Short and sweet HTML.

The mel function accepts any number of nodes and returns an HTML string. Each node is a list of the following form:

(TAG [attribute val...] CHILDREN...)

TAG must be a symbol staring with the name of an HTML tag.

The . separator can be used in a tag symbol name to indicate a class. It may be used multiple times. As a special case, if a tag symbol begins with a ., a div tag is implied.

A single # separator can be used to associate an ID with a tag. Note that the separator must be escaped with a \ in elisp. The @ separator is an alias for # which does not need to be escaped.

An optional attribute vector may be added as the second element of a node list. Each attribute must be a symbol (optionally a keyword) followed by its value. The value will be coerced to its string representation.

Any elements of a node specified after the tag and optional attribute vector are the node's children. They may be either strings or nodes.

An htmel file must contain an emacs-lisp program. When evaluated, the return value of the last expression must be a mel spec for a document. For example, the source for this page is stored in ./index.htmel. A mel file is similar to an htmel file, but the return value of each top-level sexp is collected into a list. This is useful for including partial templates within other templates (see below).

Content stored in other files can be included via the mel-read function. The mel-read function can be used to parse and load files into a template.