use strict;
use warnings;
use Carp;
+use Config;
use File::Find;
use File::Path;
use File::Spec;
use Text::Tabs;
sub DEBUG () { 0 } # Set to 0 for production; 1 for development
+my $debugging_build = $Config{"ccflags"} =~ /-DDEBUGGING/;
##########################################################################
#
# the small actual loop to process the input files and finish up; then
# a __DATA__ section, for the .t tests
#
-# This program works on all releases of Unicode through at least 5.2. The
+# This program works on all releases of Unicode through at least 6.0. The
# outputs have been scrutinized most intently for release 5.1. The others
# have been checked for somewhat more than just sanity. It can handle all
# existing Unicode character properties in those releases.
# out. But all the ones which can be used in regular expression \p{} and \P{}
# constructs will. Generally a property will have either its map table or its
# match tables written but not both. Again, what gets written is controlled
-# by lists which can easily be changed.
+# by lists which can easily be changed. Properties have a 'Type', like
+# binary, or string, or enum depending on how many match tables there are and
+# the content of the maps. This 'Type' is different than a range 'Type', so
+# don't get confused by the two concepts having the same name.
#
# For information about the Unicode properties, see Unicode's UAX44 document:
# More information on Unicode version glitches is further down in these
# introductory comments.
#
-# This program works on all properties as of 5.2, though the files for some
-# are suppressed from apparent lack of demand for them. You can change which
-# are output by changing lists in this program.
+# This program works on all non-provisional properties as of 6.0, though the
+# files for some are suppressed from apparent lack of demand for them. You
+# can change which are output by changing lists in this program.
#
-# The old version of mktables emphasized the term "Fuzzy" to mean Unocde's
+# The old version of mktables emphasized the term "Fuzzy" to mean Unicode's
# loose matchings rules (from Unicode TR18):
#
# The recommended names for UCD properties and property values are in
# Unicode_Radical_Stroke was listed in those files, so if the Unihan database
# is present in the directory, a table will be generated for that property.
# In 5.2, several more properties were added. For your convenience, the two
-# arrays are initialized with all the 5.2 listed properties that are also in
+# arrays are initialized with all the 6.0 listed properties that are also in
# earlier releases. But these are commented out. You can just uncomment the
# ones you want, or use them as a template for adding entries for other
# properties.
#
# Here are some observations about some of the issues in early versions:
#
-# The number of code points in \p{alpha} halve in 2.1.9. It turns out that
+# The number of code points in \p{alpha} halved in 2.1.9. It turns out that
# the reason is that the CJK block starting at 4E00 was removed from PropList,
# and was not put back in until 3.1.0
#
# special things
my $glob_list = 0; # ? Should we try to include unknown .txt files
# in the input.
-my $output_range_counts = 1; # ? Should we include the number of code points
- # in ranges in the output
+my $output_range_counts = $debugging_build; # ? Should we include the number
+ # of code points in ranges in
+ # the output
my $annotate = 0; # ? Should character names be in the output
# Verbosity levels; 0 is quiet
}
elsif ($arg eq '-annotate') {
$annotate = 1;
+ $debugging_build = 1;
+ $output_range_counts = 1;
}
else {
my $with_c = 'with';
'Canonical_Combining_Class=Attached_Below_Left'
}
-# These are listed in the Property aliases file in 5.2, but Unihan is ignored
+# These are listed in the Property aliases file in 6.0, but Unihan is ignored
# unless explicitly added.
if ($v_version ge v5.2.0) {
my $unihan = 'Unihan; remove from list if using Unihan';
my $other_properties = 'other properties';
my $contributory = "Used by Unicode internally for generating $other_properties and not intended to be used stand-alone";
- my $why_no_expand = "Easily computed, and yet doesn't cover the common encoding forms (UTF-16/8)",
+ my $why_no_expand = "Deprecated by Unicode: less useful than UTF-specific calculations",
%why_deprecated = (
- 'Grapheme_Link' => 'Deprecated by Unicode. Use ccc=vr (Canonical_Combining_Class=Virama) instead',
+ 'Grapheme_Link' => 'Deprecated by Unicode: Duplicates ccc=vr (Canonical_Combining_Class=Virama)',
'Jamo_Short_Name' => $contributory,
'Line_Break=Surrogate' => 'Deprecated by Unicode because surrogates should never appear in well-formed text, and therefore shouldn\'t be the basis for line breaking',
'Other_Alphabetic' => $contributory,
);
%why_suppressed = (
- # There is a lib/unicore/Decomposition.pl (used by normalize.pm) which
+ # There is a lib/unicore/Decomposition.pl (used by Normalize.pm) which
# contains the same information, but without the algorithmically
# determinable Hangul syllables'. This file is not published, so it's
# existence is not noted in the comment.
'Name' => "Accessible via 'use charnames;'",
'Name_Alias' => "Accessible via 'use charnames;'",
- # These are sort of jumping the gun; deprecation is proposed for
- # Unicode version 6.0, but they have never been exposed by Perl, and
- # likely are soon to be deprecated, so best not to expose them.
- FC_NFKC_Closure => 'Use NFKC_Casefold instead',
+ FC_NFKC_Closure => 'Supplanted in usage by NFKC_Casefold; otherwise not useful',
Expands_On_NFC => $why_no_expand,
Expands_On_NFD => $why_no_expand,
Expands_On_NFKC => $why_no_expand,
if ($v_version ge 4.0.0) {
$why_stabilized{'Hyphen'} = 'Use the Line_Break property instead; see www.unicode.org/reports/tr14';
+ if ($v_version ge 6.0.0) {
+ $why_deprecated{'Hyphen'} = 'Supplanted by Line_Break property values; see www.unicode.org/reports/tr14';
+ }
}
-if ($v_version ge 5.2.0) {
+if ($v_version ge 5.2.0 && $v_version lt 6.0.0) {
$why_obsolete{'ISO_Comment'} = 'Code points for it have been removed';
+ if ($v_version ge 6.0.0) {
+ $why_deprecated{'ISO_Comment'} = 'No longer needed for chart generation; otherwise not useful, and code points for it have been removed';
+ }
}
# Probably obsolete forever
# If you are using the Unihan database, you need to add the properties that
# you want to extract from it to this table. For your convenience, the
-# properties in the 5.2 PropertyAliases.txt file are listed, commented out
+# properties in the 6.0 PropertyAliases.txt file are listed, commented out
my @cjk_properties = split "\n", <<'END';
#cjkAccountingNumeric; kAccountingNumeric
#cjkOtherNumeric; kOtherNumeric
END
# Similarly for the property values. For your convenience, the lines in the
-# 5.2 PropertyAliases.txt file are listed. Just remove the first BUT NOT both
+# 6.0 PropertyAliases.txt file are listed. Just remove the first BUT NOT both
# '#' marks
my @cjk_property_values = split "\n", <<'END';
## @missing: 0000..10FFFF; cjkAccountingNumeric; NaN
'ReadMe.txt' => 'Just comments',
'README.TXT' => 'Just comments',
'StandardizedVariants.txt' => 'Only for glyph changes, not a Unicode character property. Does not fit into current scheme where one code point is mapped',
+ 'EmojiSources.txt' => 'Not of general utility: for Japanese legacy cell-phone applications',
+ 'IndicMatraCategory.txt' => 'Provisional',
+ 'IndicSyllabicCategory.txt' => 'Provisional',
+ 'ScriptExtensions.txt' => 'Provisional',
);
### End of externally interesting definitions, except for @input_file_objects
my $UNCONDITIONALLY = 2; # Replace without conditions.
my $MULTIPLE = 4; # Don't replace, but add a duplicate record if
# already there
+my $CROAK = 5; # Die with an error if is already there
# Flags to give property statuses. The phrases are to remind maintainers that
# if the flag is changed, the indefinite article referring to it in the
$end = min($block->containing_range($i)->end,
$unassigned_sans_noncharacters-> containing_range($i)->
end);
- } else {
- my_carp_bug("Can't figure out how to annotate"
- . sprintf("U+%04X", $i)
- . "Proceeding anyway.");
+ }
+ else {
+ Carp::my_carp_bug("Can't figure out how to annotate "
+ . sprintf("U+%04X", $i)
+ . ". Proceeding anyway.");
$viacode[$i] = 'UNKNOWN';
$annotate_char_type[$i] = $UNKNOWN_TYPE;
$printable[$i] = 0;
if ($clean_insert) {
if ($r->[$j]->standard_form ne $standard_form) {
$clean_insert = 0;
+ if ($replace == $CROAK) {
+ main::croak("The range to add "
+ . sprintf("%04X", $start)
+ . '-'
+ . sprintf("%04X", $end)
+ . " with value '$value' overlaps an existing range $r->[$j]");
+ }
}
else {
$extends_above = ($j+1 < $range_list_size
&& $r->[$j+1]->start == $end +1
&& $r->[$j+1]->standard_form eq $standard_form
- && $r->[$j-1]->type == $type);
+ && $r->[$j+1]->type == $type);
}
if ($extends_below && $extends_above) { # Adds to both
$splice_start--; # start replace at element below
trace "i =[", $i, "]", $r->[$i];
trace 'i+1=[', $i+1, ']', $r->[$i+1] if $i < @$r - 1;
trace 'i+2=[', $i+2, ']', $r->[$i+2] if $i < @$r - 2;
- trace "removed @return";
+ trace "removed ", @return if @return;
}
# An actual deletion could have changed the maximum in the list.
sub add_comment { # Adds the parameter as a comment.
+ return unless $debugging_build;
+
my $self = shift;
my $comment = shift;
Carp::carp_extra_args(\@_) if main::DEBUG && @_;
# If there is a range and doesn't need a single point range
# output
if ($start != $end && ! $range_size_1) {
- push @OUT, sprintf "%04X\t%04X\t%s", $start, $end, $value;
+ push @OUT, sprintf "%04X\t%04X", $start, $end;
+ $OUT[-1] .= "\t$value" if $value ne "";
# Add a comment with the size of the range, if requested.
# Expand Tabs to make sure they all start in the same
# Just before output, create the comment that heads the file
# containing this table.
+ return unless $debugging_build;
+
my $self = shift;
Carp::carp_extra_args(\@_) if main::DEBUG && @_;
# ones that share the same file. It lists all such tables, ordered so
# that related ones are together.
+ return unless $debugging_build;
+
my $leader = shift; # Should only be called on the leader table of
# an equivalent group
Carp::carp_extra_args(\@_) if main::DEBUG && @_;
}
}
- # This entry is still missing as of 5.2, perhaps because no short name for
+ # This entry is still missing as of 6.0, perhaps because no short name for
# it.
if (-e 'NameAliases.txt') {
my $aliases = property_ref('Name_Alias');
# file, in any order, interspersed in any way. The first time a
# property is seen, it gets information about that property and
# caches it for quick retrieval later. It also normalizes the maps
- # so that only one of many synonym is stored. The Unicode input files
- # do use some multiple synonyms.
+ # so that only one of many synonyms is stored. The Unicode input
+ # files do use some multiple synonyms.
my $file = shift;
Carp::carp_extra_args(\@_) if main::DEBUG && @_;
}
return;
}
+
+ sub filter_v6_ucd {
+
+ # Unicode 6.0 co-opted the name BELL for U+1F514, so change the input
+ # to pretend that U+0007 is ALERT instead, and for Perl 5.14, don't
+ # allow the BELL name for U+1F514, so that the old usage can be
+ # deprecated for one cycle.
+
+ return if $_ !~ /^(?:0007|1F514|070F);/;
+
+ my ($code_point, @fields) = split /\s*;\s*/, $_, -1;
+ if ($code_point eq '0007') {
+ $fields[$CHARNAME] = "ALERT";
+ }
+ elsif ($code_point eq '070F') { # Unicode Corrigendum #8; see
+ # http://www.unicode.org/versions/corrigendum8.html
+ $fields[$BIDI] = "AL";
+ }
+ elsif ($^V lt v5.15.0) { # For 5.16 will convert to use Unicode's name
+ $fields[$CHARNAME] = "";
+ }
+
+ $_ = join ';', $code_point, @fields;
+
+ return;
+ }
} # End closure for UnicodeData
sub process_GCB_test {
# implemented, it would be by hard-coding in the casing functions in the
# Perl core, not through tables. But if there is a new condition we don't
# know about, output a warning. We know about all the conditions through
- # 5.2
+ # 6.0
if ($fields[4] ne "") {
my @conditions = split ' ', $fields[4];
if ($conditions[0] ne 'tr' # We know that these languages have
# Add mappings to the property for each code point in the list
foreach my $range ($list->ranges) {
- $property->add_map($range->start, $range->end, $default);
+ $property->add_map($range->start, $range->end, $default,
+ Replace => $CROAK);
}
}
}
# Add any remaining code points to the mapping, using the default for
- # missing code points
+ # missing code points.
if (defined (my $default_map = $property->default_map)) {
foreach my $range ($property->inverse_list->ranges) {
$property->add_map($range->start, $range->end, $default_map);
# range, with their names prefaced by 'Posix', to signify that these match
# what the Posix standard says they should match. A couple are
# effectively this, but the name doesn't have 'Posix' in it because there
- # just isn't any Posix equivalent.
+ # just isn't any Posix equivalent. 'XPosix' are the Posix tables extended
+ # to the full Unicode range, by our guesses as to what is appropriate.
# 'Any' is all code points. As an error check, instead of just setting it
# to be that, construct it to be the union of all the major categories
$Lower->set_equivalent_to($gc->table('Lowercase_Letter'),
Related => 1);
}
+ $Lower->add_alias('XPosixLower');
$perl->add_match_table("PosixLower",
Description => "[a-z]",
Initialize => $Lower & $ASCII,
$Upper->set_equivalent_to($gc->table('Uppercase_Letter'),
Related => 1);
}
+ $Upper->add_alias('XPosixUpper');
$perl->add_match_table("PosixUpper",
Description => "[A-Z]",
Initialize => $Upper & $ASCII,
$Alpha += $gc->table('Nl') if defined $gc->table('Nl');
$Alpha->add_description('Alphabetic');
}
+ $Alpha->add_alias('XPosixAlpha');
$perl->add_match_table("PosixAlpha",
Description => "[A-Za-z]",
Initialize => $Alpha & $ASCII,
Description => 'Alphabetic and (Decimal) Numeric',
Initialize => $Alpha + $gc->table('Decimal_Number'),
);
+ $Alnum->add_alias('XPosixAlnum');
$perl->add_match_table("PosixAlnum",
Description => "[A-Za-z0-9]",
Initialize => $Alnum & $ASCII,
);
my $Word = $perl->add_match_table('Word',
- Description => '\w, including beyond ASCII',
+ Description => '\w, including beyond ASCII;'
+ . ' = \p{Alnum} + \pM + \p{Pc}',
Initialize => $Alnum + $gc->table('Mark'),
);
+ $Word->add_alias('XPosixWord');
my $Pc = $gc->table('Connector_Punctuation'); # 'Pc' Not in release 1
$Word += $Pc if defined $Pc;
# This is a Perl extension, so the name doesn't begin with Posix.
- $perl->add_match_table('PerlWord',
+ my $PerlWord = $perl->add_match_table('PerlWord',
Description => '\w, restricted to ASCII = [A-Za-z0-9_]',
Initialize => $Word & $ASCII,
);
+ $PerlWord->add_alias('PosixWord');
my $Blank = $perl->add_match_table('Blank',
Description => '\h, Horizontal white space',
- 0x200B, # ZWSP
);
$Blank->add_alias('HorizSpace'); # Another name for it.
+ $Blank->add_alias('XPosixBlank');
$perl->add_match_table("PosixBlank",
Description => "\\t and ' '",
Initialize => $Blank & $ASCII,
Description => '\s including beyond ASCII plus vertical tab',
Initialize => $Blank + $VertSpace,
);
+ $Space->add_alias('XPosixSpace');
$perl->add_match_table("PosixSpace",
Description => "\\t, \\n, \\cK, \\f, \\r, and ' '. (\\cK is vertical tab)",
Initialize => $Space & $ASCII,
);
# Perl's traditional space doesn't include Vertical Tab
- my $SpacePerl = $perl->add_match_table('SpacePerl',
+ my $XPerlSpace = $perl->add_match_table('XPerlSpace',
Description => '\s, including beyond ASCII',
Initialize => $Space - 0x000B,
);
- $perl->add_match_table('PerlSpace',
+ $XPerlSpace->add_alias('SpacePerl'); # A pre-existing synonym
+ my $PerlSpace = $perl->add_match_table('PerlSpace',
Description => '\s, restricted to ASCII',
- Initialize => $SpacePerl & $ASCII,
+ Initialize => $XPerlSpace & $ASCII,
);
+
my $Cntrl = $perl->add_match_table('Cntrl',
Description => 'Control characters');
$Cntrl->set_equivalent_to($gc->table('Cc'), Related => 1);
+ $Cntrl->add_alias('XPosixCntrl');
$perl->add_match_table("PosixCntrl",
Description => "ASCII control characters: NUL, SOH, STX, ETX, EOT, ENQ, ACK, BEL, BS, HT, LF, VT, FF, CR, SO, SI, DLE, DC1, DC2, DC3, DC4, NAK, SYN, ETB, CAN, EOM, SUB, ESC, FS, GS, RS, US, and DEL",
Initialize => $Cntrl & $ASCII,
Description => 'Characters that are graphical',
Initialize => ~ ($Space + $controls),
);
+ $Graph->add_alias('XPosixGraph');
$perl->add_match_table("PosixGraph",
Description =>
'[-!"#$%&\'()*+,./:;<>?@[\\\]^_`{|}~0-9A-Za-z]',
Description => 'Characters that are graphical plus space characters (but no controls)',
Initialize => $Blank + $Graph - $gc->table('Control'),
);
+ $print->add_alias('XPosixPrint');
$perl->add_match_table("PosixPrint",
Description =>
'[- 0-9A-Za-z!"#$%&\'()*+,./:;<>?@[\\\]^_`{|}~]',
$Punct->set_equivalent_to($gc->table('Punctuation'), Related => 1);
# \p{punct} doesn't include the symbols, which posix does
+ my $XPosixPunct = $perl->add_match_table('XPosixPunct',
+ Description => '\p{Punct} + ASCII-range \p{Symbol}',
+ Initialize => $gc->table('Punctuation')
+ + ($ASCII & $gc->table('Symbol')),
+ );
$perl->add_match_table('PosixPunct',
Description => '[-!"#$%&\'()*+,./:;<>?@[\\\]^_`{|}~]',
- Initialize => $ASCII & ($gc->table('Punctuation')
- + $gc->table('Symbol')),
+ Initialize => $ASCII & $XPosixPunct,
);
my $Digit = $perl->add_match_table('Digit',
Description => '[0-9] + all other decimal digits');
$Digit->set_equivalent_to($gc->table('Decimal_Number'), Related => 1);
+ $Digit->add_alias('XPosixDigit');
my $PosixDigit = $perl->add_match_table("PosixDigit",
Description => '[0-9]',
Initialize => $Digit & $ASCII,
# Hex_Digit was not present in first release
my $Xdigit = $perl->add_match_table('XDigit');
+ $Xdigit->add_alias('XPosixXDigit');
my $Hex = property_ref('Hex_Digit');
if (defined $Hex && ! $Hex->is_empty) {
$Xdigit->set_equivalent_to($Hex->table('Y'), Related => 1);
0xFF10..0xFF19, 0xFF21..0xFF26, 0xFF41..0xFF46]);
$Xdigit->add_description('[0-9A-Fa-f] and corresponding fullwidth versions, like U+FF10: FULLWIDTH DIGIT ZERO');
}
+ $perl->add_match_table('PosixXDigit',
+ Initialize => $ASCII & $Xdigit,
+ Description => '[0-9A-Fa-f]',
+ );
my $dt = property_ref('Decomposition_Type');
$dt->add_match_table('Non_Canon', Full_Name => 'Non_Canonical',
if ($table_property != $perl && $table->perl_extension) {
push @info, '(Perl extension)';
}
- push @info, "($string_count)" if $output_range_counts;
+ push @info, "($string_count)";
# Now, we have both the entry and info so add them to the
# list of all the properties.
B<Compound forms> consist of two components, separated by an equals sign or a
colon. The first component is the property name, and the second component is
the particular value of the property to match against, for example,
-'\\p{Script: Greek}' or '\\p{Script=Greek}' both mean to match characters
+'\\p{Script: Greek}' and '\\p{Script=Greek}' both mean to match characters
whose Script property is Greek.
B<Single forms>, like '\\p{Greek}', are mostly Perl-defined shortcuts for
=item Obsolete
Properties marked with $a_bold_obsolete in the table are considered
-obsolete. At the time of this writing (Unicode version 5.2) there is no
-information in the Unicode standard about the implications of a property being
obsolete.
=item Stabilized
-Obsolete properties may be stabilized. This means that they are not actively
-maintained by Unicode, and will not be extended as new characters are added to
-the standard. Such properties are marked with $a_bold_stabilized in the
-table. At the time of this writing (Unicode version 5.2) there is no further
-information in the Unicode standard about the implications of a property being
-stabilized.
+Obsolete properties may be stabilized. Such a determination does not indicate
+that the property should or should not be used; instead it is a declaration
+that the property will not be maintained nor extended for newly encoded
+characters. Such properties are marked with $a_bold_stabilized in the
+table.
=item Deprecated
-Obsolete properties may be deprecated. This means that their use is strongly
+An obsolete property may be deprecated, perhaps because its original intent
+has been replaced by another property or because its specification was somehow
+defective. This means that its use is strongly
discouraged, so much so that a warning will be issued if used, unless the
regular expression is in the scope of a C<S<no warnings 'deprecated'>>
statement. $A_bold_deprecated flags each such entry in the table, and
=back
An installation can choose to allow any of these to be matched by changing the
-controlling lists contained in the program C<\$Config{privlib}>/F<unicore/$0>
-and then re-running F<$0>. (C<\%Config> is available from the Config module).
+controlling lists contained in the program
+C<\$Config{privlib}>/F<unicore/mktables> and then re-running F<mktables>.
+(C<\%Config> is available from the Config module).
=head1 Files in the I<To> directory (for serious hackers only)
@map_tables_actually_output
An installation can choose to change which files are generated by changing the
-controlling lists contained in the program C<\$Config{privlib}>/F<unicore/$0>
-and then re-running F<$0>.
+controlling lists contained in the program
+C<\$Config{privlib}>/F<unicore/mktables> and then re-running F<mktables>.
Each of these files defines two hash entries to help reading programs decipher
it. One of them looks like this:
? \&filter_v1_ucd
: ($v_version eq v2.1.5)
? \&filter_v2_1_5_ucd
- : undef),
+
+ # And for 5.14 Perls with 6.0,
+ # have to also make changes
+ : ($v_version ge v6.0.0)
+ ? \&filter_v6_ucd
+ : undef),
# And the main filter
\&filter_UnicodeData_line,
https://perl5.git.perl.org / perl5.git / blobdiff