Re: Using prototypes...

[Vern Hamberg <vhamberg@xxxxxxxxxxx>]

One reason for using different source file names, as far as I know and 
have adopted, is that IBM distribute their examples that way, and these 
are defaulted in the various create commands for different kinds of source.

There are also differences in record length - some really matter. Yes, 
this was once based on the number of columns in Hollerith cards. So we 
had 80 columns for almost everything. Then ILE RPG came along, and we 
have 100 columns for code. C never had a specific length, and I guess 
other types can use longer record lengths and still compile just fine. 
QMQRY needs to be a record length ot 91 according to the documentation, 
QMFORM 162.

One issue for me is, if you have sample code using, say, 132 for record 
length (120 columns), your customer/downloader will have all kinds of 
problems when copying the code to source files that follow IBM's 
suggested standards.

The various source change control products I know also follow IBM's 
suggested standards. You don't HAVE to use those, but they default to that.

But you pays your money and you makes your choice. As far as I'm 
concerned, there ARE valid reasons for using the recommended record 
lengths for different kinds of source code. But there's no requirement 
to agree with that, just don't make YOUR opinion a rule for others.

Later
Vern

hr@xxxxxxxxxxxx wrote:
> I know this is not directly the subject, but I have always
> wondered why  people has different sourcefiles for different
> source types. 
>
> Yes it only takes about 5 sec to go up and change the
> sourcefile name and reposition in the new sourcefile, but
> because you do it so frequently, yoo have probably had done
> it more that 100.000 times wich equals +150 hours of doing
> absolutly nothing productive ;-)
>
> I have only two sourcefiles a QDDSSRC to DB2 definitions
> and a QSRC for the rest and all source is grouped by a
> 4-8 character prefix and the sourcefiles resides in the
> same lib as the objects they correspond.
>
> Programs are organized as ...
>
> PXGENDB2    RPGLE       powerEXT: DB2 RPC Generator BASIC 
> PXGENDB2CL  CLLE        powerEXT: DB2 RPC Generator BASIC 
> PXGENDB2CM  CMD         powerEXT: DB2 RPC Generator BASIC 
> PXGENDB2PX  CLLE        powerEXT: Create All PX-DB2 RPC into PEXTCGI
> PXGENDB2T   SQLRPGLE    powerEXT: DB2 RPC Generator BASIC Template 
> PXGENDB2TR  RPGLE       powerEXT: DB2 Default trigger 
>
> Libraries are organised as ...
>
> E600DTASDV  USR                   ALLIGATOR DB2 Database - Customer Name
> E600LIBSDV  USR                   ALLIGATOR Modified Pgms - Customer Name
> E600STD     USR                   ALLIGATOR Financials Ver. 7.0
> W600DTASDV  USR                   Way2Trade DB2 Database - Customer Name
> W600LIBSDV  USR                   Way2Trade Cgi Programs - Customer Name
> W600TMP     USR                   Way2Trade Temporary Work Files 
> PEXTDTASDV  USR                   powerEXT DB2 Database - Customer Name
> PEXTAFW     USR                   powerEXT Application Framework 1.1 
> PEXTCD2     USR                   powerEXT Core 1.5 - CGIDEV2 Version
> S600DTASDV  USR                   Way2Trade DB2 Database - Customer Name 
> S600STD     USR                   Way2Trade Application Framework Build 
> S600EDI     USR                   Way2Trade XML/EDIFACT API Library 15.0
>
> "SDV" represents a "systemID" and is just a parameter, there could be
> a "TST" for a test environment ect.
>
> Standard libraries can also be changed by adding a suffix, S600STD
> could be S600STDNEW in one setup so I'm able to run quite a diffent
> setup and many customer at the same system.
>
> And because the sourcefiles always follows the object library nobody
> can question where the source is.
>
> Regards
> Henrik