HTML to Adaptive Card JSON converter library (Demo editor)
The goal of this project is to allow integration with existing WYSIWYG editors such as CKEditor and convert their HTML output to an Adaptive Card.
Under the hood, this project has taken the Turndown code and repurposed it.
- Getting started
- API
- Currently supported HTML tags
- Known caveats
- Integrating with CKEditor
- Building it yourself
You can either install the npm package, directly use a pre-built version of the library, or use a CDN.
npm install adaptive-html
There are pre-built versions of the library for:
- Browser script (iife)
- CommonJS module environments (cjs)
- ES module environments (es)
Available in both minified and unminified formats.
<script src="/adaptive-html/dist/adaptive-html.iife.min.js"></script>
var AdaptiveHtml = require('./adaptive-html/dist/adaptive-html.cjs');
import AdaptiveHtml from './adaptive-html/dist/adaptive-html.es';
<script src="https://cdn.jsdelivr.net/npm/adaptive-html/dist/adaptive-html.iife.min.js"></script>
- toJSON(string | HTMLElement) => Adaptive Card JSON
var adaptiveCardJson = AdaptiveHtml.toJSON(` <p>Turn me into an Adaptive Card</p> `); console.log(JSON.stringify(adaptiveCardJson, null, '\t')); /* JSON returned { "type": "AdaptiveCard", "body": [ { "type": "TextBlock", "text": "Turn me into an Adaptive Card", "wrap": true } ], "actions": [], "version": "1.0" } */
- toHTML(object | string, object) => HTMLElement
- Reconstructs headings (h1 - h6), removes empty nodes, and removes attributes from nodes on top of the standard JSON to HTML conversion done by the adaptivecards library
- The second parameter is optional, but allows you to pass in a few options:
- processMarkdown (function|boolean) - A function accepting a string parameter and returning a string. Will be called for each TextBlock to process it's text and is expected to output compiled markdown or the text itself if no markdown is present
- Default value is
true
(will utilize standard process markdown function of adaptivecards library)
- Default value is
- processNode (function|boolean|object) - A function accepting an HTMLElement representing the current node being processed as the first parameter, another HTMLElement representing the card root as the second parameter, and the options object itself as the third parameter. It is not expected to return anything. Will be called for each element in the HTML output from the adaptivecards library. Allows you to manipulate the HTML output if desired. Will override the default HTML transformations done
- Default value:
{ "removeEmptyNodes": true, "reconstructHeadings": true, "removeAttributes": true }
- options object values:
- removeEmptyNodes (boolean) - whether or not to remove empty nodes in card html output
- reconstructHeadings (boolean) - whether or not to attempt to reconstruct headings from card html output
- removeAttributes (boolean|array) - whether or not to remove attributes or not. Can also pass a custom attribute white list as an array of attribute names to apply
- Default attribute whitelist ['start', 'src', 'href', 'alt']
- hostConfig (object) - An object specifying a HostConfig you desire to use when converting the Adaptive Card JSON to HTML
- Default value:
{ "fontSizes": { "small": 12, "default": 14, "medium": 17, "large": 21, "extraLarge": 26 }, "fontWeights": { "lighter": 200, "default": 400, "bolder": 600 } }
- When this option is set, the whitelisted attributes for removeAttributes is automatically updated to allow the style and class attributes through
- processMarkdown (function|boolean) - A function accepting a string parameter and returning a string. Will be called for each TextBlock to process it's text and is expected to output compiled markdown or the text itself if no markdown is present
- Note: If you want to use this method, you must also include the AdaptiveCards for Javascript library
var adaptiveHtmlOptions = { processNode: { reconstructHeadings: false }, hostConfig: { fontSizes: { small: 14, default: 17, medium: 20, large: 24, extraLarge: 28 } } }; var adaptiveCardElem = AdaptiveHtml.toHTML({ "type": "AdaptiveCard", "body": [ { "type": "TextBlock", "text": "Turn me into an Adaptive Card", "wrap": true, "weight:": "bolder", "size": "extraLarge" } ], "actions": [], "version": "1.0" }, adaptiveHtmlOptions); console.log(adaptiveCardElem.outerHTML); /* HTML returned <div class="ac-container" tabindex="0" style="display: flex; flex-direction: column; justify-content: flex-start; box-sizing: border-box; flex: 0 0 auto; padding: 15px;"> <div style="overflow: hidden; font-family: "Segoe UI", Segoe, "Segoe WP", "Helvetica Neue", Helvetica, sans-serif; text-align: left; font-size: 28px; line-height: 37.24px; color: rgb(0, 0, 0); font-weight: 400; word-wrap: break-word; box-sizing: border-box; flex: 0 0 auto;"> Turn me into an Adaptive Card </div> </div> */
- p
- br
- h1, h2, h3, h4, h5, h6
- ul, ol
- li
- a
- em, i
- strong, b
- img
The default replacement for tags not listed above depends on whether the tag refers to a block or inline level HTML element.
For block level elements, its contents are processed, and wrapped in a Container.
For inline level elements, its contents are processed and simply returned.
- Images in list steps and nested steps are pushed to the bottom of the corresponding list step
- Lists cannot contain headings
If you wish to integrate this with CKEditor it should for the most part work out of the box. However, if you are utilizing the toHTML(object | string) function to take an Adaptive Card JSON and prepopulate the CKEditor instance then you will need one extra configuration setting.
var editorConfig = {
...,
extraAllowedContent: 'ol[start]'
}
The reason this is necessary is such that ordered lists are reconstructed with the correct starting index.
If you wish to build the library yourself then you can follow these steps:
- Clone or download the repository
cd
to the repository directory via the command line/terminal- Run
npm install
to install the necessary dependencies- Note: Make sure you have Node.js installed
- Hack away
- Execute the command
npm run build
- You should now be able to view the built libraries under the
dist/
folder within your copy of the repository
To demonstrate the transformation there is a test client within the repository. To launch it follow these steps:
- Execute the command
cd client && npm install && cd ..
- This will install the test client dependencies and return to repository root
- Execute the command
npm start
- Navigate to http://localhost:3000
You can run tests by executing the command npm test
.
If you want to generate a code coverage report execute the command npm run test:report
. Launch coverage/index.html
in the browser to view the report.