Twostep compiler?

[Andre Garzia]

On Mar 21, 2004, at 12:32 PM, A.C.T. wrote:

> Again: Please excuse my potentially "offending" questions. I am trying 
> to understand the technical implications of using Revolution and to 
> find out why some issues seem ... somewhat "complicated", where they 
> do not have to be complicated :-)
>
>

Marc,

I might be steping on eggs here but I think the code is there in plain 
text for a simple reason, our apps can introspect and change their own 
code, or better, change other apps code. If the code was stored like 
bytecodes, we would need a decompilation/translation back into human 
readable code, and that would be anoying from the point of view of 
performance. This way, you can build your own "generators" and 
"pre-processors" tools, and I am not only talking about pretty-print 
tools, but real usefull tools, for example I am working with an concept 
I call "blessed client", for with my revHTTPd plataform, a user can 
easily execute code remotelly thus for increasing security  I will 
inspect a newly-uploaded/created stack and checked if it is "blessed" 
then I'll allow it to run or not. This is only possible because I can 
look into the stack code and search for desired fingerprints that will 
certify that the stack is safe. Checking bytecode is hard, cannot use 
RegEx on them... :D

Also if this is a Issue, I think in the distributuion builder there's 
an option called "Encrypt with password" that will encrypt your stack 
using a nice routine, that will render all code, better, all strings 
encrypted and only with the password you'll be able to decrypt it. 
(encrypted stacks are no way blessed! :D )

I might be wrong, but since you're keen on experimentation please try 
encrypting your stack using the distribution builder and use your 
favorite text/hex editor in it.

Cheers
Andre


-- 
Andre Alves Garzia ð 2004 ð BRAZIL
http://studio.soapdog.org