Header not coming while exporting Db2 select query results to CSV - db2

Question

I'm trying to execute the below query to export the results to CSV. I'm able to export the data to CSV but the headers were missing in the file. Is there anyway that we can achieve this? Im executing the file in the form "db2 -tvmf D:\Db.sql"
connect to ****** user ***** using ******
export to "D:\Vikas.csv" OF DEL MESSAGES
select
'ROW_NUM',
'DETAIL_TYPE_CD',
'ADMIN_FEES_TICKET',
'ADMINISTRATIVE_FEES',
'BASE_RENT',
'CITATIONS',
'COLLECTION_REPO_FEES',
'DESC',
'EFFECTIVE_DATE',
'LATE_CHARGE',
'MISC_FEE',
'STATUS_CD',
'ROW_ID',
'ROW_ID',
'BUILD',
'REVERSE_FLG',
'NSF_FLG',
'PR_CON_ID',
'PROC_DATE',
'PROPERTY_TAX',
'REGISTRATION_FEES',
'REPAIR_FEES',
'SALES_TAX',
'TERMINATION_FEES',
'TOTAL_TRANS',
'TRANSACTION_TYPE'
from sysibm.sysdummy1
UNION ALL (select
T1.ROW_NUM,
T5.DETAIL_TYPE_CD,
T1.ADMIN_FEES_TICKET,
T1.ADMINISTRATIVE_FEES,
T1.BASE_RENT,
T1.CITATIONS,
T1.COLLECTION_REPO_FEES,
T1.DESC,
T1.EFFECTIVE_DATE,
T1.LATE_CHARGE,
T1.MISC_FEE,
T2.STATUS_CD,
T4.ROW_ID,
T3.ROW_ID,
T2.BUILD,
T1.REVERSE_FLG,
T1.NSF_FLG,
T2.PR_CON_ID,
T1.PROC_DATE,
T1.PROPERTY_TAX,
T1.REGISTRATION_FEES,
T1.REPAIR_FEES,
T1.SALES_TAX,
T1.TERMINATION_FEES,
T1.TOTAL_TRANS,
T1.TRANSACTION_TYPE
FROM
SIEBEL.LSE_INPHIST_VIEW T1
LEFT OUTER JOIN SIEBEL.S_ASSET T2 ON T1.ACCOUNT_NUM = T2.ASSET_NUM
LEFT OUTER JOIN SIEBEL.S_ASSET_CON T3 ON T2.ROW_ID = T3.ASSET_ID AND
T3.RELATION_TYPE_CD = 'Obligor'
LEFT OUTER JOIN SIEBEL.S_ASSETCON_ADDR T4 ON T3.ROW_ID =
T4.ASSET_CON_ID AND T4.USE_TYPE_CD = 'Bill To'
LEFT OUTER JOIN SIEBEL.S_PROD_INT T5 ON T2.PROD_ID = T5.ROW_ID
WHERE
(T1.ACNT_ID = '01003501435'))
ORDER BY
T1.ACNT_ID DESC,T1.PROC_DATE DESC WITH UR
I have included the updated query now in the post.

Answer 1

The Db2-LUW export command lacks the ability to add columns headers to the output file. It only exports whatever is in the SELECT statement.
So when you want to have column-headers in the CSV file you have different options.
One way to do it (when there is no order by) is to make the SELECT statement into a UNION of two queries, the first query returns one row which is the list of column names, then union this with your real query. It means you must hand-craft the column-names of the first query to match the real second query. In your case for example it might look like:
SELECT 'row_num', 'detail_type_cd', ....
from sysibm.sysdummy1
UNION
SELECT t1.ROW_NUM, T5.DETAIL_TYPE_CD, ...
(you have to manually make the column-names , put them in single-quotes etc. But if you want Db2 to work out the column names you can use a technique like here ).
If you have an order by you can run two separate export commands (i.e. no union) outputting to two separate output files, and then use operating system functions to concatenate the output files like this:
export to headers.csv select 'colname1','colname2'...from sysibm.sysdummy1;
export to data.csv select ...
-- for MS-windows
!copy /a headers.csv + data.csv data_with_headers.csv ;
Another (possibly simpler) way to do it , with v11.5 (and higher) versions of Db2-LUW , is to not use the export command, but instead to create an external table, which lets you specify an option includeheader on among many other options for CSV files. You can search this site for examples, and reference the documentation.