Hi Christian,
How many entries has your table 'ct_data_tra' and what is the field size of 'trans_num' ?
That's good question, but my program is still on development system and ct_data_tra is small. It contains only 15 rows, trans_num field is 10 characters long. So I don't expect memory problems.
I activated the trace in ST05, and we see that DB-select has only 15 question marks for trans_num:
SELECT
DISTINCT "MANDT" , "DOC_TYPE" , "TRANS_NUM" , "ZZ_CV_NETDT_COR" ,
"WEEK_NUM_ZZ_CV_NETDT_COR" , "ZZ_CV_NETDT" , "NETDT_PLAN" , "VBELN" , "POSNR" ,
"LFDAT" , "EBELN" , "EBELP" , "NETWR" , "WAERS" , "BRWTR" , "ZZ_CV_FREF" ,
"LIFNR" , "SORTL" , "ZTERM" , "TEXT1" , "BUKRS" , "BUTXT" , "WERKS" , "LOEKZ" ,
"POSNR_LIPS" , "EBELP_EKPO" , "ZZ_CV_CUR_STAT"
FROM
"Z079FI_CURR_PO2"
WHERE
"MANDT" = ? AND "TRANS_NUM" IN ( ? , ? , ? , ? , ? , ? , ? , ? , ? , ? , ? , ?
, ? , ? , ? )
Variables
A0(CH,3) = '200'
A1(CH,10) = '0000000001'
A2(CH,10) = '0000000003'
A3(CH,10) = '0000000004'
A4(CH,10) = '0000000005'
A5(CH,10) = '0000000006'
A6(CH,10) = '0000000007'
A7(CH,10) = '0000000008'
A8(CH,10) = '0000000013'
A9(CH,10) = '0000000014'
A10(CH,10) = '0000000015'
A11(CH,10) = '0000000016'
A12(CH,10) = '0000000017'
A13(CH,10) = '0000000018'
A14(CH,10) = '0000000019'
A15(CH,10) = '0000000023'
I used this kind of select statement in my program and it worked - no short dump occured:
SELECT *
FROM z079fi_curr_po2
APPENDING CORRESPONDING FIELDS OF TABLE lt_tra_isub
WHERE trans_num in (
0000000001',
'0000000003',
'0000000004',
'0000000005',
'0000000006',
'0000000007',
'0000000008',
'0000000013',
'0000000014',
'0000000015',
'0000000016',
'0000000017',
'0000000018',
'0000000019' ).
Because I can't rely on FAE (I don't want to use hint &max_in_blocking_factor n& because I'm not sure if n = 5 will not stop working some day), I decided to use:
SELECT *
APPENDING CORRESPONDING FIELDS OF TABLE @lt_tra_isub
FROM z079fi_curr_po2
WHERE trans_num IN ( SELECT DISTINCT trans_num
FROM z079fi_curr_h_i
WHERE bukrs IN @ltr_bukrs
AND status_h IN @s_stath
AND payment_date IN @s_paydat
AND week_num IN @s_weekno
AND waers IN @s_waers
AND werks IN @s_werks ).
But the problem with FAE still exist.
Best regards
Rafal