Re: [netrexx-pipelines] Migrate OS/2 pipelines to NetRexx Pipes

[<hp...@we...>]

Hello René!

Am 07.06.2022 um 11:15 schrieb René Jansen:
>> [...]
> I know you are not a fan of non-GA, but in this case you might want to grab what is pointed to at https://www.rexxla.org/blog/netrexx/release404/ <https://www.rexxla.org/blog/netrexx/release404/> , bottom of page.

When I look at the pace NetRexx is updated it makes a big
difference compared to ooRexx which is Beta since 2014-02-24 --
more than eight years now. But changed almost twice a day. A never
ending story. Reassessing a statement from Frank Zappa I could
say, it's not dead but it smells funny.

Because I publish few of my programs made with ooRexx I refuse to
use its Beta version. How could I argue to use a Beta? Beta is
better? Beta is fresh and fine, everything else is frumpy? When I
install NetRexx 4.04 Alpha now it's highly likely when VilMA is
fully migrated either NetRexx 4.04 is also done and out or
finalised within few month only.

>>> addpipe "(sep !) *in:!strip!l!*in:"
>> but the compiler doesn't like the way I put it -- and I can't find
>> sufficing documentation about addpipe.
>
> There is an example of addpipe at examples/pipes/addpipetest2.njp which might help.

There are several examples beginning with addpipetest, numbered
from 1..8, but they all made me wonder about the keywords, the
options, the sequence, where all the dos and donts are documented.

In short: got it working. But... it took quite a while. The only
chance I had without description was molesting and butchering that
silly addpipetest2, trial and error and again, never give up.
Finally it returns what it should. Honestly, I detest results
based on know-how only and absolutely no know-why.
Main obstacles had been: wrong file type, must be .njp instead of
.nrx, no option (sep !) possible, no /* */ bracketed comment on
line with addpipe, and many other showstoppers I can't describe
any more.

The modified proposal for varload now looks like this (it's still
work in progress, no options for now):
> /* p4varload_2.njp: prepare for kind of VARLOAD */
> import org.netrexx.njpipes.pipes.
> class p4varload_2 extends stage
> method run()
> do /* why a DO here, for what? */
>    addpipe (bs) *in:|strip|locate|*in:      -- /* may this cause a dealay? */
>    loop forever
>       line = Rexx peekto()
>       parse line.upper sc +1 vn (sc) sv      /* UPPER to force keyword SYMBOLIC */
> /* sc <> '*' is together with the !strip!l above the default COMMENT "* " */
> /* the plan is to put the output to stack or stem and interpret it line by line */
>       if sc <> '*' then output(vn '=' sv)    /* reduced due to addpipe */
>       readto()
>    end
>    catch StageError	/* now outside the repetitive loop, is that correct? */
>    rc = rc()
> end
> exit(rc*(rc<>12))

It's same like CMS Pipelines, only different.
What could you do with it? Compile it and run it:
> C:\prgm\NetRexx\examples\VilMA>pipc p4varload_2
> addpipe (p4varload_2_bs ) *i_A:|strip|locate|*i_B:
>
> C:\prgm\NetRexx\examples\VilMA>ripe "(sep !) < vilma.ini!p4varload_2!term
> VIMDR = C:\PRGM\REXX\VILMA\
> CTLDR = C:\PRGM\HP-EMULATORS\V41\
> VILDR = C:\PRGM\HP-EMULATORS\VIRTIL\
> DELTA = 3
> ITLVL = STD.VILMA
> CTRLS = VILMA.CONTROLLERS
> TPCPT = TYPICAL.CAPTION
> DIAPOS = 321 23

Result is the same as with first version of p4varload_1 but now
with an addpipe quirk. BTW, the argument for pipe.bat (or ripe.bat
for me) only requires a double quote at the beginning but none at
the end. That's one more know-how issue, free of know-why.

Next steps? No idea. Experts -- please advice.

Best,
M.