In yesterday’s post on Octopress, I briefly mentioned the Pygments lexer documentation. One of the things I found incredibly useful was understanding all the available lexers and which language code would invoke them.
For example, let’s say you want to have a code block that highlights both PHP and HTML. How would you do this?
If you simply use the php
language token in your code block, you will get something like this:
<?php $links = $this->getNav(); ?>
<ul id="navigation">
<?php foreach ($links as $l): ?>
<li><a href="<?php echo $l->href; ?>"><?php $l->caption; ?></a></li>
<?php endforeach; ?>
</ul>
Notice how the HTML is not higlighted, but the PHP is. If you used the html
language token, you would get the reverse problem. But, if you search the Pygments lexer documentation, you will find the HtmlPhpLexer
. It first attempts to highlight all the PHP code, and then passes the remaining text to the HtmlLexer
. (Side note: the HtmlLexer
also parses CSS and Javascript). And since we’re reading the docs, there is no need to guess as to which language token will invoke this parser. It specifies the short name as html+php
.
<?php $links = $this->getNav(); ?>
<ul id="navigation">
<?php foreach ($links as $l): ?>
<li><a href="<?php echo $l->href; ?>"><?php $l->caption; ?></a></li>
<?php endforeach; ?>
</ul>
Of course, if you’re using Smarty or some other templating system, you should double-check the Pygments documentation to see if there’s a match. For example, I’m a huge fan of Twig, but in this case, there is no TwigLexer
. Twig is based on a Jinja, however, and there is a HtmlDjangoLexer
with the short name of html+jinja
or html+djanjo
. It works pretty well.
Who knew reading the docs could be so useful?