lxml.etree.XMLSyntaxError: EntityRef: expecting ';': what it means and how to fix it

What the error means

lxml.etree.XMLSyntaxError: EntityRef: expecting ';' means the XML parser found invalid markup. Most XML issues are fixable by correcting the first broken tag/character and re-validating.

Most common real-world causes

• A raw '&' appears in text (needs escaping as &), or an entity is incomplete.
• Mismatched tags (opening/closing tags do not match).
• Unescaped reserved characters (especially & in text).
• Multiple root elements (XML must have exactly one root).
• Invalid characters (control chars, bad Unicode) inside the file.
• Wrong encoding or a stray BOM/prolog mismatch.
• Cut/truncated XML (missing closing tags).

Fast debugging steps

• Search for '&' in the failing area and replace '&' with '&' when it is plain text.
• Validate the first error first. Fixing the first broken tag usually resolves many downstream errors.
• Search around the reported line/column for mismatched tags or a raw '&' that should be '&'.
• Confirm the XML starts with a single root and ends with the matching closing tag.
• If the file comes from a feed/export, try a smaller sample to isolate the first invalid record.

Code example (lxml)

from lxml import etree

try:
    # Use bytes if you're unsure about encoding
    root = etree.fromstring(xml_bytes)
    print(root.tag)
except etree.XMLSyntaxError as e:
    print('XMLSyntaxError:', e)
    # e.error_log can include line/column details
    print('error_log:', e.error_log)

Fix without uploading data

If the XML contains sensitive data, validate and convert locally. No Upload Tools runs 100% in your browser.

• XML to JSON to trigger strict parsing and see the first parser error.
• JSON Validator to validate output if you post-process to JSON.

Workflow: validate -> fix the first error -> validate again -> export/convert.

FAQ

Why do I get different errors in different tools? XML parsers differ. Always fix the first error in the raw input and re-check.

Should I regex-fix XML? Avoid blind regex edits. Validate after each change so you know what you fixed.