Naming conventions: OO names vs. Underscored names

[Former Member]

Hi all,

It seems to be a good choise to use OO naming (like we have in JAVA or .NET) to the objects in the Integration Directory and Repository. For example:

CustomerDataType

CustomerMessageType

CustomerMessageInterface

etc...

But I haven´t found names like this in XI documentation and in XI official courses. They commonly use:

DT_CUSTOMER (for a Customer Data Type)

MT_CUSTOMER (for a Customer Message Type)

MI_CUSTOMER (for a Customer Message Interface)

and so on...

What do you think about this? Would be a good choice to use OO naming stands (like in the first example I wrote)?

pls, comment.

thanks!

roberti

[henrique_pinto · Accepted solution]

I always use OO naming, like:

CustomMessage (DT)

CustomMessage (MT)

CustomMessageInboundAsync (MI)

and I use underscore for mappings:

CustomMessage1_CustomMessage2 (MM)

CustomMessage1InboundAsync_CustomMessage2OutboundSync (IM)

The problem is when I have imported RFC's, which almost always have underscore naming convention, like:

ZCUSTOM_RFC

and if I create Message Interfaces for it, it'll be like:

ZCUSTOM_RFCAbstractAsync

which is kinda weird.

Anyway, you decide what suits you better.

Regards,

Henrique.

[Former Member · Accepted solution]

Hi Henrique,

thanks for your help.

You where exactly at the point: importing RFCs (or even BAPIs) the OO naming convention can´t be maintained because the original object names. At this point the Integration Repository will have diferent naming conventions for diferent cenarios (using or not rfcs and bapis).

Anyone else has already dealed with this?

Cheers!

roberti