Safe document
The parser rejects DTD and unsafe XML, excessive nesting, unsupported content and documents that cannot be parsed.
Separate XML syntax blockers from product-data problems and see whether a readable document contains complete, consistently represented product records.
The parser rejects DTD and unsafe XML, excessive nesting, unsupported content and documents that cannot be parsed.
BOM, HTTP charset, XML declaration and byte signatures are compared before decoding.
Common product, item, entry and offer nodes are processed incrementally within depth and element limits.
Namespaced local names are normalized to IDs, content, links, images, price, availability, brand, identifiers and category.
Parsed products receive the same deterministic rules as supported CSV records.
Controlled UTF-8 recovery is visible in the scan and report; it never silently hides source damage.
<item>
<id>123</id>
<title>Running Shoes
<price>99.00 EUR</price>
</item>The title element is never closed. The following price node is nested incorrectly, so the document cannot be parsed reliably and the scan returns an XML error.
<item>
<id>123</id>
<title>Running Shoes</title>
<price>99.00 EUR</price>
</item>The fragment is well formed, but a complete product may still be missing description, link, primary image, availability, brand, GTIN/MPN and category fields checked by the audit.
| Question | Syntax and parse validation | Product-data audit |
|---|---|---|
| Document | Is it safe and well formed? | How many products were actually read? |
| Structure | Are tags nested and closed? | Are supported fields present across products? |
| Values | Can text be decoded? | Do price, availability and URLs match current rule formats? |
| Result | Accept or reject the document | Score, issue groups and affected rows |
Product Feed Scan parses public XML and CSV feeds incrementally, applies versioned rules to supported fields and groups failures by impact. It does not predict platform policy decisions, modify feeds or guarantee channel approval.
The parser recognizes common repeated product, item, entry and offer elements and normalizes supported local field names, including namespaced forms.
No. Unsafe XML constructs are rejected to preserve DTD and XXE protections.
Only isolated invalid sequences in explicitly identified UTF-8 XML may be replaced with U+FFFD within a configured limit. Recovery is always reported; systematic damage or broken XML still fails.
No. Well-formed XML can still contain missing IDs, short content, malformed prices, unsupported availability or incomplete categories.
Scan a public XML or CSV feed, review the free score and unlock the complete private report only if you need every affected product.