Incorrect remaining length is calculated when topic contains utf-8-3 or utf-8-4 characters in SUBSCRIBE/PUBLISH packets

[Eason010212]

Current version of adafruit_minimqtt.py uses len() method to calculate a packet's remaining length, for example：

Adafruit_CircuitPython_MiniMQTT/adafruit_minimqtt/adafruit_minimqtt.py

Line 671 in c55a484

packet_length += sum(len(topic) for topic, qos in topics)

Adafruit_CircuitPython_MiniMQTT/adafruit_minimqtt/adafruit_minimqtt.py

Line 592 in c55a484

remaining_length = 2 + len(msg) + len(topic)

As shown above, strings are measured instead of UTF-8 bytes. When a topic only contains English characters, numbers and punctuation marks, the calculation result is correct; However, when it contains some utf-8-3 or utf-8-4 characters (such as Chinese characters) , an incorrect calculation result of remaining length will be obtained. For example, the actual encoding length of "中文" should be 6, while the length calculated by len() is 2. As a result, such packets will be rejected by any MQTT server.

In MQTT protocol, there is no restriction on utf-8-3 or utf-8-4 characters, so I think this issue is probably a bug that needs to be fixed.

At present, in the application I developed, I rewrite the len() method to temporarily solve this problem, but I still hope that the developers of this library can make official bug fixes.

Thank you very much.

[brentru]

@Eason010212 Hi - are you interested in submitting this as a pull request? I'll review it.

[Eason010212]

@Eason010212 Hi - are you interested in submitting this as a pull request? I'll review it.
OK

[FoamyGuy]

closed by #91