BRT Community

Hello,

yeah I know this but the question is: Why there is a typecast needed?
In arg there are strange numbers.

Why can't I just let the pointer uninitialized like:
struct sub_info *subInfo;
subInfo->topic = TOPIC_CONTROL;

Is it because in arg is the adress of the pointer on the heap? And pointers need to have an address where they point to?

Dear BRT Community,

let's have a look on this callback fuction from the MQTT example:

static void client_mqtt_incoming_publish_cb(void *arg, const char *topic, u32_t tot_len)
{
   struct sub_info *subInfo = (struct sub_info *)arg;

   DEBUG_PRINTF("MQTT RECIEVE: %s (%u)\r\n", topic, (unsigned int)tot_len);

   memset(subInfo, 0, sizeof(struct sub_info));

   /* Decode topic string into a user defined reference */
   if(strcmp(topic, TOPIC_CONTROL) == 0)
   {
      subInfo->topic = TOPIC_CONTROL;
   }
   tfp_printf("%d\r\n", __LINE__);
}

I don't get what the intention of the first line of this function is.
When I'm debugging the code then there is a strange number in arg.
However, I tried to just struct sub_info *subInfo; but the result of this is that the microcontroller does a reset after the if condition.

That means this line has a purpose. But what is the purpose of this "arg"? Why does the micorocntroller a reset when not including "= (struct sub_info *)arg;"

I would really appreciate if someone could help me understanding this.

Hey,

thanks for the answer.

First I thought it makes it more complex for me to develope my MQTT aplication because I don´t have any experience with RTOS.
But now I figured out that it shall be harder to develope applications when I just use the raw API of lwIP (Bare metal). The API with an RTOS shall be easier for application developement.
Is that correct?
Probably it would be hard to port your MQTT example to bare metral right? So I should just better study how an RTOS works.

One question: Which lwIP-API do you use for your lwIP examples? Is it the Netconn API or the BSD Socket API?

Best regards

Vladi

Hello BRT community,

sorry I should have upload a zip file.
Now there is a zip file attached:

It is basically the same project you uploaded.
But I would like to change the color of the Partial Ring when a specific value is exceeded.
I could not fiigure that out yet. Where is the error in the file I attached?
Why the color doesn't change?

I also tried to implement the color change with a timer and a button but even this does not work

Best regards
Vladi

Attached you can see the two files of the ESD.

Hello guys,

I'm using the MM900EV3A Board with the FT900 microcontroller and I want to implement a MQTT communication between the board and my host computer.
Bridgetk provides an example for implementing MQTT with the lwIP library using freeRTOS.

My question is: What are the bebefits of implementing MQTT together with an RTOS?

During my research for MQTT implementations I found another very common way to implementing MQTT without lwIP and freeRTOS: Using the eclipse paho library for C.
My first impression is that it seems to be easier to implement a mqtt client with paho rather than with lwIP/freeRTOS.
But what is the disadvantage of using a "normal" mqtt library?

I hope you guys can help me understanding this

scoprioprise, could you please help me?

I'm struggeling right now with some things of the screen designer.
In the first attached picture I created a counter inside ESD and connect it to a numeric lable and the ESD_Partial_Ring. That works fine. When a threshold is exceeded the color of the partial ring changes.

But in the second attached picture this color change just doesn't work properly. It is actually the same logic, just the counter signal comes now from an external source (well, it is just a counter in an own function "readSpeed()").

Do you have an idea what could be wrong? I tried several things but I just don´t get it working...

And also: Are there opportunities to start the esd timer automatically at the beginning of the programm? I tried it with the slot-logic-function and connected it to the esd timer but then I even could not see changes at partial ring... Or even better: Are there opportunities to start this condition-logic-functions without timer? I mean another input for the "Do"-field?

I would really appreciate if you have an idea. If you want we could also discuss this via mail

Hello scoprioprise,

thank you for your hints. I will consider this when I'm working on the application.

But actually I can't find your mentioned files: Ft_Esd_Framework/Ft_Esd_app.c ;    Ft_Esd_MainLoop.c
I don´t have these files in the generated project...

I found a main-function in the file "Ft_Esd_Support" as you can see attached.

Would you suggest to implement my own functions before "Esd_Loop(&ec);" or inside this function (which you can see attached)?

Worarround attached

Hello guys,

I've exported the generated code from the EVE Screen Designer into a eclipse workspace.
I'm wondering that there is no "main.c" in the generated code...

Anyway, I would like to extend the code of the screen designer with my own functions.
For doing this I've just added a "main.c" and simple led function:

int main(void)
{
"led on"

while (1)
return 0;
}

But I could not compile the code. There was the following error: make***[project_name.elf]

How is it possible to run your own code with the screen designer code parallelly?

It would be great if anyone could answer me this simple question

Edit:
The main-function in the generated code is in the file Ft_Esd_Support.c => there you can add your personal code

And currently there is a bug with debugging the exported code from the ESD into eclipse, because the configuration of debuggin is wrong.
The exported code just has the options "debug or release" and not "FT900_debug or FT900_release"
Next post you see a workaround to overcome the problem.

Hello guys,

i'm using a Riverdi EVE 3 BT816 display in combination with the MM900EV3A.

I want to generate code with the ESD and import it to an eclipse project. But the targeted eve platforms for the Screen Designer are VM816C or ME812A.
That means I have to adapt the generated code to my Riverdi display.
First: Are the settings I chose in the ESD fine ? (see attached)
In which files do I have to make changes after export the project to eclipse?
I would assume in EVE_config.h
But which of the line I have to comment out or change? The Riverdi has a flash of 64Mb

And I have also another question: The generated source code doesn´t have a main.c. How can I run the generated code on the graphic controller?

In app_generated.c there are the following functions:

void Ft_Main__Start()

void Ft_Main__Update()

void Ft_Main__Render()

void Ft_Main__Idle()

void Ft_Main__End()

do I just have to call the functions like that in my main function?:

void main()

{
static App application;
void Ft_Main__Start();
Ft_Main__Render()

}


I would really appreciate If anyone could help me.

Best reards

Vladi

Thank you that would be great!
But actually I´ve realized that I don´t need touch functionality. Therefore the display list for the ESD Partial Ring Widget would be sufficient.

Vladi

Hello guys,

currently I'm working on programming a Riverdi TFT EVE 3 Display with an Nucleo STM32 board.
I don´t want to use the Screen Designer for my application but I would like to integrate one widget of the Screen Designer in my code.
It is the "ArcSlider" widget. As I can see in your library for the Screen Designer there is the file "Ft_Esd_ArcSlider.h" with the struct for building ArcSlider objects.
I could import the ArcSlider specific files into my workspace and then build the object like this:

Ft_Esd_ArcSlider object1, *pobject1
pobject1 = &object1;
pobject1 ->Widget.Active = 1;
pobject1 ->Widget.LocalX = 85;
pobject1 ->Widget.LocalY = 36;
pobject1 ->Widget.LocalWidth = 347;
pobject1 ->Widget.LocalHeight = 216;

or use the appropriate fuction provided by your library.

But my question is: How is it possible to display the object/ to put in the the display list ?

Best regards

Vladi