Library for interacting with ElkM1 alarm/automation panel.
- Python 3.5 (or higher)
This package is created as a library to interact with an ElkM1 alarm/automation pattern. The motivation to write this was to use with the Home Assistant automation platform. The library can be used for writing other ElkM1 integration applications. The IO with the panel is asynchronous over TCP or over the serial port.
$ pip install elkm1_libBasic connection to the Elk panel:
from elkm1_lib import Elk
import logging
# Print to STDOUT
LOG = logging.getLogger(__name__)
logging.basicConfig(level=logging.DEBUG, format='%(message)s')
# Connect to elk
elk = Elk({'url': 'elk://192.168.1.100'})
elk.connect()
elk.run()The above will connect to the Elk panel at IP address 192.168.1.100. the elk://
prefix specifies that the connect is plaintext. Alternatively, elks:// will
connect over TLS. In this case a userid and password must be specified
and the call to Elk changes to:
elk = Elk(
{'url': 'elks://192.168.1.100', 'userid': 'test', 'password': 'pass'}
)To see working example code take a look at the script bin/simple.
The Elk object supports the concept of Elements. An Element
is the base class representation of Zones, Lights, etc. So, for
example there is a list of zones: elk.zones and each zone can be
accessed by elk.zones[index]. Each element has a __str__
representation so that it is easy to print its contents.
All Elements are referenced starting at 0. Even though the Elk panel
refers to, for example, zones 1-208, the library references them
as zones 0-207. All translation from base 0 to 1 and vice-versa is
handled internally in the elkm1_lib.message module.
After creating the Elk object and connecting to the panel the
library code will synchronize all the elements to the data from the Elk panel.
Many Elk messages are handled by the library, caching their contents. When a
message causes a change to an attribute of an Element, registered
callbacks are called so that user use of the library can be notified
of changing elements. The following user code shows registering a callback:
def call_me(element, changeset):
print(changeset)
for zone_number in range(Max.ZONES.value):
elk.zones[zone_number].add_callback(call_me)The library encodes, decodes, and processes messages to/from the
Elk panel. All the encoding and decoding is done in elkm1_lib.message module.
Messages received are handled with callbacks. The library
internally registers callbacks so that decoded messages
can be used to update an Element. The user of the
library may also register callbacks. Any particular message
may have multiple callbacks.
When the message is received it is decoded and some validation is done. The message handler is called with the fields of from the decoded message. Each type of message has parameters that match the message type. All handler parameters are named parameters.
Here is an example of a message handler being registered and how it is called:
def zone_status_change_handler(zone_number, zone_status):
print(zone_number, zone_status)
elk.add_handler('ZC', zone_status_change_handler)The above code registers a callback for 'ZC' (Elk zone status change) messages. When a ZC message is received the handler functions are called with the zone_number and zone_status.
The bin directory of the library has one utility program and
a couple of example uses of the library.
The utility mkdoc creates a Markdown table of the list of Elk
messages with a check mark for those messages have encoders/decoders
and an X for those messages are not planned to be implemented.
There are no parameters to mkdoc. It outputs to stdout.
The data for the report comes from the ElkM1 library code mostly.
A couple of things are hard coded in the mkdoc script, notably
the "no plans to implement" list.
The simple Python script is a trivial use of the ElkM1 library.
It connects to the panel, syncs to internal memory, and continues
listening for any messages from the panel. The URL of the ElkM1 to
connect to is retrieved from an environment variable named ELKM1_URL.
The elk Python script is a bit of a command interpretor. It can run in
two modes. Non-interactive mode is the default. Just run the elk command.
The non-interactive mode is similar to simple except there are a
couple of message handlers (timeout and unknown handlers).
The elk can also be run in interactive mode by invoking it by
elk -i. In this mode is uses curses (full screen use of the terminal)
that has a command line and an output window. TAB switches between
the command line and output windows. In the output window the arrow keys
and scrollwheel scroll the contents of the window.
In the command line when running elk -i there are a
number of commands. Start with help. Then help <command> for
details on each command. In general there are commands to dump the internal
state of elements and to invoke any of the encoders to send a message
to the Elk panel.
For example, light <4, 8, 12-14 would invoke the __str__ method
for the light element to print the cached info for lights 0-3, 8, and 12-14.
Another example would be pf 3 which issues the pf (Turn light off)
command for light number 3 (light 4 on the panel -- remember 0
versus 1 base).
All of the commands that send messages to the panel are automatically
discovered and are all the XX_encode functions in the elkm1_lib.message
module. The docstring and the XX_encode's parameters are shown as part
of the help.
This project uses poetry for development dependencies. Installation instructions are on their website.
To get started developing:
git clone https://github.com/gwww/elkm1.git
cd elkm1
poetry install
poetry shell # Or activate the created virtual environment
make test # to ensure everything installed properly
There is a Makefile in the root directory as well. The make command
followed by one of the targets in the Makefile can be used. If you don't
have or wish to use make the Makefile serves as examples of common
commands that can be run.