Re: lp5250d as a windows service?

>
> I know nothing about windows, but when you say, "suppress the GUI" what
> GUI are you talking about?  lp5250d doesn't have a GUI.
>

It has a really simple GUI in Windows, basically just a window where it
tells the last message that it's received from the '400 and a button to
allow you to shut down the program.

>
> > 2) any scs2pdf win32 binaries out there?
>
> scs2pdf should compile if you have a unix environment for windows.  I
> think cygwin provides this.  The reason this is needed is that lp5250d
> uses a pipe to send the data to scs2pdf (at least I think this is how it
> works - maybe Scott did something special for windows?).

The Windows one doesn't use pipes, it has it's own version of scs2ascii
built-in so that it can write the text output straight to the printer.

If you wanted to change it to use a pipe, you probably could.  It would be
difficult to make it generic, however, since there's no "LPR" command to
pipe printouts to.  I suppose you could write one, though.  I seem to
remember that working with pipes in Windows is not a lot of fun...

The unix version of lp5250d may compile & run properly under Cygwin, I
don't know, I don't think I've tried it.

What we did for PDFs here is set up the Windows boxes with PostScript
drivers to print to a regular LPD on a FreeBSD box somewhere.  Then, in
/etc/printcap, we've got a filter set up that takes the input, runs it
through ps2pdf, and throws it in the user's home directory.

You could set up a similar queue that accepts jobs from the iSeries,
converts them to PDF with scs2pdf and throws them in the user's home
directory (assuming that you could figure out which user it came from.)

> > 3) paper length in scs2pdf is fixed - does it make sense to provide a way
> >   to change it? and a way to fix width too, or to switch between different
> >   fixed formats depending on the number of columns (instead of enlarging
> >   by steps the width like it's done currently, if I have read correctly
> >   the code)? And what about changing the character width?
> >   I can patch the program myself, but if it's already been
> >   done, it's unuseful to reinvent the wheel...

Probably should be variables in tn5250rc that control this if we don't
receive the info from the iSeries.  (or, maybe even if we do?)  I'll let
you and James hammer out who is going to change it. :)


> > 4) Am I wrong or in scs2pdf CRs don't do what they should (that is:
> >   carriage return and then writing OVER the previous line)?

I thought they DID do that?!  Isn't that what James is saying in the
following paragraph?

> A CR doesn't necessarily mean "go to the next line", it simply means "go
> to the left edge of the page"

Though, SCS doesn't have a CR per-se, or at least I don't think it does.
It has AVPP or something like that.  (It's been awhile)