When to Use the Scripting Toolkit

The following are examples of appropriate uses:

• You need a batch to load, extract and manage data using business rules
• Interface Salesforce with any Back-Office (rdbms, webservices, file exchange...)
• Need a lightweight interface, quick and easy development and deployment.
• If your skills are only Javascript and not other languages
• Load or extract CSV files, with transformation of field values
• Propagate your data to a database
• Create dynamically a power point presentation based on your data : using OLE to enhance your reporting experience.
• Identify and correct bad data (data quality)
• Propagate some data from one org to another
• Mass update old list values (after an update of the picklist values)
• Put it in the crontab to send emails with Key indicators on a weekly basis
• Poll data at a defined frequency to trigger an action (sending mail, running an interface..)
• Extract documents and attachments to free some storage space
• ...

The Scripting Toolkit should not be used :

• if you require user interaction in the browser : prefer the Ajax Toolkit
• if you need only to update a huge amount of Salesforce data : prefer the Apex language
• if you need to run your batchs using Unix : develop in Java instead.
• for an initial data loading : prefer the Data Loader

The following are examples of scenarios that require case-by-case analysis:

• Running time-critical processes : perhaps using both Apex and the Scripting Toolkit.
• Running as a service : either a specific development or using srvany.exe to create a scripted service.

An example of inappropriate usage is polling data every second. This would require too much API calls, and there is a risk to reach the maximum amount of API calls.

Scripting Toolkit Support Policy

The current release of the Scripting Toolkit is the only version that receives bug fixes and enhancements. When a new version is released, the previous version is not available anymore.
There is no official support of this Toolkit as it is submitted in the Salesforce Labs
By downloading the Scripting Toolkit, you get all the source code. You cannot redistribute this Toolkit, event if you have made changes.
You can use the toolkit for free, at your own risk (take it "as is")

Download the Scripting Toolkit

Latest version : 0.4.9
For each of these files, right click on them and choose "save as"
scriptingtoolkit.wsf (34 Kb) : the Scripting Toolkit engine
myscript.js (1 Kb) : sample code using the Scripting Toolkit
connection.js (54 Kb) : the Ajax ToolKit
Put these files in the same folder.

Other Resources

In addition to the content of this document, there are other resources available for you as you learn to use the Scripting Toolkit or its technology :

• Microsoft Script Center : http://www.microsoft.com/technet/scriptcenter/default.mspx
• An overview of WSH : http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnanchor/html/scriptinga.asp
• Sample WSH scripts : http://www.microsoft.com/technet/scriptcenter/csc/scripts/default.mspx

The Scripting Toolkit vs other languages

Here is a matrix that will let you choose the best system according to your needs.

Sample script Using the Scripting Toolkit

You can cut and paste the following sample code into the myscript.js file. This script queries data and displays it on the screen. For more information about using the API, see Salesforce's API documentation.

function Results(queryResult) {
	if (queryResult.size > 0) {
		var output = "";
		var records = queryResult.getArray("records");
		for (var i=0;i<records.length;i++) {
			var account = records[i];
			output += account.Id + " " + account.Name + "\r\n";
		}
		WScript.echo(output);
	}else WScript.echo("No records matched.");
	ToolKit.Quit();
}
var callback = {onSuccess : Results, onFailure : function(e){WScript.echo(e)}};
var queryResult = sforce.connection.query("Select Id,Name from Account", callback);

What are the differences between Ajax Toolkit and Scripting Toolkit ?

Running your script

The Scripting Toolkit can be run using the command-line.
Go to Start/Execute, enter "cmd" and press enter. You see a console with a prompt.
Go to the folder hosting the Scripting Toolkit and your custom script (myscript.js)
Enter the following command line :

cscript //nologo scriptingtoolkit.wsf /?

This script encapsulates Salesforce's AJAX Toolkit to provide a batch access to Salesforce
Usage : scriptingtoolkit.wsf [/user:value] [/pass:value] [/sid:value] [/server:value] [/proxyserver:value] [/proxyuser:value]
				 [/proxypass:value] [/debug] [/sandbox]

Options :

user        : Authorized user for the API.
pass        : User's password.
sid         : session id, if no user/pass provided
server      : the server of your org (ex:na3.salesforce.com or na3-api.salesforce.com)
proxyserver : hostname or IP of your proxy, and port (ex:proxy.internal.yourcompany.com:3128 or 192.168.0.7:3138)
proxyuser   : username if using a proxy with authentication
proxypass   : password if using a proxy with authentication
debug       : To dump on stderr some debug info
sandbox     : To authenticate on test.salesforce.com
Examples :      cscript //nologo scriptingtoolkit.wsf /user:myself@mycompany.com /pass:secret
                cscript //nologo scriptingtoolkit.wsf /debug /server:emea.salesforce.com /sid:Xabu.ZcS.....9AsNCSeX5jsUoLXQ=
                if no user/password and no server/sid, script tries to get an existing sid from an IE browsing session
                using a proxy : cscript //nologo scriptingtoolkit.wsf /proxyserver:192.168.0.7:3128 /proxyuser:jla /proxypass:jla
                        Proxy error ? See http://support.microsoft.com/kb/289481/en-us
                Compatibility test : cscript //nologo scriptingtoolkit.wsf //Job:diagnose

If you want to check your Windows configuration and identify any issue, run this command line :

cscript //nologo scriptingtoolkit.wsf //Job:diagnose

Note

You can redirect the output of your batch to a file using this command line :
cscript //nologo scriptingtoolkit.wsf /user:xxxx@yyy.com /pass:toto >myfile.txt
You can still use the /debug parameter as its output is redirected to stderr, not stdout

Debugging your scripts

You can use Visual Studio or Microsoft Script Debugger. You will be able to add break points and watch the value of objects.

Encoding your scripts

If you don't want somebody to see your code, you can encode it using Microsoft Script Encoder.
People will not be able to modify your code. Take care as this encoding is bijective : some tools can retrieve your source code.

Improving the speed

FOR ADVANCED USERS ONLY
The Scripting Toolkit is requesting a remote js file : connection.js
If you download this file (see the location in the source code if the Scripting Toolkit) and put it in the same folder as the Tookit, you can change the remote reference to the local file. Each time you start your batch, you will save a few seconds and bandwidth as the file will be retrieved from your local folder.

Hello World

WScript.echo("Hello World !");
ToolKit.Quit();

The first line uses the method to print a text to the console. The second line tells the Scripting Toolkit to end.
No access to salesforce data, but as it is using the Scripting Toolkit, an implicit login occurs to the application.

Retrieving data synchroneously

try{
	var queryResult = sforce.connection.query("Select Id,Name from Account");
	if (queryResult.size > 0) {
		var output = "";
		var records = queryResult.getArray("records");
		for (var i=0;i<records.length;i++) {
			var account = records[i];
			output += account.Id + " " + account.Name + "\r\n";
		}
		WScript.echo(output);
	}else WScript.echo("No records matched.");
}catch(e){
	WScript.echo(e);
}
ToolKit.Quit();

To retrieve data asynchroneously, see the sample in the previous chapter.

Benchmark your scripts

ToolKit.StartTimer();
var queryResult = sforce.connection.query("Select Id,name,Industry from Account");
ToolKit.StopTimer();
WScript.echo(ToolKit.TimeDiff()+" milliseconds");
ToolKit.Quit();

Timer functions are provided to identify how long it takes to run a few lines of code.

Extracting data to a CSV file

//var strTest=ToolKit.SOQL2CSV("Select type,Id,name,Industry from Account",["Name","Id","Type","Industry"],";");
var strTest=ToolKit.SOQL2CSV("Select Name,Owner.FirstName,Owner.LastName,Owner.Id,Account.Name from Opportunity");
WScript.echo(strTest);
if(strTest)ToolKit.StringtoFile(strTest,"test.csv");
ToolKit.Quit();

Two steps :

• Get the result set into a string in a CSV format
• Save the string to a file

Reading a file

WScript.echo(ToolKit.FileToString("test.csv",false));
ToolKit.Quit();

The file can be Ascii or Unicode (second parameter=true for unicode)

Identifying available objects

var describeGlobalResult=sforce.connection.describeGlobal();
var types=describeGlobalResult.types.toString().split(",");
for (i=0;i<types.length;i++) {
	WScript.echo(types[i]);
}
ToolKit.Quit();

You can retrieve the list of objects which you can use in your SOQL queries.

Describing an object

var describeResult=sforce.connection.describeSObject("Attachment");
var props=["activateable","createable","deletable","custom","keyPrefix","label","labelPlural","layoutable","mergeable","name"
 	,"queryable","replicateable","retrieveable","searchable","undeletable","updateable","urlDetail","urlEdit","urlNew"];
for(x=0;x<props.length;x++){
	WScript.echo(props[x]+"="+describeResult[props[x]]);
}
for(recordNumber in describeResult.fields){
	WScript.echo("Field "+recordNumber+":"+describeResult.fields[recordNumber].name+" ("+describeResult.fields[recordNumber].label+")");
	for(columns in describeResult.fields[recordNumber]){
		//if(typeof(describeResult.fields[recordNumber][columns])!='function')
			//WScript.echo(columns+"="+describeResult.fields[recordNumber][columns]);
	}
}
ToolKit.Quit();

This script lists available fields and their attributes for one sObject.

Generate an Oracle "create table" SQL query based on the sObject

SQL=ToolKit.sObject2OracleCreateTable("Account");
if(SQL){
	WScript.echo(SQL);
}else{
	WScript.echo(ToolKit.ErrText());
}
ToolKit.Quit();

This is a way to manage more records than a result set restricted to the batchsize.

Too many records

for(x=0;x<queryResult.size;x++){
	//your code here to use queryResult.records[x]);
	//...
	if(queryResult.done=='false')queryResult=sforce.connection.queryMore(queryResult.queryLocator);
}

This is a way to manage more records than a result set restricted to the batchsize.

Retrieving a document or an attachment

if(ToolKit.GetFile("Select Body from Document where Id='015300000000CJuAAM'","test.doc")){
	WScript.echo("success!")
}else{
	WScript.echo(ToolKit.ErrText());
}

ToolKit.Quit();

This will write the remote file locally.
The field "Body" should be in your SOQL. Only one file is retrieved per query : you can filter on the Id

Internal Scheduler

var frequency=10*1000;
function myScheduler(){
	WScript.echo(Date()+" scheduled.");
	//Run your scheduled code here
	window.setTimeout( "myScheduler()", frequency);
}
window.setTimeout( "myScheduler()", frequency );

You don't need a scheduler to run a script every X minutes or hours, even every minute.
The frequency is in milliseconds :

• Every 10 seconds => 10*1000
• Every hour => 60*60*1000

Sending an email

ToolKit.SendMail("Jean-Luc Antoine <antoinejl@xxartabus.com>"
	,"Jean-Luc Antoine <xxantoinexx@xxsalesforce.com>"
	,"Just a test"
	,"This is a text body\nwhich has two lines.");

ToolKit.SendMail("Jean-Luc Antoine <antoinejl@xxartabus.com>"
	,"Jean-Luc Antoine <xxantoinexx@xxsalesforce.com>"
	,"My second email"
	,"<html><body><h2>Cool !</h2>Very nice HTML</body></html>");

You can send an email with only 1 line of code (arguments : from, to, subject, body). The body can be either text or html (autodetected). The technical parameters are defined by the CDO configuration : you can dump them using the diagnose mode (see above in this page).

Deleting data based on a SOQL

if(ToolKit.DeleteFromSOQL("Select Id from Account where Name like 'ZZZyourprefix%'")){
	WScript.echo("Succefully deleted");
}else WScript.echo(ToolKit.ErrText());
ToolKit.Quit();

A quick way to drop some records with conditional filters.

Calling an Apex WebService

result = sforce.apex.execute('MyWebServiceClass' ,'MyMethod',{theParam:"Something", secondParam:"Something else"});
WScript.echo(result);
ToolKit.Quit();

Call an Apex Web Service the same way you would do it from within an SControl. executeAnonymous() is available too.

The WebService being called would be as an example :

global class MyWebServiceClass {
	webService static String MyMethod(String theParam, String secondParam) {
		return 'Hello '+theParam;
	}
}

Insert a CSV file into Salesforce

var CSVResult=ToolKit.CSV2Array("c:\\temp\\","account.csv");
if(ToolKit.InsertFromArray("Account",CSVResult,200)){
	WScript.echo("Succefully inserted");
}else WScript.echo(ToolKit.ErrText());
ToolKit.Quit();

This example uses a file with this content :

NAME,INDUSTRY
My new account 1,Manufacturing
My new account 2,Services
My new account 3,Manufacturing
My new account 4,Services