[FW-146] Deserialize packets received from DatagramSocket #22
Conversation
| self._recv_thread = threading.Thread(target=self._receive, daemon=True) | ||
| self._recv_thread.start() | ||
|
|
||
| def _receive(self): |
There was a problem hiding this comment.
why did you change DatagramSocket.py
There was a problem hiding this comment.
This is because we want DatagramSocket receive method to block on receive
|
|
||
| ds = DatagramSocket('127.0.0.1', 8800, '127.0.0.1', 8801) | ||
|
|
||
| package_data = [(100, "temp:float32,is_active:bool,state:enum(idle-fault)"), (101, "preasure:int32")] #I assume that it's the format, it might change |
There was a problem hiding this comment.
I'm not sure, however Juan told me that the package will be a dictionary where the key is the id and the values are only the data_types like this:
1: {"uint16","float32","enum(OFF,ON)"}
| if 'enum' in type: | ||
| value_data = struct.unpack('<B', buffer[:1])[0] | ||
| buffer = buffer[1:] | ||
| valores = (re.search(r'\((.*?)\)', type)).group(1).split('-') |
There was a problem hiding this comment.
wow !what a way to do it, I like it
| self.dict_measurement_types = {} #key string id of each measurament, contains an array from types | ||
| self.dict_measurement_names = {} #contains array of names of each variable | ||
| self.dict_measurement_value = {} #contains the value of each measurement |
There was a problem hiding this comment.
| self.dict_measurement_types = {} #key string id of each measurament, contains an array from types | |
| self.dict_measurement_names = {} #contains array of names of each variable | |
| self.dict_measurement_value = {} #contains the value of each measurement | |
| self.measurement_types = {} #key string id of each measurament, contains an array from types | |
| self.measurement_names = [] #contains array of names of each variable | |
| self.measurement_values = {} #contains the value of each measurement |
| if bytes_count == bytes_size: | ||
| self.deserialize(buff, id_packet) | ||
|
|
||
| buff.clear() | ||
| bytes_count = 0 | ||
| bytes_size = 0 | ||
| wait_new_packet = True |
There was a problem hiding this comment.
What happens if the bytes_count is greater than the bytes_size? This part looses packets if you have more than one in the buffer. You should add the bytes to the buffer and then do in a loop:
- Get the ID and the packet size
- If the buffer has enough bytes to deserialize the packet do that
- Remove the processed data from the buffer
- Wait for more data
| def calculate_byte_size(self, measurements): | ||
|
|
||
| size = 0 | ||
| for measure in measurements: | ||
| if measure == 'uint8' or measure == 'int8' or measure == 'bool' or 'enum' in measure: | ||
| size += 1 | ||
| elif measure == 'uint16' or measure == 'int16': | ||
| size += 2 | ||
| elif measure == 'uint32' or measure == 'int32' or measure == 'float32': | ||
| size += 4 | ||
| elif measure == 'uint64' or measure == 'int64' or measure == 'float64': | ||
| size += 8 | ||
|
|
||
| return size |
There was a problem hiding this comment.
precalculate this in the constructor, the decoding function will be called a lot of times and we want to reduce the time it takes to complete
There was a problem hiding this comment.
put this outside the vmcu package, because we don't want to import it later on, should be on src/test_decoder/...
There was a problem hiding this comment.
put this outside the vmcu package, because we don't want to import it later on, should be on src/test_decoder/...
Class decoder uses DatagramSocket and take packet UDP
includes a thread that checks packets continuously
Class Datagramsocket thread has been removed and only interact with the Decoder thread
I asume a dictionary model that could be different from the final model