Skip to content

Instantly share code, notes, and snippets.

Embed
What would you like to do?
HTML Table to Markdown Extra converter
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<title>HTML Table to Markdown Extra Table</title>
<style>
* { -moz-box-sizing: border-box; -webkit-box-sizing: border-box; box-sizing: border-box;}
body { font-family: -apple-system, "Segoe UI", Arial, Helvetica, sans-serif; line-height: 1.5;
text-rendering: optimizeLegibility; -webkit-font-smoothing: antialiased; -moz-osx-font-smoothing: grayscale; }
textarea { width: 100%; height: 15em; }
button { font-size: inherit; }
p, li { max-width: 75ch; } /* WCAG 2.0, guideline 1.4.8 */
.two-by { display: grid; grid-gap: 1rem; gap: 1rem; }
.two-by h2, .two-by p { margin: 0; }
code { color: #303; font-weight: bold; font-family: Consolas, Menlo, "Courier New", Courier, monospace; border-radius: 2px; border: 1px solid rgba(0, 0, 0, 0.2); padding: 0 0.2em; background-color: #eee; }
@media (min-width: 40rem) {
.two-by { grid-template-columns: 1fr 1fr; }
}
</style>
</head>
<body>
<h1>HTML Table to Markdown Extra Table</h1>
<p><em>Paste HTML table code into the Input, click the Convert button, and a
<a href="https://michelf.ca/projects/php-markdown/extra/#table">Markdown Extra table</a>
will be placed in the Output. This is meant as a first-pass for table conversion and will
not work for all types of tables.</em></p>
<div class="two-by">
<div>
<h2>Input</h2>
<form method="post" id="form">
<p><textarea name="in" id="in" autofocus placeholder="paste HTML table code here">&lt;table&gt;&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Header 1&lt;/th&gt;
&lt;th&gt;Header 2&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;&lt;a href=&quot;http://example.com/&quot;&gt;Cell 1, 1&lt;/a&gt;&lt;/td&gt;
&lt;td&gt;Cell &lt;em&gt;1, 2&lt;/em&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Cell &lt;code&gt;2, 1&lt;/code&gt;&lt;/td&gt;
&lt;td&gt;Cell &lt;strong&gt;2, 2&lt;/strong&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;img src=&quot;cat.png&quot; alt=&quot;cat pic&quot;&gt;&lt;/td&gt;
&lt;td&gt;&lt;img alt='another cat pic' src=&quot;kitty.gif&quot; width=&quot;300&quot;&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;&lt;/table&gt;</textarea>
<button type="submit" id="submit">Convert</button></p>
</form>
</div>
<div>
<h2>Output</h2>
<p><textarea id="out" placeholder="markdown extra converted table will appear here"></textarea></p>
</div>
</div>
<h2>Notes</h2>
<ul>
<li>Malformed HTML such as missing closing tags will likely produce undesirable results.
Tag attribute values, such as the URL for <code>href</code> are assumed to be quoted
with <code>&quot;</code> or <code>'</code>. For this tool, it only affects <code>href</code>,
<code>src</code>, and <code>alt</code> attributes.</li>
<li>No CSS styling is preserved.</li>
<li>Cell widths in the Markdown are not equalized.</li>
<li>The first row in the table is assumed to be the header row, regardless
of <code>thead</code> and <code>th</code> tags.</li>
<li>Cells containing <code>ul/ol</code>, <code>p</code> are compressed to single
line cells. It's likely that such a complex table would be better served with
different formatting (headings, paragraphs) anyway for
accessibility/readability reasons.</li>
<li>Tables with <code>colspan</code> and <code>rowspan</code> are likely to
produce undesirable results and are beyond the scope of Markdown Extra anyway.</li>
<li><code>img</code> tags are converted to Markdown only if they have a
<code>src</code> and an <code>alt</code> attribute. Why?
<a href="https://www.w3.org/TR/WCAG20/#text-equiv">WCAG 2.0, guideline 1.1</a>.
All other attributes are ignored, including <code>title</code>.</li>
<li><code>a</code>, <code>br</code>, <code>strong</code>, <code>em</code>,
and <code>code</code> tags are converted to Markdown. <strong>All other
tags are discarded.</strong></li>
</ul>
<script>
((window, document) => {
// compliments to http://stackoverflow.com/a/5450113
const repeat = (pattern, count) => {
if (count < 1) return '';
let result = '';
while (count > 1) {
if (count & 1) result += pattern;
count >>= 1, pattern += pattern;
}
return result + pattern;
}
// cache the inputs
const i = document.getElementById('in');
const o = document.getElementById('out');
// bind a submit handler to the form
document.getElementById('form').addEventListener('submit', ev => {
// stop the normal form submit process because we're doing everything here in this function
ev.preventDefault();
// get the input text
let t = i.value;
// only proceed if there is text to work with
if (t.length) {
// now perform all the changes on our t string via the r() function above
t = t.replace(/\t/g, ' '); // convert tabs to a single space
t = t.replace(/\s*[\r\n]\s*/g, ''); // remove lines
t = t.replace(/<\!--[\s\S]*?-->/g, ''); // remove html comments
t = t.replace(/ *<a[^>]* href=(["'])(.*?)\1[^>]*> *(.*?)<\/a>/ig, '[$3]($2)'); // convert anchor tags
t = t.replace(/<\/?strong.*?>/g, '**'); // convert strong to **
t = t.replace(/<\/?em.*?>/g, '_'); // convert em to _
t = t.replace(/<\/?code.*?>/g, '`'); // convert code to `
t = t.replace(/<img[^>]* src=(["'])(.*?)\1[^>]* alt=(["'])(.*?)\3[^>]*>/ig, '![$4]($2)'); // convert images with src, alt
t = t.replace(/<img[^>]* alt=(["'])(.*?)\1[^>]* src=(["'])(.*?)\3[^>]*>/ig, '![$2]($4)'); // convert images with alt, src
t = t.replace(/ *<tr[^>]*>/ig, '\n|'); // build <tr> as "\n|"
t = t.replace(/\s*<t[dh].*?>/ig, ' '); // convert <td> and <th> to a space
t = t.replace(/\s*<\/t[dh]>/ig, ' |'); // build </td> and </th> as " |"
t = t.replace(/&nbsp;/ig, ''); // drop non-breaking spaces
t = t.replace(/&amp;/ig, '&'); // de-entize ampersands
t = t.replace(/<br[^>]*>/ig, '\t'); // temporarily convert BR tags to tabs
t = t.replace(/<\/?[^>]+>/ig, ''); // drop all other tags
t = t.replace(/\t *\|/g, ' |'); // drop cell-ending BRs
t = t.replace(/\s*\t\s*/g, '<br />'); // convert tabs back to BR tags
t = t.replace(/\| {2,}/g, '| '); // tighten spacing after the pipe symbols
t = t.replace(/ {2,}\|/g, ' |'); // tighten spacing before the pipe symbols
t = t.replace(/^ +\|/gm, '|'); // trim line-leading whitespace
t = t.replace(/ {4,}/g, ' '); // convert 4+ spaces to three spaces
t = t.replace(/^\s+|\s+$/g, ''); // trim whitespace
// generate the header row separators
const lines = t.split("\n");
if (lines && lines.length) {
const segments = lines[0].split('|');
let headers = '|';
for (let j = 1; j < segments.length - 1; j++) {
headers += repeat('-', segments[j].length) + '|';
}
// console.log(headers);
t = lines[0] + "\n" + headers + "\n" + lines.slice(1).join("\n");
}
}
// put the new version into the output box
o.value = t;
// clear the old version for quick pasting of new code
i.value = '';
// select all the text in the output box and set the browser focus to the output box
o.select();
o.focus();
});
})(window, document);
</script>
</body>
</html>
<!--
LICENSE: CC0, Public Domain
CHANGE LOG:
2021-02-11, SW:
- stop paying the jQuery tax
- upgrade to ES6 syntax
- swap the CSS to use Grid instead of Flexbox and renamed to .two-by
because… reasons
- allow tick parameter value delimiter support in addition to the default
double quote delimiter; only for href, src, alt
- add some img tags to the sample table
- make other mysterious but minor tweaks to html, notes, and CSS
- add this change log
- add license
-->
@SgtPooki
Copy link

SgtPooki commented Feb 10, 2021

I've been using this for a while and it works great. Thank you!

One thing -- and I may add this if I get time -- it needs the ability to render links as absolute links instead of relative.

@sunnywalker
Copy link
Author

sunnywalker commented Feb 11, 2021

@sunnywalker
Copy link
Author

sunnywalker commented Feb 11, 2021

@sunnywalker
Copy link
Author

sunnywalker commented Feb 11, 2021

The CodePen version of this gist has been updated with these new changes as well.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment