comp.sys.ibm.as400.misc - IBM AS/400 miscellaneous topics.
Anyone deal with a full SQLRPG Video program calling another SQLRPG Video program without losing the cursor pointers in the first program? Program A is a standalone SQLRPG program which simply displays a subfile. Program A requires a call to another video program (Program B) and Program B in turn is normally a standalone program as well. Program A and Program B both have standard CONNECT RESET and DISCONNECT clauses. When program A calls program B, and then I return back to Program A, it will not scroll down and load additional records. The job logs indicate obviously that the database is not connected and/or the cursor is not open. I know why it's doing it but not necessarily know how to fix it. By calling program B, the database connection and cursor is closed in A and I doing a subfile reload is not really an option because users can change values in the subfile which a reload would wipe out. If someone can point me in the right direction and/or post a link to some type of article, it would really help! Any advice would be greatly appreciated. Eddie
Hi, we have performance problem with a query in a Cobol. The same query is really quick with strsql, but in the SQL cobol, it's very slow. Both are run interactive. Somebody has an idea ? Thanks.
I have a question about this.. For some reason when the Subprocedure calls itself it everything works great except for the fact that the SQL portion is reusing the existing cursor from the original caller. Has anyone seen this happen? The SQL statement gets updated just fine.. (I justified this using debug) The fetch bring back the next row from the original caller. Thanks, Jeff P
I am new to SQL and have been writting some programs to get a taste of SQL. I seem to be running into a wall everytime I mess with a file that has *ISO dates (2007-12-31) that don't have anything in them besides 0001-01-01. It causes an SQL error every time. Our ERP system has all kinds of date *ISO date fields that are initially set to 0001-01-01 until a real date is needed to be stored. Is there some kind of compile option or something to use to stop this error? There must be something stupid I am doing or SQL is never going to work well with our package. Anybody have any ideas?