The validation functions within the 'Check Name, NCName and QName strings' section of the file 'tree.c' do not seem to conform to the W3C XML1.0 (5th) and Namespaces in XML (3rd) editions. This causes xmllint to find false errors in atomic types for non-ASCII IDs and Names, such as those containing Japanese ideographic characters.

It seems that these the validation code, following the 'try_complex:' labels in these functions, are currently based on orphaned definitions. Suggestion is add some new macros to 'parserinternals.h' to cover the current definitions and then modify the 'try_complex:' sub-sections to utilize them e.g.

/**
 * IS_NAMESTARTCHAR:
 * @c: an xmlChar value
 *
 * [4] NameStartChar ::= ":" | [A-Z] | "_" | [a-z] | [#xC0-#xD6] | [#xD8-#xF6] | [#xF8-#x2FF] | [#x370-#x37D] | [#x37F-#x1FFF] | [#x200C-#x200D] | [#x2070-#x218F] | [#x2C00-#x2FEF] | [#x3001-#xD7FF] | [#xF900-#xFDCF] | [#xFDF0-#xFFFD] | [#x10000-#xEFFFF]
 *
 */

#define IS_NAMESTARTCHAR(c) (\
    (0x3a == (c)) ||\
    ((0x41 <= (c)) && ((c) <= 0x5a)) ||\
    (0x5f == (c)) ||\
    ((0x61 <= (c)) && ((c) <= 0x7a)) ||\
    ((0xc0 <= (c)) && ((c) <= 0xd6)) ||\
    ((0xd8 <= (c)) && ((c) <= 0xf6)) ||\
    ((0xf8 <= (c)) && ((c) <= 0x2ff))||\
    ((0x370 <= (c)) && ((c) <= 0x37d)) ||\
    ((0x37f <= (c)) && ((c) <= 0x1fff)) ||\
    ((0x200c <= (c)) && ((c) <= 0x200d)) ||\
    ((0x2070 <= (c)) && ((c) <= 0x218f)) ||\
    ((0x2c00 <= (c)) && ((c) <= 0x2fef)) ||\
    ((0x3001 <= (c)) && ((c) <= 0xd7ff)) ||\
    ((0xf900 <= (c)) && ((c) <= 0xfdcf)) ||\
    ((0xfdf0 <= (c)) && ((c) <= 0xfffd)) ||\
    ((0x10000<= (c)) && ((c) <= 0xeffff))\
)

/**
 * IS_NAMECHAR:
 * @c: an xmlChar value
 *
 * [4a] NameChar ::= NameStartChar | "-" | "." | [0-9] | #xB7 | [#x0300-#x036F] | [#x203F-#x2040]
 *
 */
    
#define IS_NAMECHAR(c)	(\
    (0x2d == (c)) ||\
    (0x2e == (c)) ||\
    ((0x30 <= (c)) && ((c) <= 0x39)) ||\
    ((0x41 <= (c)) && ((c) <= 0x5a)) ||\
    (0x5f == (c)) ||\
    ((0x61 <= (c)) && ((c) <= 0x7a)) ||\
    (0xb7 == (c)) ||\
    ((0xc0 <= (c)) && ((c) <= 0xd6)) ||\
    ((0xd8 <= (c)) && ((c) <= 0xf6)) ||\
    ((0xf8 <= (c)) && ((c) <= 0x2ff))||\
    ((0x300 <= (c)) && ((c) <= 0x36f)) ||\
    ((0x370 <= (c)) && ((c) <= 0x37d)) ||\
    ((0x37f <= (c)) && ((c) <= 0x1fff)) ||\
    ((0x200c <= (c)) && ((c) <= 0x200d)) ||\
    ((0x203f <= (c)) && ((c) <= 0x2040)) ||\
    ((0x2070 <= (c)) && ((c) <= 0x218f)) ||\
    ((0x2c00 <= (c)) && ((c) <= 0x2fef)) ||\
    ((0x3001 <= (c)) && ((c) <= 0xd7ff)) ||\
    ((0xf900 <= (c)) && ((c) <= 0xfdcf)) ||\
    ((0xfdf0 <= (c)) && ((c) <= 0xfffd)) ||\
    ((0x10000<= (c)) && ((c) <= 0xeffff))\
)


and then (in, for example,  xmlValidateName)...

try_complex:
    /*
     * Second check for chars outside the ASCII range
     */
    cur = value;
    c = CUR_SCHAR(cur, l);
    if (space) {
        while (IS_BLANK(c)) {
            cur += l;
            c = CUR_SCHAR(cur, l);
        }
    }
    if (!IS_NAMESTARTCHAR(c) )
        return(1);
    cur += l;
    c = CUR_SCHAR(cur, l);
    while (IS_NAMECHAR(c) || (c == ':')) {
        cur += l;
        c = CUR_SCHAR(cur, l);
    }

    
//    if ((!IS_LETTER(c)) && (c != '_') && (c != ':'))
//        return(1);
//    cur += l;
//    c = CUR_SCHAR(cur, l);    
//    while (IS_LETTER(c) || IS_DIGIT(c) || (c == '.') || (c == ':') ||
//	   (c == '-') || (c == '_') || IS_COMBINING(c) || IS_EXTENDER(c)) {
//	cur += l;
//	c = CUR_SCHAR(cur, l);
//    }
    if (space) {
        while (IS_BLANK(c)) {
            cur += l;
            c = CUR_SCHAR(cur, l);
        }
    }
    if (c != 0)
        return(1);
    return(0);