Code Blocks and Namespaces

I l@ve RuBoard

Code Blocks and Namespaces Code blocks are pieces of executable Python code that you can think of as collections of statements. Modules, classes, and functions are all examples. Modules may be scripts that execute only once. Since Python is unlike other languages in its lack of a main method, it starts executing code (program text) from the top of the module. To re-execute the module, you have to reload it. Modules can contain classes and functions (both code blocks), which may be executed many times. Also, functions can contain classes as well as other functions, and classes can contain methods as well as other classes. Variables defined in a code block are associated with that code block's namespace. The Main Block Python may not have a main method, but it does have something similar, the main block, which looks like this: if __name__ == "__main__" #Put main code here Modules have a __name__ attribute, which contains the module name (e.g., sys, string, mymodule). The main block is called __main__ and is executed from the command line. Since a module is a code block, anything in it will be executed when it is run from the command line or imported from another module. __name__ thus limits execution to the main module. A namespace is a way to organize variables in a large program. It provides separate contexts for individual variables so that two with the same name can be used without one adversely affecting the other. Think of namespaces as directories on your hard drive. If there are two directories, user1 and user2, and both have a file named readme.txt, you can be sure that these are two different files�one you access as \user1\readme and the other as \user2\readme. Packages provide namespaces for modules and other packages. Say you create a module called sys but sys is also the name of a standard module. This can cause problems. However, if you put your sys module in a package called mycompany, you can access it via mycompany.sys without interfering with the standard sys module. Packages also provide namespaces for other packages, just as file directories provide namespaces for each other. Modules provide namespaces for variables, classes, and functions. You might have a class called stats and download a class from the Web also called stats. You can use both classes in the same program because they're in their own modules�my_module.stats and their_module.stats. Similarly, classes provide namespaces for variables and methods. For example, a method called getMean() in a class called stat won't interfere with a function called getMean() in the same module. We'll discuss namespaces and variable scope more later. The key point here is that Python allows you a lot of flexibility in program organization. It also gives you more control over how you manipulate and introspect your namespace.

I l@ve RuBoard