simple-tem

• Remote control of PyJEM using ZeroMQ REQ/REP
• Synchronous execution for all commands
• Supports run_async = True for SetTiltXAngle
• might be turned into a conda pkg once stable
• PyJEM API: https://pyjem.github.io/PyJEM/interface/PyJEM.TEM3.html

Usage

Server

Copy tem-server.py to the machine where PyJEM is installed

Client Make sure the simple-tem folder is added to PYTHONPATH (or conda pkg installed)

from simple_tem import TEMClient
c = TEMClient("temserver", 3535)

c.ping() #check connection
c.stage_position #read position
c.SetILs(21500, 24500)


c.SetTiltXAngle(35) 

...

# Check if we can talk to the sever before issuing a command
if c.is_alive:
    c.Sec.SetTiltXAngle(40)
else:
    raise SomeKindOfException


#Waiting for a rotation to finish
c.SetTiltXAngle(10)
c.wait_until_rotate_starts() #default timeout of 2s
while c.is_rotating:
    # some checks? 

#stage is now stopped verify angle 
a = c.GetTiltXAngle()

Error handling

#Check, with a 1s timeout, if the server is responding -> True/False
>>> c.is_alive 
True

#Normal calls have a timeout of 5s and then raises a TimeoutError
>>> c.GetStagePosition()
TimeoutError: Timeout while waiting for reply from sjsjs:3535

#Each call to PyJEM could fail
>>> c.Getf1OverRateTxNum()
RuntimeError: Function execution error. function: Stage3::Getf1OverRateTxNum, code : 20483

In some functions three automatic retries with a 0.1s sleep is implemented

• c.is_rotating

But mostly the caller is responsible to handle the exceptions.

Tests

Start a local redis instance on port 5454

This is used to hold values that we pretend to send to the microscope

redis-server --port 5454

Using Python 3.5 Launch the tem server in dummy mode This is becaues the PyJEM env on the TEM PC is still Python 3.5. Of course if you want to test functtionallity without verifying python 3.5 compatibility you can use a later version of python.

python tem-server.py -d

From your normal env run the tests

python -m pytest

Implementation

ZeroMQ REQ/REP socket sending a request as a multi part message

[command_name : string][arguments : json_encoded_string]

Reply as

[status_code : string][return_value : json_encoded_string]