venkatachalam...@gmail.com writes: > ... > I am writing a python code for processing a data obtained from a sensor. The > data from sensor is obtained by executing a python script. The data obtained > should be further given to another python module where the received data is > used for adjusting the location of an object. > > For achieving this, there is a central bash script, which runs both the > python modules parallel. Something like: > > python a.py & > python b.py & > > I am trying to return the sensor data to the bash .sh file, therefore it can > be provided to the other script.
I would recommend, do not do it this way. On an abstract level, I recommend to use a communication channel between your "a" and your "b". "a" writes to it and "b" reads from it. There are many, many options for such a communication channel. One of the easiest would be to realisize "a" and "b" as tasks in a single Python process which use a queue as communication channel. The big advantage: "a" and "b" can communicate directly via python objects. Another option would be to implement the communication channel via an external queue or file; in those cases, the Python object created by "a" would need to be serialized (to be put into the communication channel) and derialized (i.e. recreated) again by "b". -- https://mail.python.org/mailman/listinfo/python-list
