Data Structure arrays are cool, but have some limitations

 

The original contents of this page have become obsolete, go to this page for up-to-date information.

 

End of Support for IBM i 6.1

Today, September 30 2015, marks the end of support for IBM i 6.1. It became available March 21 2008 and has been the second longest supported release of the operating system, for 7 years 6 months and 9 days. You can see a list of all the operating system releases with their dates here.

If you are still using 6.1 you need to think about moving to a later release, as further upgrades and PTFs not available for you.

IBM's notification of the end of support for 6.1 can be found here.

This leaves 7.1 and 7.2 as the only supported releases of IBM i.

Build Views and Views of Views

Have you ever had one of those "Aha!" moments when, as we say in England, the penny drops when someone says something and you are left thinking: "Why didn't I think of that?" I had one of those moments at the OCEAN user group technical conference during a presentation by Paul Tuohy on embedded SQL in RPG.

I have described before what SQL Views are. If you look at the object attribute it is a "LF", but it is not Logical file. Confused?

New TRs coming soon for IBM i ?

I was surfing through IBM's developerWorks website looking for information about a Technology Refresh, TR, when I noticed that pages for IBM i 7.1 TR11 and 7.2 TR3 are now on that website.

More proof that RPGIII is done

On Friday I was asked if I could help with a compilation error for a RPGIII program. When the source was compiled it produced a CPF6301 message, see below, and the programmer could not understand what it meant or how to fix it.

Message . . . . : Program PGM1 in library LIB1 is not created as CPF6301 received. Cause . . . . . : Compile terminated in phase QRGRT at MI instruction number '00000'X with message CPF6301. Text for message is: ERROR IN COMPILER. COMPILE TERMINATED.

When %LOOKUP *NE LOOKUP

 

The original contents of this page have become obsolete, go to this page for up-to-date information.

 

Putting the SQL options into the source

For many releases of IBM i we have been able to put keyword in the H-spec/Control options that are the same as the parameters in RPG's compile options, see here. The same is also available in SQL for RPG objects with embedded SQL, SQLRPGLE.

I have given some examples in the past of using the SET OPTION statement in SQL to turn off commitment control. There are a multitude of other options that mirror options in the SQLRPGLE compile commands. I am not going to list all of the options in this post as IBM provides them all on their web site, there is a link at the bottom of this article to the page. I am only going to show what I think are the most useful.

Video: Node js, DB2, and RPG talking at last

This video was produced by COMMON Europe of a presentation given by Aaron Bartell on how Node.js can interface with DB2 and RPG on the IBM i.

Having a timeout on screen

The germ for this post came from a comment made by Glenn Gundermann on last week's Display screens of results without having to press Enter:

How would you do it for the times you don't want to lock the keyboard? You might want to have a screen being displayed to allow the user to do actions but if they don't, perform an action after a timeout.

After a bit of playing I have a solution that will have a screen time out if someone does not press a key in a certain amount of time. This is just simple example to demonstrate the method I found. In the example the user will be presented with a screen they can either press Enter, F3 to exit, or do nothing. If they do nothing after two seconds the program will proceed and display a second screen.

Display screens of results without having to press Enter

I received a message asking me if there was an easy way in RPG to have a screen constantly update without the user needing to press a key. The idea is that as the program progresses it would write a message to the screen to inform the user of what is happening. Fortunately there is, otherwise I would not have be written this post.

First we need to understand what RPG's operation code Execute Format, EXFMT, does. Simply put it is a combination of a Write operation of the record format to the workstation, followed by a Read operation to retrieve the input from the record format. I can easily replace the EXFMT in any program with a WRITE followed by a READ. If I want to just write a display file's record format to the screen I could just perform a write, but how can I stop the user for using the keyboard while the record format is displayed?

Speeding up CPYF, with selection criteria

Charlie Despres posed a good question in a comment to my original post Speeding up CPYF, "Does this still work if I add conditions to the CPYF?" I assumed he would see the same benefit, but there is only one way to really know, test it!

I used the same file, BIGFILE, and only copy records into the output file, QTEMP/@BIGFILE, where the value of the field FLD1 is in the range of 1,000 to 2,000. There are 12 records in the 1 million records in BIGFILE that fit the criteria.

The only difference from the original programs shown in my last post was the Copy File command, CPYF, statements. This statement, with the default of FROMRCD(*START), was placed in one program:

Speeding up CPYF

Every developer type person who has worked on the IBM i, or its predecessors, has had to use the Copy File command, CPYF, to copy data from one file to another. Most have rarely used more that the first screen's parameters, thereby missing a way to make the command execute faster. For small files the difference is negligible, but in a file of several 100,000 records or more the method I describe below will make a noticeable difference.

I have to thank Stu Haddock for reminding me of this. It is all down to the value of the FROMRCD parameter of the CPYF command.