GTRR - is this a BUG ?

[IdealData]

The attachment should just about explain everything!

The layout I specified in the GTRR step is in the CURRENT FILE, although the table references an EXTERNAL FILE. I am deliberately not using the external files layouts for other reasons.

You can see that the script step in the background shows "Using layout: < unknown >", whereas the edit box has the correct details. Amazingly the script step functions correctly but the ScriptMaker listing is wrong.

I tested a bit further - to my horror it also managed to produce layout names from the external file with no obvious pattern even though I had not specified "Use external table's layouts", but once again the step executes correctly.

I've tested this in FMD 7.03 and FMP 8.01 (Mac OS X.3.9) and get the same results.

In my 10 years developing FM applications I've not seen one like this. Any one got any comments or advice?

Many thanks

GTRR.pdf

Edited March 5, 2006 by Guest

[LaRetta]

In my 10 years developing FM applications I've not seen one like this.

7 has only been around for a few years and this appears to be graph related. When you create a table occurrence based upon an external file, FM will nicely create a layout for it as well (if you don't have one yet). Could this be the unpredictable layout-appearance behavior you describe? You don't need to use external layouts at all if you have an occurrence of it in your graph! You didn't mention it, but did you check the layout itself to be sure it was correctly specificied to an existing table occurrence (of the external file) in your graph? It has tripped me before. Just some ideas ...

LaRetta :wink2:

[comment]

It's a known bug. Since it's purely cosmetic, I'd suggest you move on.

[IdealData]

Thanks. I don't mind the bug so long as I know about it.

Comment, is there a tech article relating to this?

[comment]

I don't know. But if you search here for "+unknown +layout" you should get a couple of threads discussing the same problem.