Skip to content

lalalic/docx2html

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

44 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

docx2html

docx2html is a javascript converter from docx to html on nodejs and browser. here's a demo.

installation

npm install docx2html

example

const docx2html=require("docx2html")
docx2html(input.files[0])
/** you can do further with utilities in converted html
	.then(html=>{
		//html.toString()
		//html.asZip/download/save
	})
*/

api

  • docx2html(docx, options), return a promise object, options support

    • container: a HTMLElement to append converted html, default value is document.body
    • asImageURL(data): to convert image data to url, only required for nodejs
  • the promise object resolved with an object with following functions

    • content: the converted dom
    • toString(/options:{template(style,body,props), extendScript:}/)
    • asZip(options)
    • download(options)
    • save(options)
    • release(): to release image resources

License

MIT, and I also provide commercial support for tickets and enhancement to pay my rent.

Feature

It is based on docx4js 1.x to parse docx, and utilize docx4js api to traverse docx models and convert docx models to html elements.

Ideally, each docx model should have a specific converter to create accordingly html elements, so the design is simply to map from type of docx model to html element constructor.

While, the difficulty is that some docx models are difficult to be expressed in html. It's luckly that we have CSS3 that make some rich styles possible in html, such as numbering, all(12) kinds of table styles.

Word shape utilizes SVG to draw lines, rects, and etc, but so far it only supports limited shapes, while the left job is time.

P of html, according to HTML specification, is restricted not to include any block container, such as div, so there's no p tag, but all div with paragraph styles, and then do some arrangement when dom is ready with a small javascript code.

It keeps header and footer for every section, but there's no conditional consideration, such as odd and even header/footer.

Word Field is kept, while so far only link is supported.

environment

  • nodejs
  • browser
    • IE9+
    • firefox
    • chrome

model

  • section
  • header
  • footer
  • paragraph
  • link
  • numbering
    • many
  • shape
    • rect
    • circle
    • round rect
  • table
  • textbox
  • inline content
  • heading
    • h1 ~ h6
  • Field
    • hyperlink
  • img

style

  • document default
  • named style
  • section style
    • page layout
    • columns
    • column style
  • paragraph style
  • inline style
  • style inheriance
  • table style
    • all(12) word built in styles
    • styles on first/last/even/odd row/column
    • styles on 4 cornor cells
  • numbering style
  • bullet style
  • shape
    • rotate
    • text direction
    • positioning
      • vertical
        • page/margin - top/bottom/absolute
      • horizontal
        • page
          • left/right/center/inside/outside/absolute
        • margin/leftMargin/RightMargin/inMargin/outMargin/column
          • left/right/center/absolute

ToDo

  • more shapes
  • word art
  • chart