"no variable or argument declarations are necessary."

[Brian Quinlan]

Paul Rubin wrote:

 >> Right now, the compiler DOES NOT examine the contents of the other
 >> modules. All it does is generate an IMPORT_NAME instruction which is
 >> evaluation during runtime.
 >
 >
 >
 > In that case the other module gets compiled when the IMPORT_NAME
 > instruction is executed.


If compilation is required to complete the import (i.e. the module has 
not already been imported and the module does not already have a 
compiled byte-code file) then that is the same.

 > That says that compile time and runtime are
 > really the same thing in the current system.


I would say that they are separate systems but that the compilation 
system is available to the runtime (but not vise-versa). In any case, 
the important thing is that the operation of these systems has an impact 
on the optional declaration proposal being discussed.

The thing that I'm trying to understand is how the proponents of such a 
system would change Python to accomodate it. No one has been able to 
articulate that with any decree of specifity ("Just do what Perl does"). 
Also, these (poorly-defined) proposals have revealed a lack of 
understanding of Python's existing workings ("The compiler would examine 
the other module when it processes the import statement, just like it 
does now.").

Without a clear idea of the nature of the proposal, it is impossible to 
assess it's costs and benefits. So could a proponent of optional 
declarations please provide a more clear proposal?

Cheers,
Brian