See #2, #1385 for context.

This is the docutils parsing, transforms and writing part, comprising

• core,
• PEP HTML Translator (Docutils doctree to HTML)
• PEP Parser (class to provide transforms to Sphinx)
• PEP Role (deals with :PEP:`XX` in RST source)
• PEP Contents (Creates table of contents without page title)
• PEP Footer (Dels with footnotes, link to source, last modified commit)
• PEP Headers (Parses RFC2822 headers)
• PEP Title (Creates PEP Title from headers)
• PEP Zero (Masks email addresses and linkifys PEP numbers)

Note: this is unlikely to successfully build, I am splitting into logical chunks as per Victor's comment.