[x3d-public] Inline, X3DUrlObject allowed to add additional filetypes, etc.

[Brutzman, Donald (Don) (CIV)]

On 7/10/2019 5:37 AM, John Carlson wrote:
> Should the Inline be checked statically or dynamically for recursion?

Yes (inclusive or).

Static:  X3D Schematron rule for Inline can check that name of a file in url field is not self-referential and matching head/meta/title or head/meta/identifier values.  Note that a static test is not rigorous since files are easily renamed.

Dynamic: important, since a recursive pathology (A loads B loads A loads B loads A...) is easily defined.  An X3D player is now required to NOT permit that.  Programming mechanisms can vary but checking can be quite similar to the event-loop-breaking rule which disallows duplicate events from being repeated for a given ROUTE during a single event cascade.

> What if there’s a switch statement in the Inline which inlines the Inline some of the time?

Still not allowed.  Dynamic checking by X3D player is expected to detect and prevent that.

Recursion testing is part of many programming-language implementations.  The X3D spec change is recognizing a common security vulnerability and noting that players must prevent it in order to be conformant.

Separate issue:

> Should we allow fractals/graftals? https://www.youtube.com/watch?v=fXu74k-77kg  How does one propose doing it without Inlines?
Fraunhofer's proposed change to X3Dv4 explicitly allows loading of other 3D representations, rather than forbidding them as exists in X3Dv3.

Note that the draft spec change does not yet say what media/model types are required to be supported.  As indicated on the x3dv4-implementations page, glTF support has been accepted by the working group and such prose is expected to be added.

Good opportunity exists here for encouraging common browser capabilities by offering best-practice open source conversion of .stl .ply etc. to X3D.  Of course such a capability is not novel since many many X3D import/export/conversion tools exist already.  Allowing X3D players to decide whether to support additional formats is good for future progress, and certainly in keeping with the "X" in X3D: Extensible.

> Thanks,
> 
> John

likewise, aloha.

all the best, Don
-- 
Don Brutzman  Naval Postgraduate School, Code USW/Br       brutzman at nps.edu
Watkins 270,  MOVES Institute, Monterey CA 93943-5000 USA   +1.831.656.2149
X3D graphics, virtual worlds, navy robotics http://faculty.nps.edu/brutzman