Closed g7gpr closed 3 years ago
Thanks - it does exactly what I need. The only thing I don't like is that it writes a log file with the data pulled from the log file, but it does at least get the log_time and time correct. I can't think of an elegant way to avoid this.
How about adding an optional argument to https://github.com/projecthorus/chasemapper/blob/master/horusmapper.py#L185 e.g. make it def handle_new_payload_position(data, log_position=True):
Then at the end of the function, where the position is logged ( https://github.com/projecthorus/chasemapper/blob/master/horusmapper.py#L273 ) make that if chase_logger and log_position
Then when you push data in from the log file, you can just set the log_position argument to False. That'll stop the data ending up in the log file.
Yes, that's was my first thought, but I'm not convinced it is the best way.
It might not be the best way, but it's the obvious solution, which is probably good enough for this situation.
Agreed. I'll implement like that.
Implemented, but it will need testing this evening.
There's a new function in auto_Rx (you may need to run a git pull) where you can 'play back' one of the auto_rx sonde log files.
Go into the radiosonde_auto_rx/auto_rx directory and run:
python3 -m autorx.emulation log/path_to_log_file.log
This will play back the log file in 'realtime' out the default UDP port of 55673. The timestamps are replaced with the current time (incrementing as appropriate).
I'll have a play with that, unfortunately out of time on this for now.
Failed testing. I think I know what the problem it.
Tested on live data from S3551348.
The predictor algorithm is running for the historical data and producing garbage as there is no vertical movement, so I've blocked the predictor running when there is 1 or fewer points in the payload track.
I think it's done.
Will have a test when I get a chance (probably Friday)
Looking good! I'll give it a test sometime today.