Not fully complete, but still a huge time saver. Tested on dozens of files that only needed minor touch ups after conversion.

Installation

npm install -g html2halvalla

Command Line Synopsys

html2halvalla test/simple.html

Example Output (for above usage)

T.doctype TODO
T.html ->
  T.head ->
    T.title 'A simple test page'
    T.style type: 'text/css', '.foo {\n        color: red\n      }'
  T.body '.awesome', ->
    T.div '#root.super.special', ->
      T.comment 'This page is rapidly becoming not-so-simple'
      T.h1 'A simple test page'
      T.p ->
        T.text 'With some awesome text, and a'
        T.a href: 'http://www.google.com', 'link'
        T.text '.'
      T.p '#paragraph_2', ->
        T.text 'And here is an image:'
        T.img src: 'fake/source', title: 'not really'
        T.text 'As well as a disabled select:'
        T.select disabled: 'disabled', ->
          T.option 'Oh boy!'
      T.script type: 'text/javascript', 'console.log("Hello there");\n        console.log("How\'s it going?");'
      T.span()

Full Command Line Usage

html2halvalla [options] <html-file>

--prefix=<string>    Prepends a string to each element function call default 'T.'
--s=slug             creates output directory for contents
--b=breakoutId       break tag with id to separate havalla source
--m=matchtext        exact match for text in html -- inserts 'id="b-value' in html tag
--e=extractList      extract repeated tags by tagname and classname RE -- inserts 'id=b-value_0123455'
--selfTest           append code to render the halvalla template for test

--

See "Supported options" below for additional details.

internal API

convert(html, stream, [options], [callback])

html must be a string.

stream is a "Writable Stream".

options is an optional hash. See next section for details.

callback is optional and passed (error) if something goes wrong.

Supported options:

• prefix prepends a string to the begining of each element functional call. (default: 'T.')

For example, using the prefix @ would result in @body ->.

• b is a string for the breakoutId. "Break Out" this tag ID element and it's children as a separate file. The HTML tag with that id is directed to a new halvalla source class and file in the slug directory

example '--b=bloviation' will direct a <tag ID='bloviation' ... to a new file in the slug directory wrapped in a class statement

• m is a string that matches the original html text and inserts the breakoutId into a tag that can only be doescribed by it's actual text of classnames.

the output files test/footer.coffee and test/header.coffee are examples.

• e is a RE of the form 'tagname..class-to-find..find-this-class-too' and inserts the matching

html elements into the Break Out file. Matched classes will have unique tag names based on the breakoutId. see the file test/blov/dicey.coffee for an example extracted from test/bs.html.

Example REPL Session

• bin/html2halvalla --b=header --m='<header' --b=footer --m='<footer' --b=bloviation --m='div class="large-8 columns"' --s=blov test/bs.html

Matches the '<header' text in the html and replace it with `<header id="#{option b.value}" <header id=header

And will decode an html file into four separate files named html.coffee, header.coffee, footer.coffee and bloviation.coffee (If the id for 'bloviation' already exists in the html file, the '-m' option is not needed.)