pymemcache-client

Extension of Python package pymemcache providing client configuration.

Badges

docs
info
tests
package
other

Install

pip install pymemcache-client

Install Memcached

To run examples and tests locally upon "localhost", memcached must be installed.

Install memcached using brew

$ brew install memcached

==> Installing memcached
==> Downloading https://homebrew.bintray.com/bottles/memcached-1.5.3.sierra.bottle.tar.gz
######################################################################## 100.0%
==> Pouring memcached-1.5.3.sierra.bottle.tar.gz
==> Caveats
To have launchd start memcached now and restart at login:
  brew services start memcached
Or, if you don't want/need a background service you can just run:
  /usr/local/opt/memcached/bin/memcached
==> Summary
🍺  /usr/local/Cellar/memcached/1.5.3: 11 files, 198.7KB

Start memcached using brew

$ brew services start memcached

==> Successfully started `memcached` (label: homebrew.mxcl.memcached)

Architecture

Extension of Python package pymemcache returning from pymemcache.client.hash import HashClient and configured using either ~/.pymemcache.json or pymemcache.json.

Functionality

class PymemcacheClient: Prepare a Pymemcache Client using available configuration.

Configuration can be provided when creating an instance of this class by:

• User-level file: ~/.pymemcache.json
• Application-level file: ./pymemcache.json
• Provided as file path value in JSON-format to parameter config_file
• Provided as dict value in JSON-format to parameter config

Configuration

This is the expected configuration that can be provided to class PymemcacheClient is in JSON format:

{
    "client_type": "[OPTIONAL str, DEFAULT 'basic']",
    "servers": [
        {
            "host": "[str]",
            "port": [int]
        }
    ],
    "server_type": "[OPTIONAL str, DEFAULT 'standard']",
    "connect_timeout": [OPTIONAL float, Defaults to forever],
    "timeout": [OPTIONAL float, Defaults to forever],
    "no_delay": [OPTIONAL bool, DEFAULT false],
    "ignore_exc": [OPTIONAL bool, DEFAULT false],
    "default_noreply": [OPTIONAL bool, DEFAULT true],
    "allow_unicode_keys": [OPTIONAL bool, DEFAULT false]
}

• "client_type": [OPTIONAL] Type of memcached client, DEFAULT "basic"

  • "basic": Memcached client using from pymemcache.client.base import Client. A client for a single memcached server.
  • "hash": Memcached client using from pymemcache.client.hash import HashClient. A client for communicating with a cluster of memcached servers.

• "servers": A list of memcached servers and requiring at least one.

  • if "client_type" is "basic", then 1 server and only 1 is required.
  • if "client_type" is "hash", then 1 or more servers is required.

• "server_type": Defining special needs for connect to memcached servers, DEFAULT "standard".

  • "standard": Pooling server connections using standard HTTP.
  • "elasticache": Pooling server connections using Auto Discovery through AWS Elasticache.

• "connect_timeout": OPTIONAL float, seconds to wait for a connection to the memcached server. Defaults to "forever" (uses the underlying default socket timeout, which can be very long), i.e., the socket is put in blocking mode waiting to connect to memcached server.
• "timeout": OPTIONAL float, OPTIONAL float, seconds to wait for send or recv calls on the socket connected to memcached. Defaults to "forever" (uses the underlying default socket timeout, which can be very long), i.e., the socket is put in blocking mode.
• "no_delay": OPTIONAL bool, set the TCP_NODELAY flag, which may help with performance in some cases. DEFAULT false.
• "ignore_exc": OPTIONAL bool, True to cause the "get", "gets", "get_many" and "gets_many" calls to treat any errors as cache misses. DEFAULT false.
• "default_noreply": OPTIONAL bool, the DEFAULT value for 'noreply' as passed to store commands. DEFAULT true.
• "allow_unicode_keys": OPTIONAL bool, support unicode (utf8) keys. DEFAULT false.

Dependencies

pymemcache-client module is built upon Python 3 and has dependencies upon several Python modules available within Python Package Index PyPI.

• pymemcache
• telnetlib3
• ujson