CharRef: invalid decimal value

What it means

A numeric reference is & in decimal or & in hexadecimal, and the semicolon is part of the syntax rather than punctuation you can drop. Without it the parser keeps consuming characters as digits until it meets one that is not.

The message says "invalid decimal value" because that is what it was left holding — the digits plus whatever followed them. The defect is the missing terminator, not the number.

What usually causes it

• Writing &#38 or &#160 without the closing semicolon.
• An escaping pass that emitted the reference and a later pass that stripped semicolons.
• Hand-written references copied from documentation that omitted the terminator.
• A truncation that landed mid-reference.

How to fix it

• Close every reference with a semicolon: & and &, never &#38.
• Prefer writing the character directly — the document is UTF-8, so © and — need no reference at all.
• Escape with a library rather than by hand; the semicolon is the part people forget.

The same error elsewhere

Different parsers, same defect. If you arrived with one of these messages, you are in the right place.

• Xerces (Java)

  The reference to entity "#38" must end with the ';' delimiter.

• expat (Python)

  not well-formed (invalid token): line 2, column 6

• .NET System.Xml

  An error occurred while parsing EntityName.