in jena-arq/src/main/java/org/apache/jena/ext/xerces_regex/RX_Token.java [735:967]
static protected RangeToken getRange(String name, boolean positive) {
if (RX_Token.categories.size() == 0) {
synchronized (RX_Token.categories) {
RX_Token[] ranges = new RX_Token[RX_Token.categoryNames.length];
for (int i = 0; i < ranges.length; i ++) {
ranges[i] = RX_Token.createRange();
}
int type;
for (int i = 0; i < 0x10000; i ++) {
type = Character.getType((char)i);
if (type == Character.START_PUNCTUATION ||
type == Character.END_PUNCTUATION) {
//build table of Pi values
if (i == 0x00AB || i == 0x2018 || i == 0x201B || i == 0x201C ||
i == 0x201F || i == 0x2039) {
type = CHAR_INIT_QUOTE;
}
//build table of Pf values
if (i == 0x00BB || i == 0x2019 || i == 0x201D || i == 0x203A ) {
type = CHAR_FINAL_QUOTE;
}
}
ranges[type].addRange(i, i);
switch (type) {
case Character.UPPERCASE_LETTER:
case Character.LOWERCASE_LETTER:
case Character.TITLECASE_LETTER:
case Character.MODIFIER_LETTER:
case Character.OTHER_LETTER:
type = CHAR_LETTER;
break;
case Character.NON_SPACING_MARK:
case Character.COMBINING_SPACING_MARK:
case Character.ENCLOSING_MARK:
type = CHAR_MARK;
break;
case Character.DECIMAL_DIGIT_NUMBER:
case Character.LETTER_NUMBER:
case Character.OTHER_NUMBER:
type = CHAR_NUMBER;
break;
case Character.SPACE_SEPARATOR:
case Character.LINE_SEPARATOR:
case Character.PARAGRAPH_SEPARATOR:
type = CHAR_SEPARATOR;
break;
case Character.CONTROL:
case Character.FORMAT:
case Character.SURROGATE:
case Character.PRIVATE_USE:
case Character.UNASSIGNED:
type = CHAR_OTHER;
break;
case Character.CONNECTOR_PUNCTUATION:
case Character.DASH_PUNCTUATION:
case Character.START_PUNCTUATION:
case Character.END_PUNCTUATION:
case CHAR_INIT_QUOTE:
case CHAR_FINAL_QUOTE:
case Character.OTHER_PUNCTUATION:
type = CHAR_PUNCTUATION;
break;
case Character.MATH_SYMBOL:
case Character.CURRENCY_SYMBOL:
case Character.MODIFIER_SYMBOL:
case Character.OTHER_SYMBOL:
type = CHAR_SYMBOL;
break;
default:
throw new RuntimeException(RX_Token.class.getName()+"#getRange(): Unknown Unicode category: "+type);
}
ranges[type].addRange(i, i);
} // for all characters
ranges[Character.UNASSIGNED].addRange(0x10000, RX_Token.UTF16_MAX);
for (int i = 0; i < ranges.length; i ++) {
if (RX_Token.categoryNames[i] != null) {
if (i == Character.UNASSIGNED) { // Unassigned
ranges[i].addRange(0x10000, RX_Token.UTF16_MAX);
}
RX_Token.categories.put(RX_Token.categoryNames[i], ranges[i]);
RX_Token.categories2.put(RX_Token.categoryNames[i],
RX_Token.complementRanges(ranges[i]));
}
}
//REVISIT: do we really need to support block names as in Unicode 3.1
// or we can just create all the names in IsBLOCKNAME format (XML Schema REC)?
//
StringBuilder buffer = new StringBuilder(50);
for (int i = 0; i < RX_Token.blockNames.length; i ++) {
RX_Token r1 = RX_Token.createRange();
int location;
if (i < NONBMP_BLOCK_START) {
location = i*2;
int rstart = RX_Token.blockRanges.charAt(location);
int rend = RX_Token.blockRanges.charAt(location+1);
//DEBUGING
//System.out.println(n+" " +Integer.toHexString(rstart)
// +"-"+ Integer.toHexString(rend));
r1.addRange(rstart, rend);
} else {
location = (i - NONBMP_BLOCK_START) * 2;
r1.addRange(RX_Token.nonBMPBlockRanges[location],
RX_Token.nonBMPBlockRanges[location + 1]);
}
String n = RX_Token.blockNames[i];
if (n.equals("Specials"))
r1.addRange(0xfff0, 0xfffd);
if (n.equals("Private Use")) {
r1.addRange(0xF0000,0xFFFFD);
r1.addRange(0x100000,0x10FFFD);
}
RX_Token.categories.put(n, r1);
RX_Token.categories2.put(n, RX_Token.complementRanges(r1));
buffer.setLength(0);
buffer.append("Is");
if (n.indexOf(' ') >= 0) {
for (int ci = 0; ci < n.length(); ci ++)
if (n.charAt(ci) != ' ') buffer.append(n.charAt(ci));
}
else {
buffer.append(n);
}
RX_Token.setAlias(buffer.toString(), n, true);
}
// TR#18 1.2
RX_Token.setAlias("ASSIGNED", "Cn", false);
RX_Token.setAlias("UNASSIGNED", "Cn", true);
RX_Token all = RX_Token.createRange();
all.addRange(0, RX_Token.UTF16_MAX);
RX_Token.categories.put("ALL", all);
RX_Token.categories2.put("ALL", RX_Token.complementRanges(all));
RX_Token.registerNonXS("ASSIGNED");
RX_Token.registerNonXS("UNASSIGNED");
RX_Token.registerNonXS("ALL");
RX_Token isalpha = RX_Token.createRange();
isalpha.mergeRanges(ranges[Character.UPPERCASE_LETTER]); // Lu
isalpha.mergeRanges(ranges[Character.LOWERCASE_LETTER]); // Ll
isalpha.mergeRanges(ranges[Character.OTHER_LETTER]); // Lo
RX_Token.categories.put("IsAlpha", isalpha);
RX_Token.categories2.put("IsAlpha", RX_Token.complementRanges(isalpha));
RX_Token.registerNonXS("IsAlpha");
RX_Token isalnum = RX_Token.createRange();
isalnum.mergeRanges(isalpha); // Lu Ll Lo
isalnum.mergeRanges(ranges[Character.DECIMAL_DIGIT_NUMBER]); // Nd
RX_Token.categories.put("IsAlnum", isalnum);
RX_Token.categories2.put("IsAlnum", RX_Token.complementRanges(isalnum));
RX_Token.registerNonXS("IsAlnum");
RX_Token isspace = RX_Token.createRange();
isspace.mergeRanges(RX_Token.token_spaces);
isspace.mergeRanges(ranges[CHAR_SEPARATOR]); // Z
RX_Token.categories.put("IsSpace", isspace);
RX_Token.categories2.put("IsSpace", RX_Token.complementRanges(isspace));
RX_Token.registerNonXS("IsSpace");
RX_Token isword = RX_Token.createRange();
isword.mergeRanges(isalnum); // Lu Ll Lo Nd
isword.addRange('_', '_');
RX_Token.categories.put("IsWord", isword);
RX_Token.categories2.put("IsWord", RX_Token.complementRanges(isword));
RX_Token.registerNonXS("IsWord");
RX_Token isascii = RX_Token.createRange();
isascii.addRange(0, 127);
RX_Token.categories.put("IsASCII", isascii);
RX_Token.categories2.put("IsASCII", RX_Token.complementRanges(isascii));
RX_Token.registerNonXS("IsASCII");
RX_Token isnotgraph = RX_Token.createRange();
isnotgraph.mergeRanges(ranges[CHAR_OTHER]);
isnotgraph.addRange(' ', ' ');
RX_Token.categories.put("IsGraph", RX_Token.complementRanges(isnotgraph));
RX_Token.categories2.put("IsGraph", isnotgraph);
RX_Token.registerNonXS("IsGraph");
RX_Token isxdigit = RX_Token.createRange();
isxdigit.addRange('0', '9');
isxdigit.addRange('A', 'F');
isxdigit.addRange('a', 'f');
RX_Token.categories.put("IsXDigit", RX_Token.complementRanges(isxdigit));
RX_Token.categories2.put("IsXDigit", isxdigit);
RX_Token.registerNonXS("IsXDigit");
RX_Token.setAlias("IsDigit", "Nd", true);
RX_Token.setAlias("IsUpper", "Lu", true);
RX_Token.setAlias("IsLower", "Ll", true);
RX_Token.setAlias("IsCntrl", "C", true);
RX_Token.setAlias("IsPrint", "C", false);
RX_Token.setAlias("IsPunct", "P", true);
RX_Token.registerNonXS("IsDigit");
RX_Token.registerNonXS("IsUpper");
RX_Token.registerNonXS("IsLower");
RX_Token.registerNonXS("IsCntrl");
RX_Token.registerNonXS("IsPrint");
RX_Token.registerNonXS("IsPunct");
RX_Token.setAlias("alpha", "IsAlpha", true);
RX_Token.setAlias("alnum", "IsAlnum", true);
RX_Token.setAlias("ascii", "IsASCII", true);
RX_Token.setAlias("cntrl", "IsCntrl", true);
RX_Token.setAlias("digit", "IsDigit", true);
RX_Token.setAlias("graph", "IsGraph", true);
RX_Token.setAlias("lower", "IsLower", true);
RX_Token.setAlias("print", "IsPrint", true);
RX_Token.setAlias("punct", "IsPunct", true);
RX_Token.setAlias("space", "IsSpace", true);
RX_Token.setAlias("upper", "IsUpper", true);
RX_Token.setAlias("word", "IsWord", true); // Perl extension
RX_Token.setAlias("xdigit", "IsXDigit", true);
RX_Token.registerNonXS("alpha");
RX_Token.registerNonXS("alnum");
RX_Token.registerNonXS("ascii");
RX_Token.registerNonXS("cntrl");
RX_Token.registerNonXS("digit");
RX_Token.registerNonXS("graph");
RX_Token.registerNonXS("lower");
RX_Token.registerNonXS("print");
RX_Token.registerNonXS("punct");
RX_Token.registerNonXS("space");
RX_Token.registerNonXS("upper");
RX_Token.registerNonXS("word");
RX_Token.registerNonXS("xdigit");
} // synchronized
} // if null
RangeToken tok = positive ? (RangeToken)RX_Token.categories.get(name)
: (RangeToken)RX_Token.categories2.get(name);
//if (tok == null) System.out.println(name);
return tok;
}