The Page Model

As mentioned in How it works, markdown fragments in source files are parsed into javascript page and fragment objects when the generator is loaded.

The representation in memory is very simple.

generator.pages is a flat array of all page objects in source order.

Each page object has:

_txt:       the unparsed page body text (usually markdown)
_hdr:       the unparsed page header
_fragments: a flat array of page fragments for rendering in order
_href:      the url path for the page
_file:      a reference to a source file object for saving edits
_parent:    the parent page in the page url hierarchy
_children:  an array of child pages in the page url hierarchy
_next:      the next sibling page in the url hierarchy
_prev:      the previous sibling page in the url hierarchy
#*:         #named fragments for rendering fragments by name
*:          named values parsed from the page header

This basic structure, which is derived from a very simple text file format, is sufficient to represent complex pages with lots of fragments, as well as complex hierarchical sites with lots of pages.

And, by breaking pages into fragments, standard markdown can be used inside fragments, without requiring additional markdown extensions to represent arbitrary layout elements like columns and sections.

Generator Plugins for themes or custom sites may manipulate page objects, creating additional properties for alternative content hierarchies or to classify various types of pages.

Plugins can also use the page structure to generate menus, sitemaps, and other navigation.