Added GEA2 Communications #4
Conversation
README.md
Outdated
| @@ -10,6 +10,9 @@ Provides a simple interface for sending and receiving GEA3 serial packets. | |||
| ### `tiny_erd_client` | |||
| Provides a simple interface for reading and writing addressable data (ERDs) over a GEA3 serial interface. | |||
|
|
|||
| ## `tiny_gea2_interface_single_wire` | |||
There was a problem hiding this comment.
| ## `tiny_gea2_interface_single_wire` | |
| ### `tiny_gea3_interface_single_wire` |
Can you put this up by the other interface?
|
The filename for the new component are inconsistent, can you get them all to match? |
|
|
||
| static void byte_received(void* context, const void* _args) | ||
| { | ||
| reinterpret(instance, context, self_t*); |
There was a problem hiding this comment.
So I remember talking to Andrew and he said that we should avoid using reinterprets. Is that not the case here too?
There was a problem hiding this comment.
Good catch! I went ahead and removed these
| tiny_fsm_send_signal(&instance->_private.fsm, signal_reflection_timeout, NULL); | ||
| } | ||
|
|
||
| static bool determine_byte_to_send_considering_escapes(self_t* instance, uint8_t byte, uint8_t* byteTosend) |
There was a problem hiding this comment.
What does it mean to have a byte escape
There was a problem hiding this comment.
The etx byte is the final byte of the protocol. It basically says the message is done now
edit: my bad Idk why i thought you were talking about the etx and not escape
There was a problem hiding this comment.
Actually, escape is how we distinguish between control characters acting as control characters (for example, start of frame, end of frame) and data that happens to look like control characters. So as an example, if we have a byte that matches STX in the middle of the payload, we'll prefix it with an ESC byte so that the receiver knows it's not really an STX control character, it's data.
| break; | ||
|
|
||
| case send_state_data: | ||
| if(determine_byte_to_send_considering_escapes(instance, instance->_private.send.buffer[instance->_private.send.offset], &byteTosend)) { |
There was a problem hiding this comment.
Going with my last comment but why would something be escaping from the rest of the data?
| { | ||
| uint8_t byteTosend = 0; | ||
|
|
||
| tiny_timer_start( |
There was a problem hiding this comment.
I noticed that this timer is meant to signal that there was a failure in the send but it doesn't seem like anything stops the timer if the next byte is sent. Is there assumption that there will always be failture?
| collision_idle_timeout); | ||
| } | ||
|
|
||
| static void state_collision_cooldown(tiny_fsm_t* fsm, const tiny_fsm_signal_t signal, const void* data) |
There was a problem hiding this comment.
so this is a single wire protocol - there is one data line. If something tries to send a message at the same time as the other one, the messages "collide" and we need to wait for a cooldown period to finish before attempting to send again
Co-authored-by: Ryan Hartlage <ryanplusplus@gmail.com>
Adding GEA2 communications to the interface.