Python for .NET

Installation
Getting Started
Importing Modules
Using Classes
Fields and Properties
Using Indexers
Using Methods
Delegates and Events
Exception Handling
Using Arrays
Using Collections
COM Components
Type Conversion
Using Assemblies
Embedding Python
License

Python for .NET is a package that gives Python programmers nearly seamless integration with the .NET Common Language Runtime (CLR) and provides a powerful application scripting tool for .NET developers. Using this package you can script .NET applications or build entire applications in Python, using .NET services and components written in any language that targets the CLR (Managed C++, C#, VB, JScript).

Note that this package does not implement Python as a first-class CLR language - it does not produce managed code (IL) from Python code. Rather, it is an integration of the C Python engine with the .NET runtime. This approach allows you to use use CLR services and continue to use existing Python code and C-based extensions while maintaining native execution speeds for Python code. If you are interested in a pure managed-code implementation of the Python language, you should check out the IronPython project, which is in active development.

Python for .NET is currently compatible with Python releases 2.3 and greater. Current releases are available at the Python for .NET website . To subscribe to the Python for .NET mailing list or read the online archives of the list, see the mailing list information page.

Installation

Python for .NET is available as a source release and as a Windows installer for various versions of Python and the common language runtime from the Python for .NET website . On Windows platforms, you can choose to install .NET-awareness into an existing Python installation as well as install Python for .NET as a standalone package.

The source release is a self-contained "private" assembly. Just unzip the package wherever you want it, cd to that directory and run python.exe to start using it. Note that the source release does not include a copy of the CPython runtime, so you will need to have installed Python on your machine before using the source release.

Running on Linux/Mono: preliminary testing shows that PythonNet will run under Mono, though the Mono runtime is not yet complete so there still may be problems. The Python for .NET integration layer is 100% managed code, so there should be no long-term issues under Mono - it should work better and better as the Mono platform matures.

It is not currently possible to *build* PythonNet using only the Mono tools, due to an issue involving the Mono assembler / disassembler. You should, however, be able to try the pre-built assembly under Mono (or compile it yourself with the MS tools and run it under Mono).

Note that if you are running under Mono on a *nix system, you will need to have a compatible version of Python installed. You will also need to create a symbolic link to the copy of libpython2.x.so (in your existing Python installation) in the PythonNet directory. This is needed to ensure that the mono interop dll loader will find it by name. For example:

    ln -s /usr/lib/libpython2.4.so ./python24.so

Getting Started

A key goal for this project has been that Python for .NET should "work just the way you'd expect in Python", except for cases that are .NET specific (in which case the goal is to work "just the way you'd expect in C#").

If you already know Python, you can probably finish this readme and then refer to .NET docs to figure out anything you need to do. Conversely if you are familiar with C# or another .NET language, you probably just need to pick up one of the many good Python books or read the Python tutorial online to get started.

A good way to start is to run python.exe and follow along with the examples in this document. If you get stuck, there are also a number of demos and unit tests located in the source directory of the distribution that can be helpful as examples.

Importing Modules

Python for .NET allows CLR namespaces to be treated essentially as Python packages. The top-level package is named CLR, and acts as the root for accessing all CLR namespaces:

    from CLR.System import String
    import CLR.System as System

Types from any loaded assembly may be imported and used in this manner. The import hook uses "implicit loading" to support automatic loading of assemblies whose names correspond to an imported namespace:

    # This will implicitly load the System.Windows.Forms assembly

    from CLR.System.Windows.Forms import Form

Python for .NET uses the PYTHONPATH (sys.path) to look for assemblies to load, in addition to the usual application base and the GAC. To ensure that you can implicitly import an assembly, put the directory containing the assembly in sys.path.

To load assemblies with names that do not correspond with a namespace, you can use the standard mechanisms provided by the CLR:

    from CLR.System.Reflection import Assembly

    a = Assembly.LoadWithPartialName("SomeAssembly")

    # now we can import namespaces defined in that assembly

    from CLR.SomeNamespace import Something

Note that CLR modules are "lazy". Because a namespace can contain a potentially very large number of classes, reflected CLR classes are created on-demand when they are requested of a CLR module. This means that using the Python dir() function in the interactive interpreter will not necessarily show all of the classes available from a given CLR module (it will only show any classes that have been referenced to that point in time).

It also means that from somemodule import * will not work as expected. It is possible that it could be made to work in the future, but for now it would impose a big performance hit and a lot of complexity to support something that is used relatively rarely and often considered bad form in Python anyway ;)

Using Classes

Python for .NET allows you to use any non-private classes, structs, interfaces, enums or delegates from Python. To create an instance of a managed class, you use the standard instantiation syntax, passing a set of arguments that match one of its public constructors:

    from CLR.System.Drawing import Point

    p = Point(5, 5)

You can also subclass managed classes in Python. See the helloform.py file in the /demo directory of the distribution for a simple Windows Forms example that demonstrates subclassing a managed class.

Fields And Properties

You can get and set fields and properties of CLR objects just as if they were regular attributes:

    from CLR.System import Environment

    name = Environment.MachineName
    Environment.ExitCode = 1

Using Indexers

If a managed object implements one or more indexers, you can call the indexer using standard Python indexing syntax:

    from CLR.System.Collections import Hashtable

    table = Hashtable()
    table["key 1"] = "value 1"

Overloaded indexers are supported, using the same notation one would use in C#:

    items[0, 2]

    items[0, 2, 3]

Using Methods

Methods of CLR objects behave generally like normal Python methods. Static methods may be called either through the class or through an instance of the class. All public and protected methods of CLR objects are accessible to Python:

    from CLR.System import Environment

    drives = Environment.GetLogicalDrives()

It is also possible to call managed methods unbound (passing the instance as the first argument) just as with Python methods. This is most often used to explicitly call methods of a base class.

Note that there is one caveat related to calling unbound methods: it is possible for a managed class to declare a static method and an instance method with the same name. Since it is not possible for the runtime to know the intent when such a method is called unbound, the static method will always be called.

The docstring of CLR a method (__doc__) can be used to view the signature of the method, including overloads if the CLR method is overloaded. You can also use the Python help method to inspect a managed class:

    from CLR.System import Environment

    print Environment.GetFolderPath.__doc__

    help(Environment)

Delegates And Events

Delegates defined in managed code can be implemented in Python. A delegate type can be instantiated and passed a callable Python object to get a delegate instance. The resulting delegate instance is a true managed delegate that will invoke the given Python callable when it is called:

    def my_handler(source, args):
        print 'my_handler called!'

    # instantiate a delegate
    d = AssemblyLoadEventHandler(my_handler)

    # use it as an event handler
    AppDomain.CurrentDomain.AssemblyLoad += d

Multicast delegates can be implemented by adding more callable objects to a delegate instance:

    d += self.method1
    d += self.method2
    d()

Events are treated as first-class objects in Python, and behave in many ways like methods. Python callbacks can be registered with event attributes, and an event can be called to fire the event.

Note that events support a convenience spelling similar to that used in C#. You do not need to pass an explicitly instantiated delegate instance to an event (though you can if you want). Events support the += and -= operators in a way very similar to the C# idiom:

    def handler(source, args):
        print 'my_handler called!'

    # register event handler
    object.SomeEvent += handler

    # unregister event handler
    object.SomeEvent -= handler

    # fire the event
    result = object.SomeEvent(...)

Exception Handling

You can raise and catch managed exceptions just the same as you would pure-Python exceptions:

    from CLR.System import NullReferenceException

    try:
        raise NullReferenceException("aiieee!")
    except NullReferenceException, e:
        print e.Message
        print e.Source

Using Arrays

Managed arrays support the standard Python sequence protocols:

    items = SomeObject.GetArray()

    # Get first item
    v = items[0]
    items[0] = v

    # Get last item
    v = items[-1]
    items[-1] = v

    # Get length
    l = len(items)

    # Containment test
    test = v in items

Multidimensional arrays support indexing using the same notation one would use in C#:

    items[0, 2]

    items[0, 2, 3]

Using Collections

Managed arrays and managed objects that implement the IEnumerable interface can be iterated over using the standard iteration Python idioms:

    domain = System.AppDomain.CurrentDomain

    for item in domain.GetAssemblies():
        name = item.GetName()

Using COM Components

Using Microsoft-provided tools such as aximp.exe and tlbimp.exe, it is possible to generate managed wrappers for COM libraries. After generating such a wrapper, you can use the libraries from Python just like any other managed code.

Note: currently you need to put the generated wrappers in the GAC, in the PythonNet assembly directory or on the PYTHONPATH in order to load them.

Type Conversion

Type conversion under Python for .NET is fairly straightforward - most elemental Python types (string, int, long, etc.) convert automatically to compatible managed equivalents (String, Int32, etc.) and vice-versa. Note that all strings returned from the CLR are returned as unicode.

Types that do not have a logical equivalent in Python are exposed as instances of managed classes or structs (System.Decimal is an example).

The .NET architecture makes a distinction between value types and reference types. Reference types are allocated on the heap, and value types are allocated either on the stack or in-line within an object.

A process called boxing is used in .NET to allow code to treat a value type as if it were a reference type. Boxing causes a separate copy of the value type object to be created on the heap, which then has reference type semantics.

Understanding boxing and the distinction between value types and reference types can be important when using Python for .NET because the Python language has no value type semantics or syntax - in Python "everything is a reference".

Here is a simple example that demonstrates an issue. If you are an experienced C# programmer, you might write the following code:

    items = CLR.System.Array.CreateInstance(Point, 3)
    for i in range(3):
        items[i] = Point(0, 0)

    items[0].X = 1 # won't work!!

While the spelling of items[0].X = 1 is the same in C# and Python, there is an important and subtle semantic difference. In C# (and other compiled-to-IL languages), the compiler knows that Point is a value type and can do the Right Thing here, changing the value in place.

In Python however, "everything's a reference", and there is really no spelling or semantic to allow it to do the right thing dynamically. The specific reason that items[0] itself doesn't change is that when you say items[0], that getitem operation creates a Python object that holds a reference to the object at items[0] via a GCHandle. That causes a ValueType (like Point) to be boxed, so the following setattr (.X = 1) changes the state of the boxed value, not the original unboxed value.

The rule in Python is essentially: "the result of any attribute or item access is a boxed value", and that can be important in how you approach your code.

Because there are no value type semantics or syntax in Python, you may need to modify your approach. To revisit the previous example, we can ensure that the changes we want to make to an array item aren't "lost" by resetting an array member after making changes to it:

    items = CLR.System.Array.CreateInstance(Point, 3)
    for i in range(3):
        items[i] = Point(0, 0)

    # This _will_ work. We get 'item' as a boxed copy of the Point
    # object actually stored in the array. After making our changes
    # we re-set the array item to update the bits in the array.

    item = items[0]
    item.X = 1
    items[0] = item

This is not unlike some of the cases you can find in C# where you have to know about boxing behavior to avoid similar kinds of lost update problems (generally because an implicit boxing happened that was not taken into account in the code).

This is the same thing, just the manifestation is a little different in Python. See the .NET documentation for more details on boxing and the differences between value types and reference types.

Using Assemblies

The Python for .NET runtime uses Assembly.LoadWithPartialName to do name-based imports, which will usually load the most recent version of an assembly that it can find.

The CLR's ability to load different versions of assemblies side-by-side is one of the (relatively few) places where the matching of meta-models between Python and the CLR breaks down (though it is unclear how often this happens in practice).

Because Python import is name-based and unaware of any concept of versioned modules, the design goal has been for name-based implicit assembly loading to do something consistent and reasonable (i.e. load most recent available based on the name).

It is possible to load a specific version of an assembly if you need to using code similar to the following:

    from CLR.System.Reflection import Assembly, AssemblyName

    name = AssemblyName(...) # set required version, etc.
    assembly = Assembly.Load(name)

    # now import namespaces from the loaded assembly

Things get a lot more complicated if you need to load more than one version of a particular assembly (or more likely, you have a dependency on some library the does so). In this case, the names you access via the CLR modules will always come from the first version of the assembly loaded (which will always win in the internals of the Python for .NET runtime).

You can still use particular versions of objects in this case - you just have to do more work to get the right versions of objects:

    from CLR.System.Reflection import Assembly, AssemblyName
    from System import Activator

    name = AssemblyName(...) # get the right version
    assembly = Assembly.Load(name)
    type = assembly.GetType("QualifiedNameOf.TheTypeINeed")
    obj = Activator.CreateInstance(type)

Embedding Python

Note: because Python code running under Python for .NET is inherently unverifiable, it runs totally under the radar of the security infrastructure of the CLR so you should restrict use of the Python assembly to trusted code.

The Python runtime assembly defines a number of public classes that provide a subset of the functionality provided by the Python C API.

These classes include PyObject, PyList, PyDict, etc. The source and the unit tests are currently the only API documentation.. The rhythym is very similar to using Python C++ wrapper solutions such as CXX.

At a very high level, to embed Python in your application you will need to:

Reference Python.Runtime.dll in your build environment
Call PythonEngine.Intialize() to initialize Python
Call PythonEngine.ImportModule(name) to import a module

The module you import can either start working with your managed app environment at the time its imported, or you can explicitly lookup and call objects in a module you import.

For general-purpose information on embedding Python in applications, use www.python.org or Google to find (C) examples. Because Python for .NET is so closely integrated with the managed environment, you will generally be better off importing a module and deferring to Python code as early as possible rather than writing a lot of managed embedding code.

Important Note for embedders: Python is not free-threaded and uses a global interpreter lock to allow multi-threaded applications to interact safely with the Python interpreter. Much more information about this is available in the Python C API documentation on the www.python.org Website.

When embedding Python in a managed application, you have to manage the GIL in just the same way you would when embedding Python in a C or C++ application.

Before interacting with any of the objects or APIs provided by the Python.Runtime namespace, calling code must have acquired the Python global interpreter lock by calling the PythonEngine.AcquireLock method. The only exception to this rule is the PythonEngine.Initialize method, which may be called at startup without having acquired the GIL.

When finished using Python APIs, managed code must call a corresponding PythonEngine.ReleaseLock to release the GIL and allow other threads to use Python.

The AcquireLock and ReleaseLock methods are thin wrappers over the unmanaged PyGILState_Ensure and PyGILState_Release functions from the Python API, and the documentation for those APIs applies to the managed versions.

License

Python for .NET is released under the open source Zope Public License (ZPL). A copy of the ZPL is included in the distribution, or you can find a copy of the ZPL online . Some distributions of this package include a copy of the C Python dlls and standard library, which are covered by the Python license .