>It certainly makes sense to me that a [JP]ythonInterpreter would have its own set of modules, etc., and as Guy pointed out, there are definite valid applications of such a construction. I wonder if we shouldn't push to evolve in that direction. The 1-1 correlation of Interpreter to Process seems artificially limiting.
>The only relevant information I could find in the Python Language Reference was in 6.12 The import statement:
>...The system maintains a table of modules that have been initialized, indexed by module name. This table is accessible as sys.modules...
>Which leaves the system == interpreter instance as a valid interpretation.
>Is there more information somewhere else?
There is also the __builtins__/__builtin__ stuff. This is related to
>Any other opinions on the utility & difficulty of separating the Interpreter instance and Process constructs?
In line with Matt Conway's suggestion it might be possible to write
a java function that starts a function on a new thread using a new
interpreter loaded with a new class loader.
It could be in 'new' module under the name 'jinterpreter'.
Have good night,
Get latest updates about Open Source Projects, Conferences and News.