put a WATCH on the key that holds our sequence value
... pipe.watch('OUR-SEQUENCE-KEY')
... # after WATCHing, the pipeline is put into immediate execution
... # mode until we tell it to start buffering commands again.
... # this allows us to get the current value of our sequence
... current_value = pipe.get('OUR-SEQUENCE-KEY')
... next_value = int(current_value) + 1
... # now we can put the pipeline back into buffered mode with MULTI
... pipe.multi()
... pipe.set('OUR-SEQUENCE-KEY', next_value)
... # and finally, execute the pipeline (the set command)
... pipe.execute()
... # if a WatchError wasn't raised during execution, everything
... # we just did happened atomically.
... break
... except WatchError:
... # another client must have changed 'OUR-SEQUENCE-KEY' between
... # the time we started WATCHing it and the pipeline's execution.
... # our best bet is to just retry.
... continue
Note that, because the Pipeline must bind to a single connection for the duration of a WATCH, care must be taken to ensure that the connection is returned to the connection pool by calling the reset() method. If the Pipeline is used as a context manager (as in the example above) reset() will be called automatically. Of course you can do this the manual way by explicity calling reset():
>>> pipe = r.pipeline()
>>> while 1:
... try:
... pipe.watch('OUR-SEQUENCE-KEY')
... ...
... pipe.execute()
... break
... except WatchError:
... continue
... finally:
... pipe.reset()
A convenience method named “transaction” exists for handling all the boilerplate of handling and retrying watch errors. It takes a callable that should expect a single parameter, a pipeline object, and any number of keys to be WATCHed. Our client-side INCR command above can be written like this, which is much easier to read:
>>> def client_side_incr(pipe):
... current_value = pipe.get('OUR-SEQUENCE-KEY')
... next_value = int(current_value) + 1
... pipe.multi()
... pipe.set('OUR-SEQUENCE-KEY', next_value)
>>>
>>> r.transaction(client_side_incr, 'OUR-SEQUENCE-KEY')
[True]