ABAP on HANA Dump

[mser]

Hi All,

the following code dumps in some cases, not always:

SELECT * FROM jest INTO TABLE tb_jest
FOR ALL ENTRIES INtb_output
WHERE objnr = tb_output-objnr
AND stat IN rs_stat AND inact = ' '.

LOOP AT tb_jest INTO wa_jest.
READ TABLE tb_output WITH KEY objnr = wa_jest-objnr.
DELETE tb_output INDEX sy-tabix.
ENDLOOP.The dump happens on line

DELETE tb_output INDEX sy-tabix.

because read table returns sy-subrc <> 0.

Can someone spot the problem ?

thanks a lot

Filippo

[horst_keller]

Debug and read the documentation about the statements involved ...

[mser]

Hi Horst thanks but not really helpful

[horst_keller]

Why? What's the contents of sy-tabix? What does the short dump say?

[mser]

found the problem.

In some minor cases, the table tb_jest have some duplicate entries(same objnr). So the first value in tb_jest deletes the corresponding entry in table tb_output and when it reaches the second value in tb_jest, it can't find a corresponding value in tb_output.

Thanks

Filippo

[matt]

Which could only be determined by knowing your data, and debugging, as Horst said...

[UweFetzer_se38]

... and please don't do SELECT * on HANA (or any other database, but especially HANA)

[Sandra_Rossi]

+1 to Horst. As the short dump indicates probably an "index error" on the DELETE statement, read the ABAP documentation - DELETE itab - itab_line - Alternative 2 INDEX idx :

If idx contains a value of 0 or less, an exception is raised that cannot be handled.

Then debug and you'll see the value of sy-tabix after READ TABLE is zero when the line is not found (if you read the documentation again, you'll see it only states that sy-tabix is set when a line is found, so consider that you shouldn't use sy-tabix if the line is not found.

[former_member428032]

Hi,

You could use if sy-subrc is initial XXX endif.

after read statement.

if sy-subrc is initial.

DELETE tb_output INDEX sy-tabix.

endif.