sllurp is a Python library to interface with RFID readers. It is a pure-Python implementation of the Low Level Reader Protocol (LLRP).
These readers are known to work well with sllurp, but it should be adaptable with not much effort to other LLRP-compatible readers:
- Impinj Speedway (R1000)
- Impinj Speedway Revolution (R220, R420)
- Impinj Speedway xPortal
- Motorola MC9190-Z (handheld)
- Kathrein ARU2400
- Zebra FX9600
File an issue on GitHub if you would like help getting another kind of reader to work.
sllurp is distributed under version 3 of the GNU General Public License. See
LICENSE.txt
for details.
This branch modifies the package heaviliy to make it suitable for fast re-configurations of the reader, i.e. connecting, changing a rospec parameter, inventoring, changing a rospec parameter, inventoring, ..., disconnecting.
As the twisted socket only allows one connection after import, we had to implement an own transport class using the standard socket of python. There is a new module reader where specific reader classes can inherit from the LLRPClient. We also modified the llrp, llrp_proto and llrp_decoder modules to handle impinj specific extensions we need. Also, llrp is now completely rewritten to be state-less and does not depend on twisted. That way, the code is much cleaner and there is no hassle with chaining deferreds.
from sllurp.reader import R420
reader = R420('192.168.4.2')
freqs = reader.freq_table
powers = reader.power_table
tags = reader.detectTags(powerDBm=powers[-1], freqMHz=freqs[0],
mode=1002, session=2, population=1, duration=0.5, searchmode=2)
for tag in tags:
print(tag)
Example code for changing Tags EPC
from sllurp.reader import R420
#import logging
#logging.basicConfig(filename='log.txt', level=logging.DEBUG)
reader = R420('192.168.4.2') # connect to reader
# setup access spec
epcLen = 12 # total number of bytes
epcRawStart = b'\x12\x34\x56\x78' # let the raw EPC URI start with these bytes
epcRawUri = epcRawStart+b'\x00'*(epcLen-len(epcRawStart)) # fill up with zeros
# note: 1 Word = 2 Bytes
writeSpecParam = {
'OpSpecID': 0,
'MB': 1,
'WordPtr': 2,
'AccessPassword': 0,
'WriteDataWordCount': len(epcRawUri)//2,
'WriteData': epcRawUri,
}
reader.startAccess(writeWords=writeSpecParam, opCount=0) # set opCount to 1 to stop after 1 write operation
# Actually adds and enables an access spec.
# It is executed with the next inventory round (reader.detectTags())
'''
readSpecParam = {
'OpSpecID': 0,
'MB': 1,
'WordPtr': 0,
'AccessPassword': 0,
'WordCount': 8
}
reader.startAccess(readWords=readSpecParam)
'''
print('Before changing:')
tags = reader.detectTags(powerDBm=16, antennas=(1,)) # remove antennas argument or set to (0,) to use all antenna ports
for tag in tags:
print(tag)
# At this point, the access spec is deleted
# Normal inventory
print('After changing:')
tags = reader.detectTags(powerDBm=16, antennas=(1,))
for tag in tags:
print(tag)
import logging
logging.basicConfig(filename='llrp.log', level=logging.DEBUG)
Currently, the GUI supports only the classes defined in reader.
from sllurp.gui import InventoryApp
app = InventoryApp() # start the GUI
The GUI can also be started with python -m sllurp.gui
Reader mode selection is confusing, not least because the LLRP specification
traditionally conflated ModeIndex
and ModeIdentifier
. If you're using
sllurp inventory
, you probably want to use --mode-identifier N
instead
of -mode-index
. Check your reader's manual to see what mode identifiers it
supports via the C1G2RFControl
parameter, or run sllurp --debug
inventory
against a reader to see a dump of the supported modes in the
capabilities description.
Want to contribute? Here are some areas that need improvement:
- Reduce redundancy in the
encode_*
anddecode_*
functions inllrp_proto.py
. - Support the AccessSpec primitive (basis for tag read and write).
- Write tests for common encoding and decoding tasks.
- Make
get_reader_config
use thefabric
library to connect to readers via SSH. - Generalize LLRP support beyond Impinj readers. Remove Impinj-specific assumptions.
Much of the code in sllurp is by Ben Ransford, although it began its life in August 2013 as a fork of LLRPyC. Many fine citizens of GitHub have contributed code to sllurp since the fork.