Re: Ramblings on "base class" for SV etc.

[Chaim Frenkel]

>>>>> "DS" == Dan Sugalski <[EMAIL PROTECTED]> writes:

>> Er, you still have to think about whether something could be tied and
>> tainted at the same time.

DS> Yeah, that's an issue. We need to hammer out how much is the responsibility 
DS> of the vtable functions, and how much is the responsibility of the op 
DS> functions.

DS> Looks like "tainted" qualifies as the first flag bit in the global flags 
DS> field...

Just a thought.

Have a tainted and untainted vtbl.

Let tainted data have a tainted vtbl. Then 'unsafe' operations simply 
call the data access routine with a 'I don't want tainted data'. Untainted
vtbls ignore the option. Tainted data, abort/scream.

<chaim>
-- 
Chaim Frenkel                                        Nonlinear Knowledge, Inc.
[EMAIL PROTECTED]                                               +1-718-236-0183