Operations » Editor » Public Comments » Public Comments Archive

There are two paragraphs that describe lax behaviour for standard rep numbers, it looks like the second is the original and should have been deleted when errata 2.21 was merged and the first created.

If 'lax' and dfdl:textNumberRep is 'standard' then grouping
separators can be omitted, decimal separator can be either '.'
or ',' (as long as this is unambiguous), leading positive sign
can be omitted, all whitespace is treated as zero, and leading
and trailing whitespace is ignored. Also the exponent is also
optional and assumed to be '1' if not supplied.