Project Specification for converting HTML documents to ePub format.

HTML documents and ePub files are similar in many respects: They both use a
markup language to structure their information and use css to style this
information.  However, the conversion process might seem straightforward,
but there will undoubtedly be some difficulties.  First, content in an ebook is
generally read from front to back, but an HTML site is not strictly
linear.  Secondly, an HTML site contains many links to navigate logically
and ePub instead uses a single table of contents and no other links between
pages.  Finally, some elements in HTML are designed to be a fixed width or
height, where ePub documents are supposed to flow text as it is resized.
We will have to make some design decisions and assumptions that attempt to
best solve these conversion issues, but the process will not convert every
page perfectly.

Since HTML and ePub both use a markup language, then we should be
able to use a tree data structure for the entire conversion process.  This
tree structure will benefit if the input is valid xhtml.  Otherwise, there
might be some unspecified behavior that we will do our best to handle.

The process of this program is generally straight-line and should occur in
the following order:

1. Determine order of pages be loading a file describing the files and
table of contents.
2. Create the table of contents and ePub manifest.
3. For each HTML page:
   - Parse the document and create the tree structure.
   - Convert Elements to ePub.
   - Append ePub elements to end of book.
4. Save the manifest and document and then zip of them. Done!