A library for parsing Dutch Smart Meter Requirements (DSMR) telegram data. It also includes client implementation to directly read and parse smart meter data.
DSMR Parser supports DSMR versions 2, 3, 4 and 5. See for the currently supported/tested Python versions here.
Serial client
Read the serial port and work with the parsed telegrams. It should be run in a separate process because the code is blocking (not asynchronous):
from dsmr_parser import telegram_specifications
from dsmr_parser.clients import SerialReader, SERIAL_SETTINGS_V4
serial_reader = SerialReader(
device='/dev/ttyUSB0',
serial_settings=SERIAL_SETTINGS_V4,
telegram_specification=telegram_specifications.V4
)
for telegram in serial_reader.read():
print(telegram) # see 'Telegram object' docs belowSocket client
Read a remote serial port (for example using ser2net) and work with the parsed telegrams. It should be run in a separate process because the code is blocking (not asynchronous):
from dsmr_parser import telegram_specifications
from dsmr_parser.clients import SocketReader
socket_reader = SocketReader(
host='127.0.0.1',
port=2001,
telegram_specification=telegram_specifications.V4
)
for telegram in socket_reader.read():
print(telegram) # see 'Telegram object' docs belowAsyncIO client
For a test run using a tcp server (lasting 20 seconds) use the following example:
import asyncio
import logging
from dsmr_parser import obis_references
from dsmr_parser.clients.protocol import create_dsmr_reader, create_tcp_dsmr_reader
logging.basicConfig(level=logging.INFO, format='%(message)s')
HOST = MY_HOST
PORT = MY_PORT
DSMR_VERSION = MY_DSMR_VERSION
logger = logging.getLogger('tcpclient')
logger.debug("Logger created")
def printTelegram(telegram):
logger.info(telegram)
async def main():
try:
logger.debug("Getting loop")
loop = asyncio.get_event_loop()
logger.debug("Creating reader")
await create_tcp_dsmr_reader(
HOST,
PORT,
DSMR_VERSION,
printTelegram,
loop
)
logger.debug("Reader created going to sleep now")
await asyncio.sleep(20)
logger.info('Finished run')
except Exception as e:
logger.error("Unexpected error: "+ e)
asyncio.run(main())Note the creation of a callback function to call when a telegram is received. In this case printTelegram. Normally the used loop is the one running.
Currently the asyncio implementation does not support returning telegram objects directly as a read_as_object() for async tcp is currently not implemented. Moreover, the telegram passed to telegram_callback(telegram) is already parsed. Therefore we can't feed it into the telegram constructor directly as that expects unparsed telegrams
However, if we construct a mock TelegramParser that just returns the already parsed object we can work around this. An example is below:
import asyncio
import logging
from dsmr_parser import telegram_specifications
from dsmr_parser.clients.protocol import create_tcp_dsmr_reader
logging.basicConfig(level=logging.INFO, format='%(message)s')
HOST = MY_HOST
PORT = MY_PORT
DSMR_VERSION = MY_DSMR_VERSION
logger = logging.getLogger('tcpclient')
logger.debug("Logger created")
class mockTelegramParser(object):
def parse(self, telegram):
return telegram
telegram_parser = mockTelegramParser()
def printTelegram(telegram):
try:
logger.info(Telegram(telegram, telegram_parser, telegram_specifications.V4))
except InvalidChecksumError as e:
logger.warning(str(e))
except ParseError as e:
logger.error('Failed to parse telegram: %s', e)
async def main():
try:
logger.debug("Getting loop")
loop = asyncio.get_event_loop()
logger.debug("Creating reader")
await create_tcp_dsmr_reader(
HOST,
PORT,
DSMR_VERSION,
printTelegram,
loop
)
logger.debug("Reader created going to sleep now")
while True:
await asyncio.sleep(1)
except Exception as e:
logger.error("Unexpected error: "+ e)
raise
if __name__ == '__main__':
try:
asyncio.run(main())
except (KeyboardInterrupt, SystemExit):
logger.info('Closing down...')
except Exception as e:
logger.error("Unexpected error: "+ e)The parsing module accepts complete unaltered telegram strings and parses these into a Telegram object.
Tip: getting full telegrams from a bytestream can be made easier by using the TelegramBuffer helper class.
from dsmr_parser import telegram_specifications
from dsmr_parser.parsers import TelegramParser
# String is formatted in separate lines for readability.
telegram_str = (
'/ISk5\\2MT382-1000\r\n'
'\r\n'
'0-0:96.1.1(4B384547303034303436333935353037)\r\n'
'1-0:1.8.1(12345.678*kWh)\r\n'
'1-0:1.8.2(12345.678*kWh)\r\n'
'1-0:2.8.1(12345.678*kWh)\r\n'
'1-0:2.8.2(12345.678*kWh)\r\n'
'0-0:96.14.0(0002)\r\n'
'1-0:1.7.0(001.19*kW)\r\n'
'1-0:2.7.0(000.00*kW)\r\n'
'0-0:17.0.0(016*A)\r\n'
'0-0:96.3.10(1)\r\n'
'0-0:96.13.1(303132333435363738)\r\n'
'0-0:96.13.0(303132333435363738393A3B3C3D3E3F303132333435363738393A3B3C3D3E'
'3F303132333435363738393A3B3C3D3E3F303132333435363738393A3B3C3D3E3F30313233'
'3435363738393A3B3C3D3E3F)\r\n'
'0-1:96.1.0(3232323241424344313233343536373839)\r\n'
'0-1:24.1.0(03)\r\n'
'0-1:24.3.0(090212160000)(00)(60)(1)(0-1:24.2.1)(m3)\r\n'
'(00001.001)\r\n'
'0-1:24.4.0(1)\r\n'
'!\r\n'
)
parser = TelegramParser(telegram_specifications.V3)
# see 'Telegram object' docs below
telegram = parser.parse(telegram_str)A Telegram has attributes for all the parsed values according to the given telegram specification. Each value is a DsmrObject which have a 'value' and 'unit' property. MBusObject's, which are DsmrObject's as well additionally have a 'datetime' property. The 'value' can contain any python type (int, str, Decimal) depending on the field. The 'unit' contains 'kW', 'A', 'kWh' or 'm3'.
Note: Telegram extends dictionary, which done for backwards compatibility. The use of keys (e.g. telegram[obis_references.CURRENT_ELECTRICITY_USAGE]) is deprecated.
Below are some examples on how to get the meter data. Alternatively check out the following unit test for a complete example: TelegramParserV5Test.test_parse
# Print contents of all available values
# See dsmr_parser.obis_name_mapping for all readable telegram values.
# The available values differ per DSMR version and meter.
print(telegram)
# P1_MESSAGE_HEADER: 42 [None]
# P1_MESSAGE_TIMESTAMP: 2016-11-13 19:57:57+00:00 [None]
# EQUIPMENT_IDENTIFIER: 3960221976967177082151037881335713 [None]
# ELECTRICITY_USED_TARIFF_1: 1581.123 [kWh]
# etc.
# Example to get current electricity usage
print(telegram.CURRENT_ELECTRICITY_USAGE) # <dsmr_parser.objects.CosemObject at 0x7f5e98ae5ac8>
print(telegram.CURRENT_ELECTRICITY_USAGE.value) # Decimal('2.027')
print(telegram.CURRENT_ELECTRICITY_USAGE.unit) # 'kW'
# All Mbus device readings like gas meters and water meters can be retrieved as follows. This
# returns a list of MbusDevice objects:
mbus_devices = telegram.MBUS_DEVICES
# A specific MbusDevice based on the channel it's connected to, can be retrieved as follows:
mbus_device = telegram.get_mbus_device_by_channel(1)
print(mbus_device.DEVICE_TYPE.value) # 3
print(mbus_device.EQUIPMENT_IDENTIFIER_GAS.value) # '4730303339303031393336393930363139'
print(mbus_device.HOURLY_GAS_METER_READING.value) # Decimal('246.138')
# DEPRECATED: the dictionary approach of getting the values by key or `.items()' or '.get() is deprecated
telegram[obis_references.CURRENT_ELECTRICITY_USAGE]The telegram object has an iterator, can be used to find all the information elements in the current telegram:
[attr for attr, value in telegram]
Out[11]:
['P1_MESSAGE_HEADER',
'P1_MESSAGE_TIMESTAMP',
'EQUIPMENT_IDENTIFIER',
'ELECTRICITY_USED_TARIFF_1',
'ELECTRICITY_USED_TARIFF_2',
'ELECTRICITY_DELIVERED_TARIFF_1',
'ELECTRICITY_DELIVERED_TARIFF_2',
'ELECTRICITY_ACTIVE_TARIFF',
'CURRENT_ELECTRICITY_USAGE',
'CURRENT_ELECTRICITY_DELIVERY',
'LONG_POWER_FAILURE_COUNT',
'VOLTAGE_SAG_L1_COUNT',
'VOLTAGE_SAG_L2_COUNT',
'VOLTAGE_SAG_L3_COUNT',
'VOLTAGE_SWELL_L1_COUNT',
'VOLTAGE_SWELL_L2_COUNT',
'VOLTAGE_SWELL_L3_COUNT',
'TEXT_MESSAGE_CODE',
'TEXT_MESSAGE',
'DEVICE_TYPE',
'INSTANTANEOUS_ACTIVE_POWER_L1_POSITIVE',
'INSTANTANEOUS_ACTIVE_POWER_L2_POSITIVE',
'INSTANTANEOUS_ACTIVE_POWER_L3_POSITIVE',
'INSTANTANEOUS_ACTIVE_POWER_L1_NEGATIVE',
'INSTANTANEOUS_ACTIVE_POWER_L2_NEGATIVE',
'INSTANTANEOUS_ACTIVE_POWER_L3_NEGATIVE',
'EQUIPMENT_IDENTIFIER_GAS',
'HOURLY_GAS_METER_READING']To install DSMR Parser:
$ pip install dsmr-parserCreate a virtualenv and activate it followed by the installation of the dsmr-parser:
python3 -m venv venv
source venv/bin/activate
pip install -e .Install tox and run it:
pip install tox
toxYou should see that the tests have succeeded:
======================================================================================================== 59 passed in 0.91s ========================================================================================================
py: commands[1]> pylama dsmr_parser test
py: OK (11.55=setup[9.73]+cmd[1.29,0.53] seconds)
congratulations :) (11.69 seconds)
Now you can make changes by editing the code and rerunning tox to verify your changes.
If the serial settings SERIAL_SETTINGS_V2_2 or SERIAL_SETTINGS_V4 don't work. Make sure to try and replace the parity settings to EVEN or NONE. It's possible that alternative settings will be added in the future if these settings don't work for the majority of meters.
