What is the error in this code?

Projects General Guidance
1

I am trying to make a IdeaSpark weather station work. I have installed all libraries and etc per instructions. When I try to compile, I get the following error. I have included the code that causes the error. Can someone point out the error and how to fix it?

Here is the error

C:\Users\hamil\Documents\Arduino\libraries\esp8266-weather-station-master\src\TimeClient.cpp: In member function 'long int TimeClient::getCurrentEpochWithUtcOffset()':

C:\Users\hamil\Documents\Arduino\libraries\esp8266-weather-station-master\src\TimeClient.cpp:124:67: error: invalid operands of types 'double' and 'long int' to binary 'operator%'

return round(getCurrentEpoch() + 3600 * myUtcOffset + 86400L) % 86400L;

Here is the erroneour routine

long TimeClient::getCurrentEpochWithUtcOffset() {
return round(getCurrentEpoch() + 3600 * myUtcOffset + 86400L) % 86400L;
}

Thanks for the help.

2

The error is saying you are using a type of 'double' with the modulus operator, but since you didn't show the entire sketch, only you will know.

Possibilities are:

  • that the function getCurrentEpoch() returns a double [probably not]
  • myUtcOffset is a constant that is of type double
  • the function round() returns a type of double [most likely cause]
3

MyUtcOffset is a float
GetCurrentEpoch is a Long
Round is an integer (per google)

So it looks like I am adding a float to an long, rounding to an integer, doing the mod thing, and returning a long. The function returns a long.

Seems OK to me. What say you??

4

MyUtcOffset is a float

This causes the other variables to be promoted to float, which round also returns, and the result is incompatible with mod.

5

But round returns integer, per google.

OK, float Plus integer = float. Right
Round (float) = integer
integer %integer = integer

integer is returned.

I guess I will have to do the change things and try method to figure out the issue.

6

In C/C++ round() returns a float; the closest integral value to the input argument (i.e. the result has no fractional component).

7

However, I cast the term myUtcOffset to be a long as shown below and it compiles. I don't get it.

long TimeClient::getCurrentEpochWithUtcOffset() {
return round(getCurrentEpoch() + 3600 * (long)myUtcOffset + 86400L) % 86400L;
}

8

The above seems a bit risky so I cast the round function as a long. It compiles but I have other issues.