Python extension that wraps protocol parsing code in hiredis. It primarily speeds up parsing of multi bulk replies.

hiredis-py is available on PyPi, and can be installed with:

pip install hiredis

hiredis-py requires Python 2.6 or higher.

Make sure Python development headers are available when installing hiredis-py. On Ubuntu/Debian systems, install them with apt-get install python-dev for Python 2 or apt-get install python3-dev for Python 3.

The hiredis module contains the Reader class. This class is responsible for parsing replies from the stream of data that is read from a Redis connection. It does not contain functionality to handle I/O.

The Reader class has two methods that are used when parsing replies from a stream of data. Reader.feed takes a string argument that is appended to the internal buffer. Reader.gets reads this buffer and returns a reply when the buffer contains a full reply. If a single call to feed contains multiple replies, gets should be called multiple times to extract all replies.

Example:

>>> reader = hiredis.Reader()
>>> reader.feed("$5\r\nhello\r\n")
>>> reader.gets()
'hello'

When the buffer does not contain a full reply, gets returns False. This means extra data is needed and feed should be called again before calling gets again:

>>> reader.feed("*2\r\n$5\r\nhello\r\n")
>>> reader.gets()
False
>>> reader.feed("$5\r\nworld\r\n")
>>> reader.gets()
['hello', 'world']

hiredis.Reader is able to decode bulk data to any encoding Python supports. To do so, specify the encoding you want to use for decoding replies when initializing it:

>>> reader = hiredis.Reader(encoding="utf-8")
>>> reader.feed("$3\r\n\xe2\x98\x83\r\n")
>>> reader.gets()
u'☃'

When bulk data in a reply could not be properly decoded using the specified encoding, it will be returned as a plain string. When the encoding cannot be found, a LookupError will be raised after calling gets for the first reply with bulk data (identical to what Python's unicode method would do).

When a protocol error occurs (because of multiple threads using the same socket, or some other condition that causes a corrupt stream), the error hiredis.ProtocolError is raised. Because the buffer is read in a lazy fashion, it will only be raised when gets is called and the first reply in the buffer contains an error. There is no way to recover from a faulty protocol state, so when this happens, the I/O code feeding data to Reader should probably reconnect.

Redis can reply with error replies (-ERR ...). For these replies, the custom error class hiredis.ReplyError is returned, but not raised.

When other error types should be used (so existing code doesn't have to change its except clauses), Reader can be initialized with the protocolError and replyError keywords. These keywords should contain a class that is a subclass of Exception. When not provided, Reader will use the default error types.

The repository contains a benchmarking script in the benchmark directory, which uses gevent to have non-blocking I/O and redis-py to handle connections. These benchmarks are done with a patched version of redis-py that uses hiredis-py when it is available.

All benchmarks are done with 10 concurrent connections.

• SET key value + GET key
  • redis-py: 11.76 Kops
  • redis-py with hiredis-py: 13.40 Kops
  • improvement: 1.1x

List entries in the following tests are 5 bytes.

• LRANGE list 0 9:
  • redis-py: 4.78 Kops
  • redis-py with hiredis-py: 12.94 Kops
  • improvement: 2.7x
• LRANGE list 0 99:
  • redis-py: 0.73 Kops
  • redis-py with hiredis-py: 11.90 Kops
  • improvement: 16.3x
• LRANGE list 0 999:
  • redis-py: 0.07 Kops
  • redis-py with hiredis-py: 5.83 Kops
  • improvement: 83.2x

Throughput improvement for simple SET/GET is minimal, but the larger multi bulk replies get, the larger the performance improvement is.

This code is released under the BSD license, after the license of hiredis.