avoid-default-key.md

code pal for ABAP > Documentation > Avoid DEFAULT KEY

Avoid default table keys

What is the intent of the check?

Default table keys (declared by the WITH DEFAULT KEY clause in a table definition) can lead to obscure mistakes because they ignore numeric data types and they make it the intent of the key unclear. SORT and DELETE ADJACENT statements without explicit field list will resort to the primary key of the internal table, which in the case of default keys can lead to unexpected results, in particular in combination with READ TABLE ... BINARY SEARCH statements.

Therefore, this check searches for internal table definitions that declare a default table key.

How to solve the issue?

Either specify the key components explicitly

DATA itab1 TYPE STANDARD TABLE OF row_type WITH NON-UNIQUE KEY comp1 comp2.

or resort to EMPTY KEY if you don't need a key at all.

DATA itab1 TYPE STANDARD TABLE OF row_type WITH EMPTY KEY.

If you declare an explicit key, also consider whether you actually should use a sorted or hashed key.

See also: Clean ABAP - Avoid DEFAULT KEY.

What to do in case of exception?

In exceptional cases, you can suppress this finding by using the pseudo comment "#EC DEFAULT_KEY which has to be placed after the declaration:

CLASS-DATA itab1 TYPE STANDARD TABLE OF row_type WITH DEFAULT KEY. "#EC DEFAULT_KEY

DATA itab1 TYPE STANDARD TABLE OF row_type WITH DEFAULT KEY. "#EC DEFAULT_KEY

TYPES: BEGIN OF type1,
         non_unique TYPE STANDARD TABLE OF row_type WITH NON-UNIQUE KEY object,
         default_key TYPE STANDARD TABLE OF row_type WITH DEFAULT KEY, "#EC DEFAULT_KEY'
         empty_key TYPE STANDARD TABLE OF row_type WITH EMPTY KEY,
       END OF type1. 

Example

Before the check:

DATA itab1 TYPE STANDARD TABLE OF row_type WITH DEFAULT KEY.

After the check:

DATA itab1 TYPE STANDARD TABLE OF row_type WITH NON-UNIQUE KEY comp1 comp2.

DATA itab1 TYPE STANDARD TABLE OF row_type WITH EMPTY KEY.

Further Readings & Knowledge

• Clean ABAP: Avoid DEFAULT KEY