die "Usage: $0 <blocks.txt>" unless @ARGV == 1;
open (BLOCKS, $ARGV[0]) || die "Can't open Unicode block file: $!\n";
my %additions = ("SYRIAC" => "1.4",
"THAANA" => "1.4",
"SINHALA" => "1.4",
"MYANMAR" => "1.4",
"ETHIOPIC" => "1.4",
"CHEROKEE" => "1.4",
"UNIFIED_CANADIAN_ABORIGINAL_SYLLABICS" => "1.4",
"OGHAM" => "1.4",
"RUNIC" => "1.4",
"KHMER" => "1.4",
"MONGOLIAN" => "1.4",
"BRAILLE_PATTERNS" => "1.4",
"CJK_RADICALS_SUPPLEMENT" => "1.4",
"KANGXI_RADICALS" => "1.4",
"IDEOGRAPHIC_DESCRIPTION_CHARACTERS" => "1.4",
"BOPOMOFO_EXTENDED" => "1.4",
"CJK_UNIFIED_IDEOGRAPHS_EXTENSION_A" => "1.4",
"YI_SYLLABLES" => "1.4",
"YI_RADICALS" => "1.4",
);
print <<'EOF';
/**
* A family of character subsets in the Unicode specification. A character
* is in at most one of these blocks.
*
* This inner class was generated automatically from
* <code>$ARGV[0]</code>, by some perl scripts.
* This Unicode definition file can be found on the
* <a href="http://www.unicode.org">http://www.unicode.org</a> website.
* JDK 1.4 uses Unicode version 3.0.0.
*
* @author scripts/unicode-blocks.pl (written by Eric Blake)
* @since 1.2
*/
public static final class UnicodeBlock extends Subset
{
/** The start of the subset. */
private final char start;
/** The end of the subset. */
private final char end;
/**
* Constructor for strictly defined blocks.
*
* @param start the start character of the range
* @param end the end character of the range
* @param name the block name
*/
private UnicodeBlock(char start, char end, String name)
{
super(name);
this.start = start;
this.end = end;
}
/**
* Returns the Unicode character block which a character belongs to.
*
* @param ch the character to look up
* @return the set it belongs to, or null if it is not in one
*/
public static UnicodeBlock of(char ch)
{
// Special case, since SPECIALS contains two ranges.
if (ch == '\uFEFF')
return SPECIALS;
// Simple binary search for the correct block.
int low = 0;
int hi = sets.length - 1;
while (low <= hi)
{
int mid = (low + hi) >> 1;
UnicodeBlock b = sets[mid];
if (ch < b.start)
hi = mid - 1;
else if (ch > b.end)
low = mid + 1;
else
return b;
}
return null;
}
EOF
my $seenSpecials = 0;
my $seenSurrogates = 0;
my $surrogateStart = 0;
my @names = ();
while (<BLOCKS>) {
next if /^\ my ($start, $end, $block) = split(/; /);
next unless defined $block;
chomp $block;
$block =~ s/ *$//;
if (! $seenSpecials and $block =~ /Specials/) {
$seenSpecials = 1;
next;
}
if ($block =~ /Surrogates/) {
$seenSurrogates++;
if ($seenSurrogates == 1) {
$surrogateStart = $start;
next;
} elsif ($seenSurrogates == 2) {
next;
} else {
$start = $surrogateStart;
$block = "Surrogates Area";
}
}
$block =~ s/(Private Use)/$1 Area/;
(my $name = $block) =~ tr/a-z -/A-Z__/;
push @names, $name;
my $since = (defined $additions{$name}
? "\n * \@since $additions{$name}" : "");
my $extra = ($block =~ /Specials/ ? "'\\uFEFF', " : "");
print <<EOF;
/**
* $block.
* $extra'\\u$start' - '\\u$end'.$since
*/
public final static UnicodeBlock $name
= new UnicodeBlock('\\u$start', '\\u$end',
"$name");
EOF
}
print <<EOF;
/**
* The defined subsets.
*/
private static final UnicodeBlock sets[] = {
EOF
foreach (@names) {
print " $_,\n";
}
print <<EOF;
};
} // class UnicodeBlock
EOF