Data URL Extractor from HTML

Data URLs (also called data URIs) are a URL scheme defined by RFC 2397 that allows embedding data directly into HTML documents rather than referencing external files. A data URL follows the format: data:[<MIME-type>][;base64],<data>. For example, data:image/png;base64,iVBORw0KGgo... embeds a complete PNG image as a base64-encoded string directly in the HTML.

Data URLs are commonly used for small images (icons, sprites), fonts (woff2), inline SVGs, and HTML email templates where external resources may be blocked. While they eliminate HTTP requests, they add approximately 33% overhead from base64 encoding and cannot be cached independently by the browser. Our data URL extractor from HTML helps you find and analyze all embedded data: URIs, understand their size impact, and decide whether to keep them inline or migrate to external files.

1. 1. Enter HTML: Paste your HTML document (or any text containing data: URIs) into the input panel. The tool scans the content for 14 common HTML patterns where data URLs typically appear: img src (images), link href (icons, stylesheets), CSS url() declarations (backgrounds, fonts), source srcset (responsive images), iframe src (inline frames), script src (embedded scripts), a href (download links), and object data (plugins).
2. 2. Extract & Analyze: For each found data URL, the extractor parses the MIME type (image/png, text/html, application/pdf, etc.), encoding type (base64 vs percent-encoded), calculates the encoded size (the full data URL string), decoded size (actual data after decoding), and estimates the original file size (adjusted for base64 overhead). A MIME type breakdown shows the distribution of content types across all extracted URLs.
3. 3. Review & Export: Browse extracted data URLs in a card layout with color-coded MIME type categories. Each card shows the URL details, size information, and surrounding HTML context. Download individual files with auto-detected file extensions, download all files at once, or copy all data URL strings to your clipboard for further processing.

• Encoded Size: The full byte length of the data URL as it appears in the HTML, including the data: prefix, MIME type declaration, encoding marker (;base64), and the base64 or percent-encoded data. This is the actual contribution to your HTML page weight from this data URL.
• Decoded Size: The actual binary size of the data after decoding from base64 or percent-encoding. For base64 data, this is approximately 75% of the encoded data length (base64 expands binary data by ~33%). For percent-encoded data, this is the decoded UTF-8 string length.
• Estimated Original Size: An approximation of what the file size would be if it were stored as an external file rather than embedded as a data URL. This corrects for the ~33% base64 encoding overhead plus the metadata (MIME type, encoding markers). This helps you estimate the actual data footprint and the savings from externalizing data URLs.
• MIME Type Distribution: The breakdown of content types across all extracted data URLs helps you understand what kinds of data are embedded — images, fonts, scripts, styles, or other content — and their relative contribution to the total embedded data size.

The data URL extractor is 100% free with no signup required. All HTML analysis is performedlocally in your browser — your HTML documents, extracted data URLs, and analysis results never leave your device. There areno file size limits or usage caps. The tool scans 14 different HTML tag patterns, provides MIME type breakdown, encoded/decoded size analysis, individual and batch download capabilities, and works completely offline after the first load. Use it as many times as you need to analyze HTML documents with embedded data URLs.