How to use HTML Microdata to JSON-LD Extractor
- Paste HTML with itemscope/itemprop declarations and supported itemtype, itemid and itemref relationships. Existing JSON-LD script blocks are not extracted.
- Set the HTTP(S) base for relative URL values and enter the JSON context object to include. Context values remain local and are never fetched.
- Inspect the updated property map and copyable JSON-LD draft. Check missing references, cycles, duplicate IDs and unattached properties against the source.
- Download the JSON draft, full property CSV and text report. The result table shows at most 200 properties; review semantics and eligibility separately before use.
Example: HTML Microdata to JSON-LD Extractor
Extract an Article with a nested Person, an itemref-supplied topic and a separate Thing from the provided HTML.
Options
- Base address
- Relative supported URL values resolve locally against the entered base. No address or context is requested.
- Item relationships
- Nested items and supported itemref values follow the declared bounded rules. Missing or cyclic references are reported for review.
- HTML source
- Inert HTML up to 200,000 characters. Microdata attributes are read; scripts are never run and resources are never requested.
- Base URL
- Relative itemtype, itemid, href and src values resolve against this address locally.
- JSON-LD context to include
- The entered JSON object supplies context values for the draft. An object without @context is wrapped in it; original whitespace is not retained. No context is fetched.
Supported inputs and limits
- Up to 200,000 HTML characters, 100 items, 2,000 properties, 6,000 nodes and 20 item levels.
- Context JSON is limited to 20,000 characters; a supplied @graph is refused because extracted items supply the graph.
- Only the supported microdata/value subset is read.
- No RDFa or JSON-LD script extraction, remote-context retrieval, full semantic validation or rich-result eligibility check.
Where your input is processed
This tool does not upload your inputs. See the limits for anything saved on your device.
Extraction does not validate a rich result
This conversion reads microdata attributes, not RDFa or existing JSON-LD scripts. It resolves supported relative URLs against the entered base and builds a new JSON-LD graph using the supplied context. A syntactically valid draft can still use unsuitable properties or omit required facts. Review the source map, actual entity meanings and the relevant search feature’s requirements separately.
Questions about HTML Microdata to JSON-LD Extractor
Does it scrape a URL?
It reads pasted HTML without fetching a page.
Will RDFa also be converted?
This extractor handles the supported microdata subset only.
Can I publish the draft without checking it?
Review the property map, omissions and context against the original HTML first.