Re: RUNSQLSTM Source File Record Length

eftimios pantzopoulos wrote:

I'm generating SQL statements to a source file in order to run that source file member through RUNSQLSTM. The SQL statements are about 150 characters long so I created a source physical file with a record length of 200 thinking that this would solve my problem. My SQL statements were now complete but when I ran the source member through RUNSQLSTM it seemed to ignore everything past character 79! What documentation I could find said the same thing about QM QRY.

Mike:

In V5R4, a documentation reference is here:

http://publib.boulder.ibm.com/infocenter/iseries/v5r4/index.jsp?topic=/sqlp/rbafyrunstat.htm
or
http://tinyurl.com/24egkz

It isn't for RUNSQLSTM, but rather for the "SQL statement processor" which is invoked by RUNSQLSTM. I see it documented back as far as V4R3 in the SQL Programming manual.

I also see that the docs are saying 80 characters, but you say 79. I've known the limit was 79 for QM SQL, but never checked RUNSQLSTM. I'll take your word that position 80 is dropped by RUNSQLSTM -- this should be reported to IBM.

I'm using SQL to insert into the source physical file. Is there an easy way to generate each SQL statement so that it runs over two lines without resorting to another pass over the resulting member with another program?

AFAIK, there is no continuation syntax other than breaking at white-space or simply putting the first 80 chars in the first source record, the next 80 in the next record, and so forth. I.e., just break it into 80-char substrings and don't worry about it -- it has worked for me. The breaks can be in the middle of a word as long as the first part of the word ends in position 80 and the rest of the word begins in position 1 of the next source record.

Tom Liotta