Adding agent to monitor external fuel tanks level.

[ntampier]

Description

Motivation and Context

Resolves #568.

How Has This Been Tested?

Types of changes

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)

Checklist:

• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.

[BrianJKoopman]

Thanks for the PR @ntampier! A few comments below. Happy to discuss.

[BrianJKoopman]

This /dev/ttyUSB1 will need to be configurable, as it is not guaranteed to be the same every time the device is plugged in/the computer is rebooted. I'd recommend passing it as an argument when instantiating a TankLevelMonitor object.

[BrianJKoopman]

One overall comment here is that we need a documentation page for the Agent. I described this recently in #638, so I'll link you there: #638 (comment)

[BrianJKoopman]

This self.unit is unused. (It looks like it's leftover from the Agent this one was based on.) Remove, along with the unit argument.

[BrianJKoopman]

Remove the --unit argument.

[BrianJKoopman]

Last place to remove unit.

[BrianJKoopman]

Let's replace this deep nesting with a series of checks. Starting with the assumption that self.tank_data_ok = False, each check should return if the condition means the data is not ok. If all checks pass, then set self.tank_data_ok = True, like this:

    def tank_data_checker(self, tank_num):
        self.tank_data_ok = False

        if len(self.full_data) != 89:
            return

        if self.full_data[18:19] == b'':
            return

        if self.verbose:
            print("IN DATA CHECKER 1: ", self.full_data[0:2])
        if self.full_data[0:2] != b'\x01i':
            return

        if self.verbose:
            print("IN DATA CHECKER 2:", self.full_data[88:89])
        if self.full_data[88:89] != b'\x03':
            return

        if self.verbose:
            print("IN DATA CHECKER 3:", self.full_data[18:19])
        if int(self.full_data[18:19]) != int(tank_num):
            return

        if self.verbose:
            print("DATA IS OK")
        self.tank_data_ok = True

[BrianJKoopman]

I'm a little concerned about this loop, as it potentially goes on forever if the tank data is always bad. How often is the tank data bad? And when it's bad, do you know why?

I would rather not loop here, do the write/read once, check the data quality, and raise an exception if it fails. The main acq Process is already looping to repeat the read. The exception can be caught there and a warning printed about the data being bad, and you can move on to the next loop in acq, while possibly trying to fix the issue causing the data read to fail.

[ntampier]

The problem is that tank data ib bad very often, so it breaks my code after some minutes, data is bad like once per 5 minutes.

[BrianJKoopman]

Gotcha, yeah, I'm not proposing we allow it to break your code, just that it gets caught in the loop at the Agent level so we can tell when it's bad.

[BrianJKoopman]

Similar to above, this line 94 isn't needed and can be removed.

[BrianJKoopman]

A verbosity flag works for debugging, but I'll make the same optional suggestion as I made in #640 -- an alternative approach would be to use debug logging. This would allow you to drop the tracking of verbose/verbosity and instead use:

self.log.debug('debug statement')

The advantage here is that the LOGLEVEL environment variable can be changed without modification of this verbosity constant in the code.

If you want to implement this check out the Logging documentation. (@felipecarrero also just implemented this, you can ask him as well.) I wouldn't let it hold up the PR though.