Copyright	(c) 2021-2022 Mirko Westermeier
License	MIT
Stability	experimental
Safe Haskell	None
Language	Haskell2010

HTMell

Contents

Loading a content tree from filesystem
Accessing single content nodes

Description

Minimal filesystem powered markdown content management system

Synopsis

loadHTMell :: FilePath -> IO (Maybe (HTree MarkdownContent))
loadHTMellContent :: HTMellContent c => FilePath -> IO (Maybe (HTree c))
get :: HTMellContent c => HTree c -> String -> Maybe (HTree c)
getHTML :: HTMellContent c => HTree c -> String -> Maybe Text

Loading a content tree from filesystem

A content tree can be built from any directory, given by a FilePath. Each file or subdirectory in that content directory can become a content node in the resulting tree. Subdirectories can become inner nodes with children, but they can also have content, that is loaded from their "index" child, if it exists. Directory or file names become the node's addresses in the content tree. Leading numbers, followed by "_" are used for sorting in listings, and are stripped from the names, as well as extensions.

Example:

Filesystem:                 Content tree:

content                     Content root node
  |--- 1_foo.md             Content node "/foo"
  |--- index.md             Content of the root node "/"
  '--- 2_bar                Inner node "/bar", no content
        |--- 17_baz.md      Content node "/bar/baz"
        |--- 42_quux.md     Content node "/bar/quux"
        '--- xnorfzt.md     Content node "/bar/xnorfzt"

Caveat: Don't use directories named "index"!

In listings, the nodes below "/bar" will appear in this order, defined by their (optional) ordering numbers:

/bar/xnorfzt: because it's loaded from xnorfzt.md and no number implies 0
/bar/baz: because it's loaded from 17_baz.md and 17 < 42
/bar/quux: because it's loaded from 42_quux.md and 42 > 17

The accepted file extensions are the responsibility of the HTMellContent instance used. The default instance, MarkdownContent, accepts .md, RawHTMLContent reads .html files only.

loadHTMell #

Arguments

:: FilePath	The directory to read the content tree from.
-> IO (Maybe (HTree MarkdownContent))	The loaded content tree, if possible.

Creates a markdown content tree

loadHTMellContent :: HTMellContent c => FilePath -> IO (Maybe (HTree c)) #

Creates a content tree just like loadHTMell, but it works for any HTMellContent instance, for example RawHTMLContent. The type needs to be declared:

maybeTree <- loadHTMell "content" :: IO (Maybe (HTree RawHTMLContent))
let tree = fromJust maybeTree

Accessing single content nodes

get #

Arguments

:: HTMellContent c
=> HTree c	The content tree to search in
-> String	A query string of the desired subtree or content node
-> Maybe (HTree c)	The desired content node, if it exists

Finds a subtree or leaf (content node) in a given content tree. The query String consists of child node names, separated with "/". See loadHTMell for details of node names.

getHTML :: HTMellContent c => HTree c -> String -> Maybe Text #

The same as get, but accessing the content of the desired content node directly.