Re: LX source question

Hi Sean,

It's used internally only - the definitions are shipped to customers in DDS format. It simply links in to a lot of the development tools that already exist, so saves a lot of time and re-write of tools used to package and ship the product. As far as I know at the time I left, the new Unicode extension files are defined natively.

There would have been value in adding in that support for a Unicode database if AS/SET was also being used to do the coding as well, which in a lot of ways would have been pretty cool because then there would be the possibility to automatically regenerate a lot of the existing action diagrams into ILE RPG that was able to use Unicode tables and to in fact change large parts of the main database to Unicode, rather than only coding to the Unicode extension files. Note that it's very different RPG code when dealing with Unicode rather than native CCSIDs and many complex problems arise, and this is why the whole Unicode extension file concept came up to start with, due to lack of ADK in 8.3. It solves the 'pain points' where customers most require this support (but it requires a lot of work to develop due to hand coded changes now required).

Many of us 'tech experts' at old SSA Global did not agree that the removal of the development tool set was a wise decision for ease of future maintenance with such a large code base at stake, so you will find I am slightly biased in this regard. But if anyone at Infor wants to hire me on a contract to do design changes to the toolset with the idea of retrofitting to 8.3 or regenerating 8.2, just let me know. :-)

Take care,

Genyphyr Novak
iSeries technical consultant
+33 662 15 02 02


message: 4
date: Fri, 15 May 2009 21:43:20 +0100
from: "McGovern, Sean" <Sean.McGovern@xxxxxxxxxxxx>
subject: Re: [BPCS-L] LX source question

Genyphyr,

I'm
sure I'll have a few questions regarding multi languages over the
coming months when planning the upgrade to LX ! I'm keen to protect the
database from any corruption from invariant characters not being
translated correctly.

When I initially read the documentation
on MLS recently, it seemed to me that not enough had been done. If I
recall correctly, only master file Names, Addresses and Descriptions
are now held in Unicode tables. Seems like a lot more should be
converted to Unicode, Notes for example.

I'm surprised to learn
that AS/SET is still being used to maintain the database definitions.
Was AS/SET upgraded to be able to cope with the MLS Unicode tables ?

Cheers,
Sean

________________________________