http://www.adobe.com/devnet/opentype/afdko/topic_feature_file_syntax.html#10
For example, @mhosken is interested in experimenting with inlining
custom syntax (such as Python snippets) into feature files. After this
change, such experiments can be done on top of feaLib because the
Abstract Syntax Tree now contains the tag and content of `anonymous`
blocks.
Since py23 modifies some essential builtins, it's safe to import
everything all the time. At least, that's how it was designed.
It's a bug if importing * breaks some code.
This simplifies the public API to the library. For clients, it does
not matter which exact component was detecting an error. And we will
soon have more components; there would be little point in declaring
CompilerError, TableBuilderError, and so forth.
The OpenType Feature File specification does not specify hex numbers
as explicit token type, but they appear several times in the examples
given by the spec.
When the lexer encounters the "include" keyword, it now enters
a special mode for scanning file names. After having scanned over
the file name, the lexer goes back to normal. The exact format
of file name strings is not defined by the OpenType feature file
specification, so we accept any character that is not a closing
parenthesis.
This simplifies the implementation of the parser for
OpenType feature files, since it can now just keep
token locations returned by the lexer. Before this
change, the parser had to un-pack the location tuples
and build new tuples that included the file path.
