[comp.windows.x] PEX question

zrjm03@apctrc.trc.amoco.com (Bob J. Mars) (11/21/89)

It is my understanding that PEX will provide a network-transparent 3D
graphics system.  One of the deficiencies of PHIGS (rightly so with respect
to its design goals) is that it does not allow for "image" (raster) data
display and handling.  However, X does provide a raster data display
capability.

In merging PHIGS with X, will it be possible within the confines of existing
or upcoming ANSI/ISO PHIGS/X standards to underlay, overlay, or mix "image"
data with the 3D graphics provided by PHIGS?  Sun is providing this
capability with an escape function in their version 1.2 release of PHIGS.
Even though this may be a standard way of implementing non-standard features,
it remains non-standard.  Will the PEX "standard" resolve this discrepancy or
will this be a vendor-implementation dependency?
  -Mars-
  All interstellar communication is solely my own.
  zrjm03@apctrc.trc.amoco.com    ..!uunet!apctrc!zrjm03

spencer@eecs.umich.edu (Spencer W. Thomas) (11/28/89)

> In merging PHIGS with X, will it be possible within the confines of existing
> or upcoming ANSI/ISO PHIGS/X standards to underlay, overlay, or mix "image"
> data with the 3D graphics provided by PHIGS?

The "API" (application programmer interface) for (the 3-D part of) PEX
is officially that of PHIGS+.  So there will not be a "PHIGS-like" way
to merge raster data with 3-D data.  If it could get hold of the
drawable that PEX is using, a "PEX aware" application could presumably
use standard X calls to merge image data with 3-D data.  The main
problem would be one of intercepting expose events to the drawable so
that the X data could be redrawn (since presumably this interception
is happening somewhere in the guts of the API).

As far as I know, the ways in which a "PEX aware" application could
interact with a PEX drawable have not been defined yet.  In fact, I
think that the whole issue of combining the an external (to the API) X
event loop with the API internal event processing has not been resolved.

--
=Spencer (spencer@eecs.umich.edu)