Ah - then I misunderstood your earlier post. Sorry about that.
In that case, the script could be easily used for OVDC clients in general by either changing the "S1" to
"MD5" or changing the "insert_token" to "terminal_cid" (Joerg's right, using "MD5" for
the token type is probably best), but there's no easy way to distinguish iPad OVDC clients from PC OVDC clients.
-Bob
On 08/01/11 03:46, Jörg Barfurth wrote:
Bob Doolittle schrieb:
I created the ATI interface for one specific (big!) customer way back when, and
I'd love to see it prove more widely useful. So in the interest of promoting
its use I've written this script for you. Attached.
I'd attach the script to my blog, along with the AMGH examples and How-To I'd
provided, but that was all lost in the move from Sun since I hadn't updated it
recently (so it was presumed inactive/unimportant :-( ).
This script does rely on an assumption that iPad OVDC always creates tokens of the form
"S1.*" and that no other OVDC clients do so. I don't know if this is true,
since I don't have an iPad. Nor do I know if it's expected to be true going into the
future - the token naming is not a documented/public interface so OVDC hasn't provided
any stability guarantee regarding it (AFAIK), and is therefore free to change the naming
convention in future. So caveat emptor and all that.
"S1" is not a token prefix, so your script won't work as is. Like any other client OVDC gets
assigned a "pseudo" token for non-card access. "S1" is the prefix of the model
identifier. Perceiving this as part of some kind of token probably comes from 'utwho -c', which creates a
token-like string from model type identifier and the id part of the terminal CID. AFAICT that combination is
used nowhere else.
Even this use of "S1" can't be used to distinguish iPads from laptops or PCs,
as it is used by all versions of OVDC. I don't know of any insert parameters that allows
distinguishing iPad type clients from PC/laptop clients. Maybe the version (sw) parameter
gives a clue, but even if it does, it probably isn't designed to be parseable.
IIRC the model identification or version parameters aren't readily available to ATI
scripts. The most reliable - and documented, so reasonably stable - way to detect an OVDC
client (versus a DTU) is the "MD5" prefix on the terminal CID.
- Jörg
_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users
