It's a good idea to create parallel execution with your code. When ? if you want to generate in the same time, the file header and the implementation file, you should use paralleling. However, you will need to tell to the compiler what do you want it writes.

Where to write ?

A new command called create writer name for file could do these things : tell to the compiler to create a new writer into a file named from the dictionary or the scope and to identify it with the name stored in the scope.

When writing in that file ?
You write in a file, that's all. By default, the file written has the same name than the source file name.
A new command can be created : default writer is name
With that command, you just tell to the compiler to write in that writer (just declared before).

Careful : the default writer is only known by the current process and sub-processes. An another process will not write in that file but in the default file.

How ?

On the starting of parallel process, you have a new command :
start in parallel

And into that command, you can add one or more commands, especially a process call.

Each sub-command is processed in parallel with others. And for each command, the scope is the same. This means that when you want to update in parallel your values from a process to an another, you can.

Last edited Aug 1, 2014 at 12:29 PM by skercrow, version 2