invalid format exception when parsing float

Questions & Answers
1

Hello all,

I’ve been creating a tele-transportation script that gets the coordinates from an sql server based database.
After downloading the data and splitting it with Regex.Split, I have to parse it into a float so the vector3 gets the numbers.

My problem is that, even though it works and during gameplay everything functions, in the console I get format exception messages.

“FormatException: Invalid format.
System.Double.Parse (System.String s, NumberStyles style, IFormatProvider provider)”

function teletransportation(searchname){
    //create the form with the name of the point to which we want to teletransport
	var sendPoint : WWWForm = new WWWForm();
		sendPoint.AddField("PointName",searchname);
		
	//send the WWWForm via WWW
	var getCoords : WWW = new WWW("../DataBase/FindPoint.php",sendPoint);
	yield getCoords; //Wait for the data to return
	
	//split downloaded data
	var received_data = Regex.Split(getCoords.text/*data*/,"</next>");
	//divide split data into vars 
	pointCoords.x = float.Parse(received_data[0]);
	pointCoords.y = float.Parse(received_data[1]);
	pointCoords.z = float.Parse(received_data[2]);

	//traslation function
	gameObject.Find("Player").transform.position.x = pointCoords.x;
	gameObject.Find("Player").transform.position.y = pointCoords.y;
	gameObject.Find("Player").transform.position.z = pointCoords.z;
}
3

Well, wouldn’t it be interesting how your returned data looks like? That’s the real problem.

I guess it’s something like:

"<next>10</next>  <next>20</next>  <next>30</next>"

If you split it at “</next>” the 4 strings would look like this:

received_data[0] = "<next>10"    
received_data[1] = "  <next>20"    
received_data[2] = "  <next>30"
received_data[3] = ""

Why do you use xml for returning 3 values? Just 3 comma-seperated-values would be enough. If you have to use xml, why do you parse the xml string manually? There are many xml parsers out there.

Some further side-notes:

  • I’m not sure if you need pointCoords outside of the function, but it looks like it shoule be alocal variable.
  • pointCoords is a vector3, so why do you assign it component by component to Transform.position? gameObject.Find("Player").transform.position = pointCoords; is enough.
  • Find() is a static function of GameObject and you should use the classname instead of an instance. In other words : gameObject.Find( → GameObject.Find(
  • GameObject.Find is a quite low function. It’s ok to call it once every level change, but avoid it in Update() or other cyclic called functions.
  • To see what float.Parse() actually have to parse, you can put a Debug.Log(received_data[0]); in your code.
2

Ok, I don’t know WHY that works, but I’ve looked at other answers and found that parseFloat() doesn’t give that error, so I just used that and I no longer receive the error.

Sorry 'bout that :slight_smile:

*note: if the admins want to delete this question, feel free.