Hi,

In our project we have cloud code that sends emails to players in french through mailgun. It's working fine in english but in french the resulting emails are badly encoded. Every accentuated characters are replaced with �. I've tried to set content-type header to utf-8 , use postJson instead of postForm with no luck. Do you know what is the cause of the problem?  

here's a simplify version of our code cloud code:

var form = {
	"to":"jdoe@mycompany.com", 
	"from":"cgauthier@mycompany.com",
	"fromname":"John Doe",
	"subject": "Probl\u00e8me avec les \u00e0",
	"text":"\u00e9l\u00e8ve"
};

var httpSender = Spark.getHttp("https://api.mailgun.net/v3/mg.mycompany.com/messages");
httpSender.setBasicAuth("api","key-9x9x9x9x9x9x9x9x9");

var response = httpSender.postForm(form);
var responseCode = response.getResponseCode();
if (responseCode !== 200){
    Spark.setScriptError("error", "Send Email failed, error code: " + responseCode + " : " + response.getResponseString());
}
else {
    Spark.setScriptData("data", "success");
}

Expected result:

Subject: Problème avec les à

Body: élève

Actual result:

Subject: Probl�me avec les �

Body: �l�ve

I've tried with the extra parameters and it's not working. By the way i'm not posting json, i'm using postForm() not postJson()

Using postman, it's working great. Here's the request outputed in Curl by Postman:

curl -X POST -H "Authorization: Basic B----------------------------U" -H "Cache-Control: no-cache" -H "Content-Type: multipart/form-data; boundary=----WebKitFormBoundary7MA4YWxkTrZu0gW" -F "to=cgauthier@mycompany.com" -F "from=cgauthier@mycompany.com" -F "fromname=Christian Gauthier" -F "subject=Problème avec les à" -F "text=élève" "https://api.mailgun.net/v3/mg.teamupmobileapp.com/messages"

So here's what I tried so far and nothing's working:

A) var httpSender = Spark.getHttp("https://api.mailgun.net/v3/mg.teamupmobileapp.com/messages?characterEncoding=UTF-8&contentType=application/json");

  email still has �

B) var httpSender = Spark.getHttp("https://api.mailgun.net/v3/mg.teamupmobileapp.com/messages?characterEncoding=UTF-8");

  email still has �

C) httpSender.setHeaders({"content-type": "application/json; charset='utf-8'"});

  email still has �

D) httpSender.setHeaders({"content-type": "charset='utf-8'"});

  mailgun returns an error: from parameter is missing

E) httpSender.setHeaders({"content-type": "form-data; charset='utf-8'"});

  mailgun returns an error: from parameter is missing

F) httpSender.setHeaders({"Content-Disposition": "form-data; charset='utf-8'"});

  email still has �

I'm out of idea :-(

Hello guys,

I know this is a tough one :-) but do you have any hints on this one?

While searching for solution I tried your built-in sendGrid solution and I discovered an interesing problem. Accent (é for example) entered through your web site interface (NoSQL, Test harness or Cloud Code editor) are not encoded in a way that outlook likes (no problem with gmail)

The fist image show 2 documents:

code:"TEST0" was added using an UTF8 encoded json file and a REST Api call

code:"TEST1" was added using the browser NoSQL interface

If you run the following code (notice the subject has an é that will also be badly encoded since it's was entered in the "cloud code" interface):

var msg0 = Spark.metaCollection("EmailDebugging").findOne({"code":"TEST0"}).value;
var msg1 = Spark.metaCollection("EmailDebugging").findOne({"code":"TEST1"}).value;

var mySendGrid = Spark.sendGrid("xxxxxxx", "yyyyyyyyyyy");
mySendGrid.setFrom("notimportant@somewhere.com", "Tester");
mySendGrid.addTo("anyhotmailrecipient@hotmail.com","Your Name");
mySendGrid.setSubject("This é will fail");
mySendGrid.setText(msg0 + "-" + msg1);
var response = mySendGrid.send(); 
response = JSON.parse(response); 
if (response["message"] !== "success"){
    Spark.getLog().error("Send Email fail: " + response.message);
}

 While searching for solution I tried your built-in sendGrid solution and I discovered an interesing problem. Accent (é for example) entered through your web site interface (NoSQL, Test harness or Cloud Code editor) are not encoded in a way that outlook likes (no problem with gmail)

The fist image show 2 documents:

code:"TEST0" was added using an UTF8 encoded json file and a REST Api call

code:"TEST1" was added using the browser NoSQL interface

If you run the following code (notice the subject has an é that will also be badly encoded since it's was entered in the "cloud code" interface):

The resulting email from outlook showing the encoding problem :