Thank you Mike, Your suggestion works perfectly!
From: Mike Beckerle <mbecke...@apache.org> Sent: Sunday, August 24, 2025 3:05 PM To: users@daffodil.apache.org Subject: Re: Using UTF-8 in a regular expression? No clue about "option setting". I think that's spurious. I think the notation for the characters you want in an XSD pattern regex in XSD is "A" same as in any XSD string literal. But the maximum number of hex digits is 6, because the highest codepoint value in the Unicode standard is U+10FFFF, so your "\u10FFFFFC" has too many hex digits. But even correcting that, I am not sure you can get away with "􏿼" just because that requires two 16-bit Java JVM characters - a high and a low surrogate, to express, and I don't know that the regex engine is going to support that. Hopefully it does. So your regex should be "[A-􏿼]*" Note that ChatGPT5 seems to think this will work with XercesJ validator (which is the one built into Daffodil for "full" validation). See https://chatgpt.com/share/68ab6236-cd84-800f-801f-a08977e6c4dc On Fri, Aug 22, 2025 at 4:01 PM Mark Kozak <mark.ko...@adeptus-cs.com <mailto:mark.ko...@adeptus-cs.com> > wrote: Hello folks, I am suing version 3.10. When using a pattern restriction as shown below: <element name="value" dfdl:terminator="%NL;"> <simpleType> <restriction base="xs:string"> <pattern value="[\u00000041-\u10FFFFFC]*"/> </restriction> </simpleType> </element> I get the following error: [error] Schema Definition Error: Error loading schema due to InvalidRegex: Pattern value '[\u00000041-\u10FFFFFC]*' is not a valid regular expression. The reported error was: 'This expression is not supported in the current option setting.'. What is the option setting I need to change to make that work? Mark Kozak Director of Engineering Adeptus Cyber Solutions Adeptus-CS.com
smime.p7s
Description: S/MIME cryptographic signature
