texme

v1.2.2

Published

2 years ago

Self-rendering Markdown + LaTeX documents

Downloads

475

0High
0Medium
0Low

susam

markdown commonmark tex latex browser node

TeXMe

TeXMe is a lightweight JavaScript utility to create self-rendering Markdown + LaTeX documents.

Get Started

Copy and paste the code below into an HTML file with .html as the extension name:

<!DOCTYPE html><script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

# Euler's Identity

In mathematics, **Euler's identity** is the equality
$$ e^{i \pi} + 1 = 0. $$

## Explanation

Euler's identity is a special case of Euler's formula from complex
analysis, which states that for any real number $ x $,
$$ e^{ix} = \cos x + i \sin x. $$

This file contains one line of HTML code followed by Markdown + LaTeX content.

Open this HTML file with a web browser. It renders itself to look like this: get-started.html. Here is a screenshot of the output:

TeXMe Screenshot

There are three simple rules to remember while using TeXMe:

TeXMe removes any leading and trailing whitespace in the content before rendering the content to HTML.
TeXMe uses the first non-empty line of the content to set the page title if no explicit <title> element is specified. Any leading and trailing whitespace and hash (#) characters are removed while setting the page title.
If there is a Markdown element such as code span/block or image with LaTeX delimiters in it (e.g., $, $$, etc.), TeXMe may interpret it as LaTeX which may lead to incorrect rendering of the document. To prevent this issue, put such Markdown element within a special purpose md environment supported by TeXMe, for example, \begin{md}`echo $foo`\end{md}. If you do not have such Markdown elements with LaTeX delimiters, you may ignore this rule. See the Markdown Priority Environment section to see more details about this. Note: For most documents, you don't have to worry about this point.

If you do not like to start your document with HTML tags, you can write your content first and add the <script> tag in the end but this approach has some limitations. See the Content in Body section for more details about it.

CDN URLs

Use the following URL in the <script> tag to load version 1.2.2 (the current version at this time) of TeXMe:

https://cdn.jsdelivr.net/npm/[email protected]

Use the following URL in the <script> tag to always load the latest version of TeXMe:

https://cdn.jsdelivr.net/npm/texme

If you need something really easy to remember, use this URL to load the latest version of TeXMe:

https://unpkg.com/texme

Valid HTML5

The Get Started section earlier shows how we can create a self-rendering document with a single line of HTML code but this brevity comes at the cost of standard conformance. For example, the required <title> element is missing from the code. Further the <textarea> element is not closed.

For the sake of completeness and correctness, here is a minimal but complete and valid HTML5 example:

<!DOCTYPE html>
<html lang="en">
<title>Notes on Euler's Identity</title>
<script src="https://cdn.jsdelivr.net/npm/[email protected]"></script>
<textarea>

# Euler's Identity

In mathematics, **Euler's identity** is the equality
$$ e^{i \pi} + 1 = 0. $$

## Explanation

Euler's identity is a special case of Euler's formula from complex
analysis, which states that for any real number $ x $,
$$ e^{ix} = \cos x + i \sin x. $$

</textarea>

Here is the output: valid-html5.html.

It has a few more lines of code to ensure that this HTML5 code validates successfully at validator.w3.org. As a result, this example does not look as concise as the one in the previous section.

In case you are wondering, a valid HTML5 document does not require explicit <head>, <body>, or the closing </html> tags, so they have been omitted for the sake of brevity while maintaining completeness and correctness.

In practice though, it is not necessary to write verbose code like this. All browsers follow the robustness principle, so they can render the shorter example in the Get Started section just fine.

Use TeXMe in Web Pages

Style

TeXMe renders the document on a white pane against a gray background by default. This is due to a configuration option named style that is set to 'viewer' by default.

To render the document with a minimal style on a completely plain white background, set the style configuration option to 'plain'. Here is an example:

<!DOCTYPE html>
<script>window.texme = { style: 'plain' }</script>
<script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

# Euler's Identity

In mathematics, **Euler's identity** is the equality
$$ e^{i \pi} + 1 = 0. $$

## Explanation

Euler's identity is a special case of Euler's formula from complex
analysis, which states that for any real number $ x $,
$$ e^{ix} = \cos x + i \sin x. $$

Here is the output: style-plain.html.

To render the document with absolutely no style, set style to 'none'. The 'none' style option is useful to disable the default 'viewer' style set by TeXMe before defining a custom style with regular CSS code. Here is an example:

<!DOCTYPE html>
<script>window.texme = { style: 'none' }</script>
<script src="https://cdn.jsdelivr.net/npm/[email protected]"></script>
<style>
body {
  background: lightcyan;
}
main {
  max-width: 20em;
  padding: 1em;
  border: medium double gray;
  margin: 2em auto;
  background: lightyellow;
}
</style>
<textarea>

# Euler's Identity

In mathematics, **Euler's identity** is the equality
$$ e^{i \pi} + 1 = 0. $$

## Explanation

Euler's identity is a special case of Euler's formula from complex
analysis, which states that for any real number $ x $,
$$ e^{ix} = \cos x + i \sin x. $$

Here is the output: style-custom.html.

Note that the rendered content is displayed within a <main> element inside the <body>. That is why these elements are being styled in the above example.

Render Markdown Without MathJax

To render Markdown-only content without any mathematical content at all, set useMathJax and protectMath options to false:

<!DOCTYPE html>
<script>window.texme = { useMathJax: false, protectMath: false }</script>
<script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

# Atomic Theory

**Atomic theory** is a scientific theory of the nature of matter, which
states that matter is composed of discrete units called *atoms*. It
began as a philosophical concept in ancient Greece and entered the
scientific mainstream in the early 19th century when discoveries in the
field of chemistry showed that matter did indeed behave as if it were
made up of atoms.

Here is the output: markdown-only.html.

Skip Automatic Rendering on Load

When TeXMe loads, it begins rendering the document automatically. This automatic rendering may be skipped by setting renderOnLoad option to false. Here is an example that disables automatic rendering and then invokes rendering later on the click of a button by using the texme.renderPage() function from the TeXMe API:

<!DOCTYPE html>
<script>window.texme = { renderOnLoad: false }</script>
<script src="https://cdn.jsdelivr.net/npm/[email protected]"></script>
<script>
window.onload = function () {
  var button = document.getElementById('button')
  button.onclick = function () {
    button.remove()
    texme.renderPage()
  }
}
</script>
<textarea>

# Euler's Identity

In mathematics, **Euler's identity** is the equality
$$ e^{i \pi} + 1 = 0. $$

## Explanation

Euler's identity is a special case of Euler's formula from complex
analysis, which states that for any real number $ x $,
$$ e^{ix} = \cos x + i \sin x. $$

</textarea>
<div><button id="button">Render</button></div>

Here is the output: skip-render.html.

Set Options After Loading

When we load TeXMe with the <script> tag, it begins rendering the document as soon as it loads. Therefore in the above examples, we define the configuration options prior to loading TeXMe. We do this by defining an object named window.texme with the configuration options defined as properties in this project.

However if we set the renderOnLoad option to false, we prevent TeXMe from rendering the document after it loads. We now have the control to invoke the rendering at a later time, e.g., on the click of a button. In this case, it is possible to set configuration options after loading TeXMe with the texme.setOption() function. This function takes two parameters: option name as a string and option value.

Here is an example that skips automatic rendering on load and sets the style to 'plain' using this function:

<!DOCTYPE html>
<script>window.texme = { renderOnLoad: false }</script>
<script src="https://cdn.jsdelivr.net/npm/[email protected]"></script>
<script>
window.onload = function () {
  var button = document.getElementById('button')
  button.onclick = function () {
    button.remove()
    texme.setOption('style', 'plain')
    texme.renderPage()
  }
}
</script>
<textarea>

# Euler's Identity

In mathematics, **Euler's identity** is the equality
$$ e^{i \pi} + 1 = 0. $$

## Explanation

Euler's identity is a special case of Euler's formula from complex
analysis, which states that for any real number $ x $,
$$ e^{ix} = \cos x + i \sin x. $$

</textarea>
<div><button id="button">Render</button></div>

Here is the output: set-options.html.

Content in Body

If you do not like to start your document with HTML tags, you can write your content first and add the <script> tag in the end like this:

# Euler's Identity

In mathematics, **Euler's identity** is the equality
$$ e^{i \pi} + 1 = 0. $$

## Explanation

Euler's identity is a special case of Euler's formula from complex
analysis, which states that for any real number $ x $,
$$ e^{ix} = \cos x + i \sin x. $$

<script src="https://cdn.jsdelivr.net/npm/[email protected]"></script>

Here is the output: content-in-body.html.

Although, the code looks neater in this example, there is a limitation associated with this form of writing content: Since the content is part of the HTML <body> element (there is no <textarea> element in this code), the content should be written carefully, so that it does not have any HTML syntax error.

Caveats

While using the content-in-body method of using TeXMe, an HTML syntax error in the content can produced mangled output. For example, the following input is not rendered as expected because the content is in the <body> element, so the browser interprets this content as HTML and encounters the beginning of a start tag that is not closed properly:

Here is some unusual code:

```
print('unusual <string')
```

<script src="https://cdn.jsdelivr.net/npm/[email protected]"></script>

Here is the broken output: unusual-code-body.html.

The <string part of the code is interpreted as the opening of a start tag by the browser. What looks like a fragment of Python code to a human ends up being parsed as an HTML tag by the browser that looks like this:

<string') ```="" <script="" src="https://cdn.jsdelivr.net/npm/[email protected]">

This mangled form of the input is then rendered leading to unexpected results. To resolve this, put the content within a <textarea> element (as shown in the very first example in this document). The following input is fine because the content is put inside a <textarea> element.

<!DOCTYPE html><script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

Here is some unusual code:

```
print('unusual <string')
```

Here is the output: unusual-code-textarea.html.

Since the content occurs within the <textarea> element, the browser does not parse it as HTML and therefore does not mangle it.

Use TeXMe as a Library

The examples so far use TeXMe as a utility. The previous examples load TeXMe in a web page and then TeXMe automatically picks the <textarea> element containing Markdown + LaTeX code and renders it as HTML.

In this section, we see how to use TeXMe as a library and invoke its functions. These examples would run as is on Node.js.

Install TeXMe

Enter the following command to install TeXMe:

npm install texme

Render Markdown and LaTeX

Markdown + LaTeX content can be rendered to HTML by simply invoking the texme.render() function. It accepts the Markdown + LaTeX content as a string and returns the rendered HTML as a string. Here is an example:

var texme = require('texme')
var input = '**Foo** $$ {a}_{1} {a}_{2} $$'
var output = texme.render(input)
console.log(output)

The above example produces the following output:

<p><strong>Foo</strong> $$ {a}_{1} {a}_{2} $$</p>

Configuration Options

Here is a quick reference for all the supported configuration options:

useMathJax (true by default): Load MathJax and run it to render LaTeX when set to true. Do not load or run MathJax when set to false.
protectMath (true by default): Prevent Markdown renderer from seeing LaTeX code when set to true. Therefore LaTeX content that may contain text that could be interpreted as Markdown (e.g., $$ {a}_{1} {a}_{2} $$) remains intact as LaTeX when set to true. Let Markdown renderer look for Markdown text within LaTeX code and render it when set to false. For example, $$ {a}_{1} {a}_{2} $$ is rendered as <p>$$ {a}<em>{1} {a}</em>{2} $$</p> when this option is set to false.
style ('viewer' by default): Three values are supported: 'viewer', 'plain', and 'none'. The viewer style displays the rendered document on a white pane against a gray background. The plain style displays the content with a very minimal style that does not change the background style. If set to 'none', no style whatsoever is applied and the document is displayed with the browser's default style.
renderOnLoad (true by default): Begins rendering the document automatically on load when set to true. Skips rendering automatically when set to false.
onRenderPage (undefined by default): A callback function that is automatically invoked after TeXMe completes rendering the page. It is guaranteed that TeXMe has completed rendering the page before invoking this callback. If useMathJax option is true, it is also guaranteed that TeXMe has invoked typesetting LaTeX with MathJax before invoking this callback. However it is not guaranteed that MathJax has completed typesetting the page before this callback is invoked. MathJax typesetting occurs asynchronously and may complete after this callback is invoked. This callback runs only when the texme.renderPage() function runs in web browser either due to automatic rendering on load or due to explicit call to this function.
markdownURL (CDN URL of minified marked.js by default): URL to load marked.js while running in a web browser.
MathJaxURL (CDN URL of MathJax by default): URL to load MathJax while running in a web browser.

Self-Hosting TeXMe

Some users of TeXMe want to know if TeXMe can be hosted on one's own web server such that TeXMe does not load resources from any other web server while rendering a document, i.e., any requests to load resources must be made to the same web server from which TeXMe is loaded. Yes, it is possible to self-host TeXMe in this manner. Here are the steps:

Clone copies of TeXMe and its dependencies to your own server at a location from where you want to serve the files:

git clone https://github.com/susam/texme.git
git clone https://github.com/markedjs/marked.git
git clone https://github.com/mathjax/mathjax.git

Then create a self-rendering document, say, euler.html like this:

<!DOCTYPE html>
<script>
window.texme = {
  markdownURL: 'marked/marked.min.js',
  MathJaxURL: 'mathjax/es5/tex-mml-chtml.js'
}
</script>
<script src="texme/texme.min.js"></script>
<textarea>

# Euler's Identity

In mathematics, **Euler's identity** is the equality
$$ e^{i \pi} + 1 = 0. $$

## Explanation

Euler's identity is a special case of Euler's formula from complex
analysis, which states that for any real number $ x $,
$$ e^{ix} = \cos x + i \sin x. $$

</textarea>

Now, open euler.html with a web browser and it should self-render fine. All resources will be loaded from the local disk.
Now test euler.html by serving it via a web server. Assuming Python 3 is installed, here is one really easy way to test it:
```
python3 -m http.server
```
Then open https://localhost:8000/euler.html using a web server. The network tab in the browser's developer tools should show that all resources are loaded from the same web server and no requests to any other server are made.

Markdown Priority Environment

TeXMe provides a special LaTeX-like environment named md. This is the markdown priority environment. We will see what this term means in the next section. Let us first see what this environment does by looking at a few examples of when this special environment can be useful.

TeXMe introduces the special purpose md environment to protect portions of Markdown content from being interpreted as LaTeX. In most documents, the use of this environment is not required. This environment is useful only in a handful of scenarios where a Markdown element like code span, code block, link, image, etc. may contain content with LaTeX delimiters that may get interpreted as LaTeX by TeXMe thereby leading to a broken rendering of the Markdown element. This environment protects the content of one or more Markdown elements from being interpreted as LaTeX. Let us see a few examples in the next two subsections.

Protect Dollar Sign in Code

The md environment is useful when Markdown code spans or code blocks contain LaTeX delimiters. This environment prevents the content of Markdown code spans and code blocks from being interpreted as LaTeX. Here is an example:

<!DOCTYPE html><script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

# Using Variables

To expand a variable in shell script, prefix the variable name with a
dollar sign. For example:

```
foo=hello
echo $foo
```

The variable `$foo` is substituted with its value, if any, after the
expansion. In the above example, `$foo` expands to the string `hello`,
so the output looks like this:

```
hello
```

The above code fails to render as expected because the TeXMe tokenizer parses out everything between $foo and `$ (inclusive) and interprets it as possible LaTeX code and prevents the Markdown parser from seeing it. As a result, the Markdown parser does not see the triple backticks (```) just after echo $foo and the document gets rendered in an unexpected manner. Here is how the output looks: shell-script-unprotected.html.

A rendering issue like this can be prevented with the use of the markdown priority environment like this:

<!DOCTYPE html><script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

# Using Variables

To expand a variable in shell script, prefix the variable name with a
dollar sign. For example:

\begin{md}
```
foo=hello
echo $foo
```
\end{md}

The variable \begin{md}`$foo`\end{md} is substituted with its value, if
any, after the expansion. In the above example, \begin{md}`$foo`\end{md}
expands to `hello`, so the output looks like this:

```
hello
```

The \begin{md} and \end{md} delimiters create a markdown priority environment that prevents TeXMe from interpreting anything within it as LaTeX. Here is how the output looks now: shell-script-protected.html.

Protect Dollar Sign in Image Description

Here is another example that shows how rendering can break when LaTeX delimiter is found in a Markdown element such as within image description and how the usage of the md environment can fix it. Here is an example:

<!DOCTYPE html><script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

# Metasyntactic Variable

![Screenshot of variable $foo assigned and echoed in shell][1]

The screenshot above shows an example usage of a metasyntactic variable
`$foo` in an interactive shell session.

[1]: https://i.imgur.com/iQx46hd.png

The above input fails to render as expected because the TeXMe tokenizer parses out everything between the first occurrence of $foo within the image description and the next occurrence of `$ (inclusive). As a result, the Markdown parser does not see the closing bracket of the image description and does not recognize the image element. This leads to a broken rendering of the document. Here is how the output looks: img-alt-unprotected.html.

The md environment can be used to fix the rendering like this:

<!DOCTYPE html><script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

# Metasyntactic Variable

\begin{md}
![Screenshot of variable $foo assigned and echoed in shell][1]
\end{md}

The screenshot above shows an example usage of a metasyntactic variable
\begin{md}`$foo`\end{md} in an interactive shell session.

[1]: https://i.imgur.com/iQx46hd.png

Here is how the output looks now: img-alt-protected.html.

Parsing Precedence

In this subsection, we dive a little deeper into what the md environment is. First, we need to understand how TeXMe renders a document. TeXMe performs the following steps while rendering a document:

At first, the tokenizer looks for anything that looks like LaTeX and masks them, that is, it substitutes all LaTeX snippets in the content with mask literal. In case you are curious, the mask literal is ::MASK:: but this detail should not matter to you while using TeXMe.
Then it feeds the masked input to Markdown parser. The Markdown parser cannot see any LaTeX code anymore because they are all masked, so it cannot accidentally render any portion of the LaTeX code as Markdown. The Markdown parser returns a rendered HTML.
The rendered HTML is then unmasked, that is, all mask literals in the rendered HTML are substituted with the original LaTeX snippets.
At this point, TeXMe rendering is complete. Now TeXMe invokes MathJax to render all LaTeX content in the HTML obtained from the previous step.

It is important to note that TeXMe does not implement a Markdown parser of its own. It relies on an existing popular and stable Markdown parser that conforms to the GitHub Flavored Markdown (GFM) specification and has stood the test of time. Note that GFM is a strict superset of CommonMark. TeXMe only parses out content within LaTeX delimiters and masks it, so that the Markdown parser cannot see such content. As a result of this, step 1 can be a problem when there are LaTeX delimiters like $, $$, etc. within a Markdown code span/block. The TeXMe tokenizer interprets the delimiter and the content after it as LaTeX if it finds the corresponding closing delimiter too later in the document. This can break the Markdown rendering of the code span/block. An example of this was discussed in the previous section. This issue occurs because TeXMe parses out and masks the LaTeX snippet before invoking the Markdown parser. The md environment prevents TeXMe from looking for LaTeX content within the environment.

The md environment ensures that anything within \begin{md} and \end{md} is not searched for LaTeX delimiters. Anything within this environment is fed to the Markdown parser intact. This is why this environment is known as the Markdown priority environment.

Unlimited Variants

In the previous two subsections we saw how the Markdown priority environment, that is, the md environment is used and what it does but that is not the entire story. TeXMe provides an unlimited number of variants of the md environment. In fact, any environment name that starts with the string md is a Markdown priority environment, that is, all of \begin{md*}, \begin{md**}, \begin{mdfoo}, \begin{mdbar}, etc. start Markdown priority environments provided the corresponding \end commands also exist. The \end command for a Markdown priority environment must use the exact same environment name as the \begin delimiter.

The availability of unlimited variants of the Markdown priority environment is useful when we have a Markdown code span/block that itself contains code with Markdown priority environment in it such as perhaps a code example that explains how TeXMe works. Consider the following example:

<!DOCTYPE html><script src="https://cdn.jsdelivr.net/npm/[email protected]"></script><textarea>

# Markdown Priority Environment

Here is an example usage of Markdown priority environment:

\begin{md*}
````
\begin{md}
```
foo=hello
echo $foo
```
\end{md}
````
\end{md*}

The above example shows how to protect \begin{md}`$`\end{md},
\begin{md*}`\begin{md}`\end{md*}, and \begin{md*}`\end{md}`\end{md*} in
a Markdown code block.

Here is the output: texme-code-protected.html.

If we start the Markdown priority environment with \begin{md}, then we cannot have \end{md} anywhere within the environment because the first occurrence of it would end the environment. That is why we use \begin{md*} and \end{md*} to create a Markdown priority environment. Now we can safely write \end{md} within it.

In case you are wondering what the quadruple backticks are doing in the above code example, it is a feature defined in the CommonMark specification. It creates a code block within which we can safely use triple backticks. CommonMark allows us to start a code span/block with an arbitrary number of backticks such the code span/block may safely contain consecutive backticks. To be precise a code span that starts with M backticks can safely contain N consecutive backticks as long as M ≠ N. Similarly, a code block that starts with M backticks (M ≥ 3) can safely contain N consecutive backticks as long as M > N. All of this is standard CommonMark and not something introduced by TeXMe. TeXMe only introduces the special purpose md environment and its unlimited variants.

License

This is free and open source software. You can use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of it, under the terms of the MIT License. See LICENSE.md for details.

This software is provided "AS IS", WITHOUT WARRANTY OF ANY KIND, express or implied. See LICENSE.md for details.

Support

To report bugs, suggest improvements, or ask questions, create issues.

Channels

The author of this project hangs out at the following places online:

Website: susam.net
Twitter: @susam
GitHub: @susam
Matrix: #susam:matrix.org
IRC: #susam:libera.chat

You are welcome to subscribe to, follow, or join one or more of the above channels to receive updates from the author or ask questions about this project.

See MathB, a mathematics pastebin built using TeXMe. This is the oldest mathematics pastebin that is still alive on the web and serving its community of users.

See Muboard, a self-rendering and distributable chalkboard built using TeXMe.