Best practice to pass multiple parameters to Spark.Function()

Firmware
1

Dear all,

I want to pass from a html form multiple parameters to one spark cloud function.
Best would be to have always a tuple of values (e.g. parameter name & value) so that in later processing it you have a high flexibility to react on different number of parameters.

As I understood the only way is to pass one string to this function separated by a delimiter character.

Is there any best practice on how to parse this?
Is this even the right way to do it?

2

Depending on what you want to do, you could pass some integers which each represent a certain function. So you could send 345, go through some if statements ( if > 300 do this, - 300 etc) and execute your function.

Or did I misunderstand you? :wink:

3

I have a feeling he’s looking more into passing multiple parameters at once. Like for example color values, which could be “red:10,green:75,blue:255”. There are some examples of that floating around the forum, one of them being the FireTorch. Personally, I’ve put the arguments in a JSON style, since that’s relatively easy to parse on most platforms/languages.

1 Like
4

Thanks for your answers.
Indded I am looking into something like @Moors7 descibed. Exactly about passing RGB values, but also additional information like which led strip to control, what kind of action the LED strip should perform, on/off signals etc.

5

I have a few sketches doing what you want... I'm sending multiple 'commands' to spark. I created a Class to decode String messages. Here is an example:

In Header:

class httpCommand {
  String argument;
public:
  void extractValues(String);
  String mssgCommand (void) {
    return argument.substring(argument.indexOf("command#") + 8, argument.indexOf("#text="));
  }
  String mssgText (void) {
    return argument.substring(argument.indexOf("#text=") + 6, argument.indexOf("#value0="));
  }
  int mssgValue0 (void) {
    return (argument.substring(argument.indexOf("#value0=") + 8, argument.indexOf("#value1="))).toInt();
  }
  int mssgValue1 (void) {
    return (argument.substring(argument.indexOf("#value1=") + 8, argument.indexOf("?"))).toInt();
  }
};
//
void httpCommand::extractValues (String stringPassed){
  argument = stringPassed;
}

In Setup:

Spark.function("httpRequest", httpRequest);

It is looking for me to send to Spark:

params=command#MY_COMMAND#text=MY_TEXT#value0=MY_INT#value1=MY2ND_INT?

it is certainly extensible...

I sort incoming commands like this:

Spark.function( ) function:

int httpRequest(String mssgArgs)
{
  DEBUG_PRINTLN("command recieved...");
  oldData = false;
  lastDataTransmitTime = millis();
  httpCommand command;
  command.extractValues(mssgArgs);
  DEBUG_PRINTLN(command.mssgCommand());
  boolean badMessage = true;
  for (int i = 0; i < NUMBER_OF_MESSAGE_TYPES; i++)
  {
    if (command.mssgCommand().equals(messageType[i]))
    {
      DEBUG_PRINTLN("Valid Message Recieved...");
      DEBUG_PRINTLN("Message type:");
      DEBUG_PRINTLN(command.mssgCommand());
      DEBUG_PRINT("messgText = ");
      DEBUG_PRINTLN(command.mssgText());
      DEBUG_PRINT("messgValue0 = ");
      DEBUG_PRINTLN(command.mssgValue0());
      DEBUG_PRINT("messgValue1 = ");
      DEBUG_PRINTLN(command.mssgValue1());
      updateVariables(i, command.mssgText(), command.mssgValue0(), command.mssgValue1()); // Pass decoded variables to function
      badMessage = false;
    }
  }
  if (badMessage)
  {
    DEBUG_PRINTLN("non-conforming message attempt...");
    return -1;
  }
  else return 1;
}

against an array of commands like this:

in the header:

const String messageType[NUMBER_OF_MESSAGE_TYPES] = {
  "ledStatus", "alarmState", "garageState", "guestGarageState", "weatherCondition", \
  "outsideTemp", "outsideHumid", "airconSetpoint", "weatherForecast", "messageData", \
  "todayHigh", "todayLow", "windSpeed", "windDirection", \
  "relayState", "brightLevel", "emailCount", "resetSpark", "resetHour" };
1 Like
6

Hi @BulldogLowell,

I think this is a interesting solution. I haven’t had yet time to test this but do I understand correctly that you even don’t use the Spark Cloud but sending the http request directly to the Spark?

7

no, I am using Spark Cloud with a poorly named function :frowning: that was a carryover from the same program I wrote for my ethernet connected Arduino.

Sorry to confuse you... lazy on my part.