Port FlatBuffers to Python.

Implement code generation and self-contained runtime library for Python.

The test suite verifies:
  - Correctness of generated Python code by comparing output to that of
    the other language ports.
  - The exact bytes in the Builder buffer during many scenarios.
  - Vtable deduplication correctness.
  - Edge cases for table construction, via a fuzzer derived from the Go
    implementation.
  - All code is simultaneously valid in Python 2.6, 2.7, and 3.4.

The test suite includes benchmarks for:
  - Building 'gold' data.
  - Parsing 'gold' data.
  - Deduplicating vtables.

All tests pass on this author's system for the following Python
implementations:
  - CPython 2.6.7
  - CPython 2.7.8
  - CPython 3.4.2
  - PyPy 2.5.0 (CPython 2.7.8 compatible)

docs/source/PythonUsage.md

@@ -0,0 +1,115 @@
+# Use in Python
+
+There's experimental support for reading FlatBuffers in Python. Generate
+code for Python with the `-p` option to `flatc`.
+
+See `py_test.py` for an example. You import the generated code, read a
+FlatBuffer binary file into a `bytearray`, which you pass to the
+`GetRootAsMonster` function:
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.py}
+    import MyGame.Example as example
+    import flatbuffers
+
+    buf = open('monster.dat', 'rb').read()
+    buf = bytearray(buf)
+    monster = example.GetRootAsMonster(buf, 0)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Now you can access values like this:
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.py}
+    hp = monster.Hp()
+    pos = monster.Pos()
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To access vectors you pass an extra index to the
+vector field accessor. Then a second method with the same name suffixed
+by `Length` let's you know the number of elements you can access:
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.py}
+    for i in xrange(monster.InventoryLength()):
+        monster.Inventory(i) # do something here
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You can also construct these buffers in Python using the functions found
+in the generated code, and the FlatBufferBuilder class:
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.py}
+    builder = flatbuffers.NewBuilder(0)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Create strings:
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.py}
+    s = builder.CreateString("MyMonster")
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Create a table with a struct contained therein:
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.py}
+    example.MonsterStart(builder)
+    example.MonsterAddPos(builder, example.CreateVec3(builder, 1.0, 2.0, 3.0, 3.0, 4, 5, 6))
+    example.MonsterAddHp(builder, 80)
+    example.MonsterAddName(builder, str)
+    example.MonsterAddInventory(builder, inv)
+    example.MonsterAddTest_Type(builder, 1)
+    example.MonsterAddTest(builder, mon2)
+    example.MonsterAddTest4(builder, test4s)
+    mon = example.MonsterEnd(builder)
+
+    final_flatbuffer = bulder.Output()
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Unlike C++, Python does not support table creation functions like 'createMonster()'.
+This is to create the buffer without
+using temporary object allocation (since the `Vec3` is an inline component of
+`Monster`, it has to be created right where it is added, whereas the name and
+the inventory are not inline, and **must** be created outside of the table
+creation sequence).
+Structs do have convenient methods that allow you to construct them in one call.
+These also have arguments for nested structs, e.g. if a struct has a field `a`
+and a nested struct field `b` (which has fields `c` and `d`), then the arguments
+will be `a`, `c` and `d`.
+
+Vectors also use this start/end pattern to allow vectors of both scalar types
+and structs:
+
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~{.py}
+    example.MonsterStartInventoryVector(builder, 5)
+    i = 4
+    while i >= 0:
+        builder.PrependByte(byte(i))
+        i -= 1
+
+    inv = builder.EndVector(5)
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+The generated method 'StartInventoryVector' is provided as a convenience
+function which calls 'StartVector' with the correct element size of the vector
+type which in this case is 'ubyte' or 1 byte per vector element.
+You pass the number of elements you want to write.
+You write the elements backwards since the buffer
+is being constructed back to front. Use the correct `Prepend` call for the type,
+or `PrependUOffsetT` for offsets. You then pass `inv` to the corresponding
+`Add` call when you construct the table containing it afterwards.
+
+There are `Prepend` functions for all the scalar types. You use
+`PrependUOffset` for any previously constructed objects (such as other tables,
+strings, vectors). For structs, you use the appropriate `create` function
+in-line, as shown above in the `Monster` example.
+
+Once you're done constructing a buffer, you call `Finish` with the root object
+offset (`mon` in the example above). Your data now resides in Builder.Bytes.
+Important to note is that the real data starts at the index indicated by Head(),
+for Offset() bytes (this is because the buffer is constructed backwards).
+If you wanted to read the buffer right after creating it (using
+`GetRootAsMonster` above), the second argument, instead of `0` would thus
+also be `Head()`.
+
+## Text Parsing
+
+There currently is no support for parsing text (Schema's and JSON) directly
+from Python, though you could use the C++ parser through SWIG or ctypes. Please
+see the C++ documentation for more on text parsing.
+