Dynamic Selection of DataSource Program in Exit

[Former Member]

Dear Experts,

I have gone through the below Ho-To document on SDN.

But to be honest, I could not understand how it is beneficial over the Include approach that is generally used and is also documented in this How-To.

https://www.sdn.sap.com/irj/sdn/index?rid=/library/uuid/2d99121a-0e01-0010-e78c-b1ae566a2413

Any expert opinions please...

Regards

[former_member205352 · Accepted solution]

That approach is benificial in following ways :

1) Multiple users can edit the user exit ZXRSRU02,Which wont be possible other wise.

2) When you are in the middle of some enhancement and other person wants his code to be moved to production.You need to delete or comment your code.With above approach you neednot put your include.

3) Untill your program is complete you neednot include in user exit.

Actually SAP had better method for this where we neednot manipulate the user exit itself we had to write codes in function modules but I dont have the document.

Hope this helps.

Edited by: Praveen G on Oct 7, 2008 5:12 AM

[Former Member · Accepted solution]

From my point of view of having multiple developers - it stops the leap frogging of cmods into production

[Former Member · Accepted solution]

Let me assume, I am still using an old approach and I have to 2 separate includes ZBW_0MATERIAL_ATTR and ZBW_0CUSTOMER_ATTR.

Now if I am a developer who is working on Material Attribute enhancement and you are a developer working on Customer Attribute enhancement. Then I can just modify and activate my include (Executable Type) ZBW_0MATERIAL_ATTR and you can independently work in your include (Executable Type) ZBW_0CUSTOMER_ATTR .

Can you please explain what is the problem with the above old approach of modular programming?

Regards

[Former Member · Accepted solution]

You can if you as a developer remmber to transport the lower level and not higher level objects

In addition what are you going to do if you have two new cmod includes?

[Former Member · Accepted solution]

I am happy to know that we are on the same page. As long as, I as a developer transport only my include and my transport request does not contain Exit ZXRSRU02, then there is no problem with old approach.

I also appreciate your 2nd point. If I have to create 2 new includes, then I will have to first hard-code them in User ZXRSRU02 which means my transport request has to include Exit ZXRSRU02 and it can get problematic now.

I guess, another advantage I would have with new approach is that code in exit ZXRSRU02 would be quite cleaner and just contain basic 8-10 lines as described in the How-To.

If there are any other benefits with new approach that you have understood then please share.

Regards

[Former Member · Accepted solution]

Excellent - we are on the same page

My job as a architect is to mitigate risk and anything that involves risk aversion and stopping leap frogging without anybody really having to worry - then I am all for

[Former Member · Accepted solution]

I think the person who wrote that How-To document has done fantastic job but if he/she wishes the document to be understood by larger audience then probably detailed scenario-wise explanation of the pitfalls of older approach would have made it possible for ordinary consultants like me to understand it in the first round of reading.

Anyways, thanks to you and Pravin for help from you guys.

Regards,

[Former Member]

Doubts are clarified now.