From e6918187568dbd01842d8d1d2c808ce16a894239 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Sun, 21 Apr 2024 13:54:28 +0200 Subject: Adding upstream version 18.2.2. Signed-off-by: Daniel Baumann --- src/utf8proc/data/Makefile | 69 ++++++ src/utf8proc/data/charwidths.jl | 184 ++++++++++++++++ src/utf8proc/data/data_generator.rb | 426 ++++++++++++++++++++++++++++++++++++ 3 files changed, 679 insertions(+) create mode 100644 src/utf8proc/data/Makefile create mode 100644 src/utf8proc/data/charwidths.jl create mode 100644 src/utf8proc/data/data_generator.rb (limited to 'src/utf8proc/data') diff --git a/src/utf8proc/data/Makefile b/src/utf8proc/data/Makefile new file mode 100644 index 000000000..1b2472807 --- /dev/null +++ b/src/utf8proc/data/Makefile @@ -0,0 +1,69 @@ +# Unicode data generation rules. Except for the test data files, most +# users will not use these Makefile rules, which are primarily to re-generate +# unicode_data.c when we get a new Unicode version or charwidth data; they +# require ruby, fontforge, and julia to be installed. + +# programs +CURL=curl +RUBY=ruby +PERL=perl +MAKE=make +JULIA=julia +FONTFORGE=fontforge +CURLFLAGS = --retry 5 --location + +.PHONY: clean + +.DELETE_ON_ERROR: + +utf8proc_data.c.new: data_generator.rb UnicodeData.txt GraphemeBreakProperty.txt DerivedCoreProperties.txt CompositionExclusions.txt CaseFolding.txt CharWidths.txt emoji-data.txt + $(RUBY) data_generator.rb < UnicodeData.txt > $@ + +# GNU Unifont version for font metric calculations: +UNIFONT_VERSION=11.0.01 + +unifont.ttf: + $(CURL) $(CURLFLAGS) -o $@ $(URLCACHE)https://mirrors.kernel.org/gnu/unifont/unifont-$(UNIFONT_VERSION)/unifont-$(UNIFONT_VERSION).ttf + +unifont_upper.ttf: + $(CURL) $(CURLFLAGS) -o $@ $(URLCACHE)https://mirrors.kernel.org/gnu/unifont/unifont-$(UNIFONT_VERSION)/unifont_upper-$(UNIFONT_VERSION).ttf + +%.sfd: %.ttf + $(FONTFORGE) -lang=ff -c "Open(\"$<\");Save(\"$@\");Quit(0);" + +CharWidths.txt: charwidths.jl unifont.sfd unifont_upper.sfd EastAsianWidth.txt + $(JULIA) charwidths.jl > $@ + +# Unicode data version +UNICODE_VERSION=11.0.0 + +UnicodeData.txt: + $(CURL) $(CURLFLAGS) -o $@ -O http://www.unicode.org/Public/$(UNICODE_VERSION)/ucd/UnicodeData.txt + +EastAsianWidth.txt: + $(CURL) $(CURLFLAGS) -o $@ -O $(URLCACHE)http://www.unicode.org/Public/$(UNICODE_VERSION)/ucd/EastAsianWidth.txt + +GraphemeBreakProperty.txt: + $(CURL) $(CURLFLAGS) -o $@ -O $(URLCACHE)http://www.unicode.org/Public/$(UNICODE_VERSION)/ucd/auxiliary/GraphemeBreakProperty.txt + +DerivedCoreProperties.txt: + $(CURL) $(CURLFLAGS) -o $@ -O $(URLCACHE)http://www.unicode.org/Public/$(UNICODE_VERSION)/ucd/DerivedCoreProperties.txt + +CompositionExclusions.txt: + $(CURL) $(CURLFLAGS) -o $@ -O $(URLCACHE)http://www.unicode.org/Public/$(UNICODE_VERSION)/ucd/CompositionExclusions.txt + +CaseFolding.txt: + $(CURL) $(CURLFLAGS) -o $@ -O $(URLCACHE)http://www.unicode.org/Public/$(UNICODE_VERSION)/ucd/CaseFolding.txt + +NormalizationTest.txt: + $(CURL) $(CURLFLAGS) -o $@ -O $(URLCACHE)http://www.unicode.org/Public/$(UNICODE_VERSION)/ucd/NormalizationTest.txt + +GraphemeBreakTest.txt: + $(CURL) $(CURLFLAGS) $(URLCACHE)http://www.unicode.org/Public/$(UNICODE_VERSION)/ucd/auxiliary/GraphemeBreakTest.txt | $(PERL) -pe 's,÷,/,g;s,×,+,g' > $@ + +emoji-data.txt: + $(CURL) $(CURLFLAGS) -o $@ -O $(URLCACHE)http://unicode.org/Public/emoji/`echo $(UNICODE_VERSION) | cut -d. -f1-2`/emoji-data.txt + +clean: + rm -f UnicodeData.txt EastAsianWidth.txt GraphemeBreakProperty.txt DerivedCoreProperties.txt CompositionExclusions.txt CaseFolding.txt NormalizationTest.txt GraphemeBreakTest.txt CharWidths.txt unifont*.ttf unifont*.sfd emoji-data.txt + rm -f utf8proc_data.c.new diff --git a/src/utf8proc/data/charwidths.jl b/src/utf8proc/data/charwidths.jl new file mode 100644 index 000000000..7b3d15874 --- /dev/null +++ b/src/utf8proc/data/charwidths.jl @@ -0,0 +1,184 @@ +# Following work by @jiahao, we compute character widths using a combination of +# * advance widths from GNU Unifont (advance width 512 = 1 en) +# * UAX 11: East Asian Width +# * a few exceptions as needed +# Adapted from http://nbviewer.ipython.org/gist/jiahao/07e8b08bf6d8671e9734 +# +# Requires Julia (obviously) and FontForge. + +############################################################################# +CharWidths = Dict{Int,Int}() + +############################################################################# +# Use ../libutf8proc for category codes, rather than the one in Julia, +# to minimize bootstrapping complexity when a new version of Unicode comes out. +catcode(c) = ccall((:utf8proc_category,"../libutf8proc"), Cint, (Int32,), c) + +# use Base.UTF8proc module to get category codes constants, since +# we won't change these in utf8proc. +import Base.UTF8proc + +############################################################################# +# Use a default width of 1 for all character categories that are +# letter/symbol/number-like, as well as for unassigned/private-use chars. +# This can be overriden by Unifont or UAX 11 +# below, but provides a useful nonzero fallback for new codepoints when +# a new Unicode version has been released but Unifont hasn't been updated yet. + +zerowidth = Set{Int}() # categories that may contain zero-width chars +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_MN) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_MC) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_ME) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_SK) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_ZS) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_ZL) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_ZP) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_CC) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_CF) +push!(zerowidth, UTF8proc.UTF8PROC_CATEGORY_CS) +for c in 0x0000:0x110000 + if catcode(c) ∉ zerowidth + CharWidths[c] = 1 + end +end + +############################################################################# +# Widths from GNU Unifont + +#Read sfdfile for character widths +function parsesfd(filename::AbstractString, CharWidths::Dict{Int,Int}=Dict{Int,Int}()) + state=:seekchar + lineno = 0 + codepoint = width = nothing + for line in readlines(open(filename)) + lineno += 1 + if state==:seekchar #StartChar: nonmarkingreturn + if contains(line, "StartChar: ") + codepoint = nothing + width = nothing + state = :readdata + end + elseif state==:readdata #Encoding: 65538 -1 2, Width: 1024 + contains(line, "Encoding:") && (codepoint = parse(Int, split(line)[3])) + contains(line, "Width:") && (width = parse(Int, split(line)[2])) + if codepoint!=nothing && width!=nothing && codepoint >= 0 + w=div(width, 512) # 512 units to the en + if w > 0 + # only add nonzero widths, since (1) the default is zero + # and (2) this circumvents some apparent bugs in Unifont + # (https://savannah.gnu.org/bugs/index.php?45395) + CharWidths[codepoint] = w + end + state = :seekchar + end + end + end + CharWidths +end +CharWidths=parsesfd("unifont.sfd", CharWidths) +CharWidths=parsesfd("unifont_upper.sfd", CharWidths) + +############################################################################# +# Widths from UAX #11: East Asian Width +# .. these take precedence over the Unifont width for all codepoints +# listed explicitly as wide/full/narrow/half-width + +for line in readlines(open("EastAsianWidth.txt")) + #Strip comments + (isempty(line) || line[1] == '#') && continue + precomment = split(line, '#')[1] + #Parse code point range and width code + tokens = split(precomment, ';') + length(tokens) >= 2 || continue + charrange = tokens[1] + width = strip(tokens[2]) + #Parse code point range into Julia UnitRange + rangetokens = split(charrange, "..") + charstart = parse(UInt32, "0x"*rangetokens[1]) + charend = parse(UInt32, "0x"*rangetokens[length(rangetokens)>1 ? 2 : 1]) + + #Assign widths + for c in charstart:charend + if width=="W" || width=="F" # wide or full + CharWidths[c]=2 + elseif width=="Na"|| width=="H" + CharWidths[c]=1 + end + end +end + +############################################################################# +# A few exceptions to the above cases, found by manual comparison +# to other wcwidth functions and similar checks. + +for c in keys(CharWidths) + cat = catcode(c) + + # make sure format control character (category Cf) have width 0 + # (some of these, like U+0601, can have a width in some cases + # but normally act like prepended combining marks. U+fff9 etc + # are also odd, but have zero width in typical terminal contexts) + if cat==UTF8proc.UTF8PROC_CATEGORY_CF + CharWidths[c]=0 + end + + # Unifont has nonzero width for a number of non-spacing combining + # characters, e.g. (in 7.0.06): f84,17b4,17b5,180b,180d,2d7f, and + # the variation selectors + if cat==UTF8proc.UTF8PROC_CATEGORY_MN + CharWidths[c]=0 + end + + # We also assign width of one to unassigned and private-use + # codepoints (Unifont includes ConScript Unicode Registry PUA fonts, + # but since these are nonstandard it seems questionable to use Unifont metrics; + # if they are printed as the replacement character U+FFFD they will have width 1). + if cat==UTF8proc.UTF8PROC_CATEGORY_CO || cat==UTF8proc.UTF8PROC_CATEGORY_CN + CharWidths[c]=1 + end + + # for some reason, Unifont has width-2 glyphs for ASCII control chars + if cat==UTF8proc.UTF8PROC_CATEGORY_CC + CharWidths[c]=0 + end +end + +#Soft hyphen is typically printed as a hyphen (-) in terminals. +CharWidths[0x00ad]=1 + +#By definition, should have zero width (on the same line) +#0x002028 '
' category: Zl name: LINE SEPARATOR/ +#0x002029 '
' category: Zp name: PARAGRAPH SEPARATOR/ +CharWidths[0x2028]=0 +CharWidths[0x2029]=0 + +#By definition, should be narrow = width of 1 en space +#0x00202f ' ' category: Zs name: NARROW NO-BREAK SPACE/ +CharWidths[0x202f]=1 + +#By definition, should be wide = width of 1 em space +#0x002001 ' ' category: Zs name: EM QUAD/ +#0x002003 ' ' category: Zs name: EM SPACE/ +CharWidths[0x2001]=2 +CharWidths[0x2003]=2 + +############################################################################# +# Output (to a file or pipe) for processing by data_generator.rb, +# encoded as a sequence of intervals. + +firstc = 0x000000 +lastv = 0 +uhex(c) = uppercase(hex(c,4)) +for c in 0x0000:0x110000 + v = get(CharWidths, c, 0) + if v != lastv || c == 0x110000 + v < 4 || error("invalid charwidth $v for $c") + if firstc+1 < c + println(uhex(firstc), "..", uhex(c-1), "; ", lastv) + else + println(uhex(firstc), "; ", lastv) + end + firstc = c + lastv = v + end +end diff --git a/src/utf8proc/data/data_generator.rb b/src/utf8proc/data/data_generator.rb new file mode 100644 index 000000000..8bd87e8da --- /dev/null +++ b/src/utf8proc/data/data_generator.rb @@ -0,0 +1,426 @@ +#!/usr/bin/env ruby + +# This file was used to generate the 'unicode_data.c' file by parsing the +# Unicode data file 'UnicodeData.txt' of the Unicode Character Database. +# It is included for informational purposes only and not intended for +# production use. + + +# Copyright (c) 2018 Steven G. Johnson, Tony Kelman, Keno Fischer, +# Benito van der Zander, Michaël Meyer, and other contributors. +# Copyright (c) 2009 Public Software Group e. V., Berlin, Germany +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE +# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER +# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING +# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER +# DEALINGS IN THE SOFTWARE. + + +# This file contains derived data from a modified version of the +# Unicode data files. The following license applies to that data: +# +# COPYRIGHT AND PERMISSION NOTICE +# +# Copyright (c) 1991-2007 Unicode, Inc. All rights reserved. Distributed +# under the Terms of Use in http://www.unicode.org/copyright.html. +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of the Unicode data files and any associated documentation (the "Data +# Files") or Unicode software and any associated documentation (the +# "Software") to deal in the Data Files or Software without restriction, +# including without limitation the rights to use, copy, modify, merge, +# publish, distribute, and/or sell copies of the Data Files or Software, and +# to permit persons to whom the Data Files or Software are furnished to do +# so, provided that (a) the above copyright notice(s) and this permission +# notice appear with all copies of the Data Files or Software, (b) both the +# above copyright notice(s) and this permission notice appear in associated +# documentation, and (c) there is clear notice in each modified Data File or +# in the Software as well as in the documentation associated with the Data +# File(s) or Software that the data or software has been modified. +# +# THE DATA FILES AND SOFTWARE ARE PROVIDED "AS IS", WITHOUT WARRANTY OF ANY +# KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OF +# THIRD PARTY RIGHTS. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR HOLDERS +# INCLUDED IN THIS NOTICE BE LIABLE FOR ANY CLAIM, OR ANY SPECIAL INDIRECT OR +# CONSEQUENTIAL DAMAGES, OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF +# USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER +# TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR +# PERFORMANCE OF THE DATA FILES OR SOFTWARE. +# +# Except as contained in this notice, the name of a copyright holder shall +# not be used in advertising or otherwise to promote the sale, use or other +# dealings in these Data Files or Software without prior written +# authorization of the copyright holder. + + +$ignorable_list = File.read("DerivedCoreProperties.txt")[/# Derived Property: Default_Ignorable_Code_Point.*?# Total code points:/m] +$ignorable = [] +$ignorable_list.each_line do |entry| + if entry =~ /^([0-9A-F]+)\.\.([0-9A-F]+)/ + $1.hex.upto($2.hex) { |e2| $ignorable << e2 } + elsif entry =~ /^[0-9A-F]+/ + $ignorable << $&.hex + end +end + +$grapheme_boundclass_list = File.read("GraphemeBreakProperty.txt") +$grapheme_boundclass = Hash.new("UTF8PROC_BOUNDCLASS_OTHER") +$grapheme_boundclass_list.each_line do |entry| + if entry =~ /^([0-9A-F]+)\.\.([0-9A-F]+)\s*;\s*([A-Za-z_]+)/ + $1.hex.upto($2.hex) { |e2| $grapheme_boundclass[e2] = "UTF8PROC_BOUNDCLASS_" + $3.upcase } + elsif entry =~ /^([0-9A-F]+)\s*;\s*([A-Za-z_]+)/ + $grapheme_boundclass[$1.hex] = "UTF8PROC_BOUNDCLASS_" + $2.upcase + end +end + +$emoji_data_list = File.read("emoji-data.txt") +$emoji_data_list.each_line do |entry| + if entry =~ /^([0-9A-F]+)\.\.([0-9A-F]+)\s*;\s*Extended_Pictographic\W/ + $1.hex.upto($2.hex) { |e2| $grapheme_boundclass[e2] = "UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC" } + elsif entry =~ /^([0-9A-F]+)\s*;\s*Extended_Pictographic\W/ + $grapheme_boundclass[$1.hex] = "UTF8PROC_BOUNDCLASS_EXTENDED_PICTOGRAPHIC" + elsif entry =~ /^([0-9A-F]+)\.\.([0-9A-F]+)\s*;\s*Emoji_Modifier\W/ + $1.hex.upto($2.hex) { |e2| $grapheme_boundclass[e2] = "UTF8PROC_BOUNDCLASS_EXTEND" } + elsif entry =~ /^([0-9A-F]+)\s*;\s*Emoji_Modifier\W/ + $grapheme_boundclass[$1.hex] = "UTF8PROC_BOUNDCLASS_EXTEND" + end +end + +$charwidth_list = File.read("CharWidths.txt") +$charwidth = Hash.new(0) +$charwidth_list.each_line do |entry| + if entry =~ /^([0-9A-F]+)\.\.([0-9A-F]+)\s*;\s*([0-9]+)/ + $1.hex.upto($2.hex) { |e2| $charwidth[e2] = $3.to_i } + elsif entry =~ /^([0-9A-F]+)\s*;\s*([0-9]+)/ + $charwidth[$1.hex] = $2.to_i + end +end + +$exclusions = File.read("CompositionExclusions.txt")[/# \(1\) Script Specifics.*?# Total code points:/m] +$exclusions = $exclusions.chomp.split("\n").collect { |e| e.hex } + +$excl_version = File.read("CompositionExclusions.txt")[/# \(2\) Post Composition Version precomposed characters.*?# Total code points:/m] +$excl_version = $excl_version.chomp.split("\n").collect { |e| e.hex } + +$case_folding_string = File.open("CaseFolding.txt", :encoding => 'utf-8').read +$case_folding = {} +$case_folding_string.chomp.split("\n").each do |line| + next unless line =~ /([0-9A-F]+); [CF]; ([0-9A-F ]+);/i + $case_folding[$1.hex] = $2.split(" ").collect { |e| e.hex } +end + +$int_array = [] +$int_array_indicies = {} + +def str2c(string, prefix) + return "0" if string.nil? + return "UTF8PROC_#{prefix}_#{string.upcase}" +end +def pushary(array) + idx = $int_array_indicies[array] + unless idx + $int_array_indicies[array] = $int_array.length + idx = $int_array.length + array.each { |entry| $int_array << entry } + end + return idx +end +def cpary2utf16encoded(array) + return array.flat_map { |cp| + if (cp <= 0xFFFF) + raise "utf-16 code: #{cp}" if cp & 0b1111100000000000 == 0b1101100000000000 + cp + else + temp = cp - 0x10000 + [(temp >> 10) | 0b1101100000000000, (temp & 0b0000001111111111) | 0b1101110000000000] + end + } +end +def cpary2c(array) + return "UINT16_MAX" if array.nil? || array.length == 0 + lencode = array.length - 1 #no sequence has len 0, so we encode len 1 as 0, len 2 as 1, ... + array = cpary2utf16encoded(array) + if lencode >= 7 #we have only 3 bits for the length (which is already cutting it close. might need to change it to 2 bits in future Unicode versions) + array = [lencode] + array + lencode = 7 + end + idx = pushary(array) + raise "Array index out of bound" if idx > 0x1FFF + return "#{idx | (lencode << 13)}" +end +def singlecpmap(cp) + return "UINT16_MAX" if cp == nil + idx = pushary(cpary2utf16encoded([cp])) + raise "Array index out of bound" if idx > 0xFFFF + return "#{idx}" +end + +class UnicodeChar + attr_accessor :code, :name, :category, :combining_class, :bidi_class, + :decomp_type, :decomp_mapping, + :bidi_mirrored, + :uppercase_mapping, :lowercase_mapping, :titlecase_mapping, + #caches: + :c_entry_index, :c_decomp_mapping, :c_case_folding + def initialize(line) + raise "Could not parse input." unless line =~ /^ + ([0-9A-F]+); # code + ([^;]+); # name + ([A-Z]+); # general category + ([0-9]+); # canonical combining class + ([A-Z]+); # bidi class + (<([A-Z]*)>)? # decomposition type + ((\ ?[0-9A-F]+)*); # decompomposition mapping + ([0-9]*); # decimal digit + ([0-9]*); # digit + ([^;]*); # numeric + ([YN]*); # bidi mirrored + ([^;]*); # unicode 1.0 name + ([^;]*); # iso comment + ([0-9A-F]*); # simple uppercase mapping + ([0-9A-F]*); # simple lowercase mapping + ([0-9A-F]*)$/ix # simple titlecase mapping + @code = $1.hex + @name = $2 + @category = $3 + @combining_class = Integer($4) + @bidi_class = $5 + @decomp_type = $7 + @decomp_mapping = ($8=='') ? nil : + $8.split.collect { |element| element.hex } + @bidi_mirrored = ($13=='Y') ? true : false + # issue #130: use nonstandard uppercase ß -> ẞ + @uppercase_mapping = ($16=='') ? (code==0x00df ? 0x1e9e : nil) : $16.hex + @lowercase_mapping = ($17=='') ? nil : $17.hex + @titlecase_mapping = ($18=='') ? (code==0x00df ? 0x1e9e : nil) : $18.hex + end + def case_folding + $case_folding[code] + end + def c_entry(comb_indicies) + " " << + "{#{str2c category, 'CATEGORY'}, #{combining_class}, " << + "#{str2c bidi_class, 'BIDI_CLASS'}, " << + "#{str2c decomp_type, 'DECOMP_TYPE'}, " << + "#{c_decomp_mapping}, " << + "#{c_case_folding}, " << + "#{singlecpmap uppercase_mapping }, " << + "#{singlecpmap lowercase_mapping }, " << + "#{singlecpmap titlecase_mapping }, " << + "#{comb_indicies[code] ? comb_indicies[code]: 'UINT16_MAX'}, " << + "#{bidi_mirrored}, " << + "#{$exclusions.include?(code) or $excl_version.include?(code)}, " << + "#{$ignorable.include?(code)}, " << + "#{%W[Zl Zp Cc Cf].include?(category) and not [0x200C, 0x200D].include?(category)}, " << + "#{$charwidth[code]}, 0, " << + "#{$grapheme_boundclass[code]}},\n" + end +end + +chars = [] +char_hash = {} + +while gets + if $_ =~ /^([0-9A-F]+);<[^;>,]+, First>;/i + first = $1.hex + gets + char = UnicodeChar.new($_) + raise "No last character of sequence found." unless + $_ =~ /^([0-9A-F]+);<([^;>,]+), Last>;/i + last = $1.hex + name = "<#{$2}>" + for i in first..last + char_clone = char.clone + char_clone.code = i + char_clone.name = name + char_hash[char_clone.code] = char_clone + chars << char_clone + end + else + char = UnicodeChar.new($_) + char_hash[char.code] = char + chars << char + end +end + +comb1st_indicies = {} +comb2nd_indicies = {} +comb2nd_indicies_sorted_keys = [] +comb2nd_indicies_nonbasic = {} +comb_array = [] + +chars.each do |char| + if !char.nil? and char.decomp_type.nil? and char.decomp_mapping and + char.decomp_mapping.length == 2 and !char_hash[char.decomp_mapping[0]].nil? and + char_hash[char.decomp_mapping[0]].combining_class == 0 and + not $exclusions.include?(char.code) + + dm0 = char.decomp_mapping[0] + dm1 = char.decomp_mapping[1] + unless comb1st_indicies[dm0] + comb1st_indicies[dm0] = comb1st_indicies.keys.length + end + unless comb2nd_indicies[dm1] + comb2nd_indicies_sorted_keys << dm1 + comb2nd_indicies[dm1] = comb2nd_indicies.keys.length + end + comb_array[comb1st_indicies[dm0]] ||= [] + raise "Duplicate canonical mapping: #{char.code} #{dm0} #{dm1}" if comb_array[comb1st_indicies[dm0]][comb2nd_indicies[dm1]] + comb_array[comb1st_indicies[dm0]][comb2nd_indicies[dm1]] = char.code + + comb2nd_indicies_nonbasic[dm1] = true if char.code > 0xFFFF + end + char.c_decomp_mapping = cpary2c(char.decomp_mapping) + char.c_case_folding = cpary2c(char.case_folding) +end + +comb_indicies = {} +cumoffset = 0 +comb1st_indicies_lastoffsets = [] +comb1st_indicies_firstoffsets = [] +comb1st_indicies.each do |dm0, index| + first = nil + last = nil + offset = 0 + comb2nd_indicies_sorted_keys.each_with_index do |dm1, b| + if comb_array[index][b] + first = offset unless first + last = offset + last += 1 if comb2nd_indicies_nonbasic[dm1] + end + offset += 1 + offset += 1 if comb2nd_indicies_nonbasic[dm1] + end + comb1st_indicies_firstoffsets[index] = first + comb1st_indicies_lastoffsets[index] = last + raise "double index" if comb_indicies[dm0] + comb_indicies[dm0] = cumoffset + cumoffset += last - first + 1 + 2 +end + +offset = 0 +comb2nd_indicies_sorted_keys.each do |dm1| + raise "double index" if comb_indicies[dm1] + comb_indicies[dm1] = 0x8000 | (comb2nd_indicies[dm1] + offset) + raise "too large comb index" if comb2nd_indicies[dm1] + offset > 0x4000 + if comb2nd_indicies_nonbasic[dm1] + comb_indicies[dm1] = comb_indicies[dm1] | 0x4000 + offset += 1 + end +end + +properties_indicies = {} +properties = [] +chars.each do |char| + c_entry = char.c_entry(comb_indicies) + char.c_entry_index = properties_indicies[c_entry] + unless char.c_entry_index + properties_indicies[c_entry] = properties.length + char.c_entry_index = properties.length + properties << c_entry + end +end + +stage1 = [] +stage2 = [] +for code in 0...0x110000 + next unless code % 0x100 == 0 + stage2_entry = [] + for code2 in code...(code+0x100) + if char_hash[code2] + stage2_entry << (char_hash[code2].c_entry_index + 1) + else + stage2_entry << 0 + end + end + old_index = stage2.index(stage2_entry) + if old_index + stage1 << (old_index * 0x100) + else + stage1 << (stage2.length * 0x100) + stage2 << stage2_entry + end +end + +$stdout << "static const utf8proc_uint16_t utf8proc_sequences[] = {\n " +i = 0 +$int_array.each do |entry| + i += 1 + if i == 8 + i = 0 + $stdout << "\n " + end + $stdout << entry << ", " +end +$stdout << "};\n\n" + +$stdout << "static const utf8proc_uint16_t utf8proc_stage1table[] = {\n " +i = 0 +stage1.each do |entry| + i += 1 + if i == 8 + i = 0 + $stdout << "\n " + end + $stdout << entry << ", " +end +$stdout << "};\n\n" + +$stdout << "static const utf8proc_uint16_t utf8proc_stage2table[] = {\n " +i = 0 +stage2.flatten.each do |entry| + i += 1 + if i == 8 + i = 0 + $stdout << "\n " + end + $stdout << entry << ", " +end +$stdout << "};\n\n" + +$stdout << "static const utf8proc_property_t utf8proc_properties[] = {\n" +$stdout << " {0, 0, 0, 0, UINT16_MAX, UINT16_MAX, UINT16_MAX, UINT16_MAX, UINT16_MAX, UINT16_MAX, false,false,false,false, 1, 0, UTF8PROC_BOUNDCLASS_OTHER},\n" +properties.each { |line| + $stdout << line +} +$stdout << "};\n\n" + + + +$stdout << "static const utf8proc_uint16_t utf8proc_combinations[] = {\n " +i = 0 +comb1st_indicies.keys.each_index do |a| + offset = 0 + $stdout << comb1st_indicies_firstoffsets[a] << ", " << comb1st_indicies_lastoffsets[a] << ", " + comb2nd_indicies_sorted_keys.each_with_index do |dm1, b| + break if offset > comb1st_indicies_lastoffsets[a] + if offset >= comb1st_indicies_firstoffsets[a] + i += 1 + if i == 8 + i = 0 + $stdout << "\n " + end + v = comb_array[a][b] ? comb_array[a][b] : 0 + $stdout << (( v & 0xFFFF0000 ) >> 16) << ", " if comb2nd_indicies_nonbasic[dm1] + $stdout << (v & 0xFFFF) << ", " + end + offset += 1 + offset += 1 if comb2nd_indicies_nonbasic[dm1] + end + $stdout << "\n" +end +$stdout << "};\n\n" -- cgit v1.2.3