Comparison between Workflow, Process Flow, Visual flow

Feature	Workflow	Process Flow	Visual Flow / Visual Workflow
Description	Workflow allow users to set up criteria and actions	Process Flow automates your business processes by providing a powerful and user-friendly graphical representation of your process as you build it.	Visual Workflow enables you to create flows, which are triggered by users rather than events
UI	Runs in Background and take the action for ex field update	Runs in Background take the action for ex. Post chatter feed	Allows user to interact from UI and take input from user.
Initiated By	Triggered by events. For ex. Record save	Triggered by events. For ex. Record save	Triggered by users rather than events. For ex. Survey button click on lead object.
Supported Actions	·   Create a task ·   Create an email alert ·   Field Update New ·   Outbound Message ·   New Flow Trigger	·   Create a record ·   Update fields on any related record—not just the record or its parent. ·   Launch a flow—as an immediate or a scheduled action. ·   Send an email ·   Post to Chatter ·   Submit for approval ·   Call apex methods ·   The process builder doesn’t support outbound message	·   Lookup any object ·   Create Flow variables ·   Do CURD on any object
Scope	Limited to object in context and its master	Process Flow is not tied to any object	Visual Flow is not tied to any object
Interface	Point and click, multi-screen	Drag drop, Single Screen, easy to change & visualize	Drag drop, Single Screen, easy to change & visualize
Scheduling	Immediate and scheduled actions	Immediate and scheduled actions	Only Immediate actions
Use case	Send a web service callout to Inventory management system to update stock when  opportunity close successfully	Send a chatter feed to group of users when high value opportunity stage is changes	Wizard for self-help portal if it does not resolve issue then log a case with all the details.

Using NTLM authentication from Salesforce apex

We had requirement of integrating web-services that used NTML authentication from Salesforce.

The whole process of this integration was very interesting and so I thought of writing article which may help someone on how to tackle this type of integration.

Here are points we will cover:

1) What is NTML authentication and how it works

2) How we can use NTML authentication from Apex?

3) If you face any issues how to resolve them?

Ok, Lets start with point #1



1) What is NTML authentication and how it works: 

In a Windows network, NT LAN Manager (NTLM) is a suite of Microsoft security protocols that provides authentication, integrity, and confidentiality to users. There are various versions of it  NTLMv1, NTLMv2 & NTLM session.  Full details of NTLM is out of scope of this article. you can find more details at https://en.wikipedia.org/wiki/NT_LAN_Manager

How it works: 

A) The client sends an initial message to the server, advertising certain options or capabilities such as cryptographic algorithm support .

B) The server creates a challenge, c, and returns the challenge and the options or capabilities that it can support to the client .

C) The client computes a function on the challenge, resp = f(c, password), and sends the results to the server, along with the user's textual name and domain .

D)The server looks up the user (by the name passed) and computes the same function, f(c, user's password). If the result matches resp (that is, what the client sent in step C), the passwords are presumed to match, and the user is authenticated.

More Details: https://msdn.microsoft.com/en-us/library/cc239684.aspx




2) How we can use NTML authentication from Apex?

Well its dead simple. Nate Wallace have wrote apex classes that takes care of all over heads and you just need to deploy these classes to your org and use them.

Here is sample code on how to use these classes and give a call to NTLM SOAP service.


String body = 'XML_REQUEST_BODY_XXXXXXXXXXXXXXXXXXXX';
String ep = 'END_POINT_URL_XXXXXXXXXXXXXX';

HttpClient httpObj = new HttpClient('DOMIN_XXXX\\USERNAME_XXXXXX','PASSWORD_XXXXXXXX',True);  /* Note two forward slashes between domain and username \\  */
HttpRequest req = new HttpRequest();
req.setBody(body);
req.setMethod('POST');
req.setEndpoint(ep);

req.setHeader('Content-Type','text/xml;charset=UTF-8');

HttpResponse res = httpObj.send(req);
System.debug('Response**:' + res);
System.debug('Response**:' + res.getBody());

This should give you a XML response. Well but what if you did not get valid response... this bring us to next topic


3) If you face any issues how to resolve them?

When working on integration there are generally (at least) two components. First is Server and second is client. In our case our endpoint is server and client is in apex. When we face issue it becomes difficult to identify the issue. To resolve issue we should first decide where the issue is? in client or in server?

To find that out we can use tool like SoapUi to replace our client. SoapUi is tested and trusted tool so when we use it properly we eliminate the possibility of client side issue for ex. incorrect  XML.

To use SoapUI with NTLM authentication you need do following:

a) Import WSDL into SoapUI

b) Click on any service and open Request1

c) On left hand bottom side of Request1 there is authorization tab, click on it.

d)  Select add new authorization and select type As NTLM and provide the credentials

e) Now you can replace the ? in your Request1 with actual values and click submit request button.


If you get positive response when you use SoapUi then it means that there is something wrong in your client:
Examples of client side issues:
1) Invalid XML
2) Incorrect parameter in XML
3) Invalid Header
4) 401 error : authentication failed due to user name/ password / domain
5) 500 Internal server error due to wrong data

If you get negative response when you use SoapUi then it means that there is something wrong in your Server:

1) 500 Internal server error due long processing time
2) 401 error:  network restriction like IP restriction  

Conclusion:While working on integration you need to isolation the issue whether its in client or server and then proceed further. 

Difference between ActionFuction And ActionSupport

Difference between ActionFuction And ActionSupport

Visualforce provides three great components to call controller methods they are ActionPoller, ActionFuction & ActionSupport. Among these three components ActionFuction & ActionSupport are very similar and in most of cases can be used interchangeably. Following is example of same.

VF1– Action Function

=================================

<apex:page controller="exampleCon">           

    <p><apex:outputText value="Clicked? {!state}" id="showstate" /></p>              

    <apex:outputPanel onclick="methodOneInJavascript()" styleClass="btn"> Click Me </apex:outputPanel>     

                        <apex:form >

                                                <apex:actionFunction action="{!methodOne}" name="methodOneInJavascript" rerender="showstate">

                                                                        <apex:param name="firstParam" assignTo="{!state}" value="" />

                                                </apex:actionFunction>

                      </apex:form>

</apex:page>

VF2 – Action Support

=================================

<apex:page controller="exampleCon">

<apex:form>           

    <p><apex:outputText value="Clicked? {!state}" id="showstate" /></p>               

    <apex:outputPanel styleClass="btn"> Click Me        

        <apex:actionSupport event="onclick" action="{!method2}" rerender="showstate">

            <apex:param name="firstParam" value="Yes" />

         </apex:actionSupport>

    </apex:outputPanel>   

</apex:form>  

</apex:page>

Controller

=================================

public class exampleCon {

private String state = 'no';         

    public void setState(String n) {

        state = n;

    }

public String getState() {

        return state;

    }            

public PageReference methodOne() {

        return null;

    }

public PageReference method2() {

        state = ApexPages.CurrentPage().getParameters().get('firstParam');

        return null;

    }

}

As you can see I can use both Action fiction and Action support to archive the same functionality.  So the question is they truly interchange? Well the answer is No!

When should I use Action Function and when to use Action support?

When you want to add ACTION to any (single) apex component then you should use Action support

When there multiple components which needs to call controller action then you should use Action Function.

Why?

The reason lies in the background implementation. When you define Action Support in your VF it generates AJAX code with function name that call controller method. If you use this in loop (repeat) this defines multiple JS function with same name which throws error in JS and your page do not work properly.

In Case of Action Function it is defined at one place and can be called from many places. So if you have loop (repeat) in your code then calling code can be repeated in loop and that is why Action Function should be used with Repeat.

Salesforce Static Code Analysis using codescan.villagechief.com

Salesforce Static Code Analysis .

Like many other sophisticated programming language there is Static Code Analysis tool available by codescan.villagechief.com. This code review tool can give us lots of benefits. I have listed some of them bellow.

• Warning on some governor limits  - for ex SOQL in loops

• Coding standards: naming conventions, fields at top of file, etc

• Redundant code

• Code simplification suggestions

• Clean code, less mistakes less bugs

So, now the question is hot to set up Static Code Analysis tool? 

There are 2 ways you can use a) Eclips IDE  2) Set up SonarQube Server

Setting SonarQube Server is little tricky and it need licence from codescan.villagechief.com so for now lets focus on how to set up Eclips IDE

Steps to install CodeScan Apex PMD Eclipse Plugin:

1.In Eclipse go to Help -> Install new Software-> use http://codescan.villagechief.com/eclipse/ link to install plugin

2) Select the plugin and click next next...

3) After successful installation of plugin, Go to project Explorer -> Right click on your sales-force project -> ApexPDM -> run ApexPDM

4) To view Codescan result, goto ApexPDM view 

5) You can run ApexPDM on specific class also

You can scan your code online.Goto http://codescan.villagechief.com/report/index Paste your class for code analysis

5) 

Salesforce oAuth Username Password Flow is security risk?

Salesforce oAuth Username Password Flow is security risk?

When user login from the UI, SFDC checks if its first successful login attempt from that machine. If it is true then it asks for addition security code which is sent to your email address. The only exception to this is made when that machine’s IP is in trusted IP range.

Similar additional check is implemented for SOAP API login call. It requires User name, password & security token to get access to any data.

If you consider above security measures then Salesforce oAuth Username Password flow seems vulnerable because you do not need any additional information to call Username Password flow.

I mean you defiantly need client_id & client_secret but anyone can create connected app in developer org and they can bypass the security if they just have user name & password.

Am I missing something?