Larry Byler wrote:

> Jeff's question show a bit of confusion about what happens when.  The
> treatment of ENV files when the spool file is created depends *entirely*
> on what kind of ENV file the file system is dealing with.  It has nothing
> to do with the target device, about which no assumptions are made at spool
> file creation time because it could change before printing.  So, PENV
> files are processed by one routine, TTYPE files by another, and so on.
> At the bottom of the sieve are network printing ENV files; it is these
> files that must have a filecode of 0 <= n <= 1023 to be accepted.

I've obviously been ambiguous in my points.  Let me retry:
* Network printer drivers ignore TTYPE files (and perhaps true PENV files).
* Network printer drivers do page level recovery by PJL *only*.
* Formation/etc packages that process special PENV/TTYPE files do so by
  inserting raw PCL (Formation does at FOPEN time, ESPUL does at output
  time <but that's not a netprint issue>).  If you use output reports that
  start with %61 <top-of-form> CCTL, you will get an extra blank page if
  there is *any* leading setup information in the spoofle.
* Network printing ENV files are the only things network print drivers will
  insert "silently" before the user data (and in most cases handle the
  initial form-feed suppression like the serial printer drivers).

Jeff Kell <[log in to unmask]>