replacing MIT's ASN.1 code

[Ezra Peisach]

Kevin Coffman wrote:
> Sorry that I failed to mention this before.  There is another free
> asn.1 compiler/library which has been under development for a while.
> I don't know whether it meets the needs of Kerberos.
>
> http://lionet.info/asn1c/
> _______________________________________________
> krbdev mailing list             krbdev at mit.edu
> https://mailman.mit.edu/mailman/listinfo/krbdev
>   
I have been looking at asn1c as well... A quick examination of the code 
shows that it should be able to handle unsigned vs signed integers with the
high bit set.  Helper function handle conversion of ulong and long to 
INTEGER types, etc.

According to the FAQ - the asn1c compiler is tied to GCC-specific 
extensions - so it might not be so portable - but I suspect that the
generated code will be.

The generated code provides for constraint checking (for integers etc) - 
but I do not know if it has to be explicitly tested for.
I do not know how optional fields are stored - intialized, etc.

There is a ASN_THREAD_SAFE definition.  Looks like it is used for 
GeneralizedTime - in converters to and from struct tm...

Is this code robust enough? I do not know...  a2c is clearly still in 
development - and so is this..

Ezra