On Sun, 20 Jun 2021, at 20:03, Jeremy Nicoll wrote:
> I have in the past rather than using just numbered elements of stems eg
>
> thing.1 = "The problem here is"
> thing.2 = 21
> thing.3 = 25
>
> typically defined
>
> _errpfx = 1 ; _xcoord = 2 ; _ycoord = 3 ...
I slightly mis-remembered. What I actually did was placed in each stem
that the program used a list of subscript names, eg from the initial
comment where I noted how I was going to use a "fld." stem:
/* Create a table containing field data layout information:
/*
/* fld.0 - the number of defined fields
/*
/* fld.0.1 - list of !xxxx values shared by mainline code and
functions, eg "!pp !qq !rr !ss"
/*
/* fld.n.!xxxx: - properties of the n'th defined field, where "xxxx" can be:
/* !name - before data file is inspected: anticipated name of
field
/* !minw - before data file is inspected: anticipated minimum
width
/* !maxw - before data file is inspected: anticipated maximum
width
/* !strt - in the data file, the starting column
/* !wdth - in the data file, the actual width
then in the mainline code
fld. = 0
fld.0.1 = "!name !minw !maxw !strt !wdth"
do ff = 1 to words(fld.0.1) /* assign values 1, 2, 3... to vars named in
"fld.0.1" */
ffthvnam = word(fld.0.1,ff)
waste = value(ffthvnam,ff)
end
then in a procedure that uses that stem, the whole stem is exposed, and
right after that I repeat the same loop as I had in the mainline code.
--
Jeremy Nicoll - my opinions are my own.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
