HTML Minifier

About the HTML Minifier

Shrinking your markup means stripping HTML comments and collapsing redundant whitespace, but the contents of <pre>, <textarea>, <script>, and <style> tags are left completely alone so meaningful spacing and code inside them survive intact. Conditional comments like <!--[if IE]> stick around on purpose, since some browsers still depend on them.

The whole process happens right in your browser, so nothing you paste in ever leaves your machine.

A simpler regex approach, deliberately, unlike the site's HTML Formatter

This tool's sibling, the HTML Formatter, runs a full tag-aware tokenizer that tracks nesting depth and nested elements to reprint markup with correct indentation. Minification is a narrower problem, all it needs to do is protect a handful of special blocks and generically collapse whitespace elsewhere, so this tool uses a more targeted set of regular expressions instead of building a complete parse tree. That's a deliberate scope match: parsing arbitrary nested HTML in general genuinely does need a real tokenizer (regex alone famously can't reliably handle arbitrarily nested structure), but finding "everything between this specific opening tag and its matching closing tag" and then collapsing whitespace around what's left is squarely within what a well-constructed regex can do reliably.

How the backreference keeps mismatched tags from corrupting the output

The regex that protects pre, textarea, script, and style content, <(pre|textarea|script|style)(\s[^>]*)?>[\s\S]*?<\/\1>, captures the tag name in a group and then reuses that exact captured name via \1 when matching the closing tag. That backreference guarantees a <script> block only gets closed by an actual matching </script>, never accidentally by an unrelated </style> that happens to appear nearby, which would otherwise scramble which content gets protected and which gets minified.

Why whitespace between tags collapses to one space, not zero

The tool replaces runs of whitespace between tags with a single space (> <) rather than removing that whitespace entirely (><). This distinction matters for how a page actually renders: HTML treats whitespace between inline elements as visually significant, <span>A</span> <span>B</span> renders with a visible gap between A and B, but stripping that space entirely would visually merge them together with no gap at all. Keeping exactly one space is the safe middle ground, it still removes the bulk of excess formatting whitespace while never accidentally changing how inline content is displayed.

Why conditional comments get rescued before the general comment-stripper runs

Conditional comments (<!--[if IE]> ... <![endif]-->) were a legacy mechanism that let developers target markup specifically at old versions of Internet Explorer, since removed from modern browsers but still present in some codebases maintaining compatibility with older systems. This tool's minifier stashes any conditional comment block safely aside using a dedicated pattern before its general "strip every comment" regex runs, so a genuinely functional conditional comment survives the minification pass intact instead of getting deleted along with ordinary, purely explanatory <!-- notes --> comments.

Two separate cleanup passes for two different kinds of extra whitespace

Horizontal whitespace between tags and vertical whitespace from blank lines are handled as two distinct steps. Inter-tag spacing gets collapsed to a single space as described above, while a separate pass splits the text into lines, trims each one, and discards any line that's now empty, removing the extra vertical blank lines that indentation-heavy source HTML often accumulates. Treating these as separate problems keeps each regex focused and easier to reason about, rather than trying to handle both spacing directions with one overly clever pattern, and it's why the output stays multi-line and readable rather than collapsing into one dense unbroken string.