mirror of
https://github.com/tateisu/SubwayTooter
synced 2024-12-30 18:57:30 +01:00
291 lines
7.9 KiB
Perl
291 lines
7.9 KiB
Perl
#!perl --
|
|
use strict;
|
|
use warnings;
|
|
use utf8;
|
|
use LWP::Simple;
|
|
use JSON;
|
|
use Data::Dump qw(dump);
|
|
use Encode;
|
|
use File::Copy;
|
|
|
|
=tmp
|
|
|
|
pngフォルダにある画像ファイルを参照する
|
|
emoji-data/emoji.json を参照する
|
|
|
|
以下のjavaコードを生成する
|
|
- UTF-16文字列 => 画像リソースID のマップ。同一のIDに複数のUTF-16文字列が振られることがある。
|
|
- shortcode => 画像リソースID のマップ。同一のIDに複数のshortcodeが振られることがある。
|
|
- shortcode中の区切り文字はハイフンもアンダーバーもありうる。出力データではアンダーバーに寄せる
|
|
- アプリはshortcodeの探索時にキー文字列の区切り文字をアンダーバーに正規化すること
|
|
|
|
=cut
|
|
|
|
|
|
|
|
sub loadFile{
|
|
my($fname)=@_;
|
|
open(my $fh,"<",$fname) or die "$fname $!";
|
|
local $/ = undef;
|
|
my $data = <$fh>;
|
|
close($fh) or die "$fname $!";
|
|
return $data;
|
|
}
|
|
|
|
sub parseCodePoint($){
|
|
my($src)=@_;
|
|
return () if not $src;
|
|
|
|
my @chars = map{ lc $_ } ( $src =~ /([0-9A-Fa-f]+)/g );
|
|
return () if not @chars;
|
|
|
|
return \@chars;
|
|
}
|
|
|
|
sub encodeCodePoint($){
|
|
my($chars)=@_;
|
|
return join '-', @$chars;
|
|
}
|
|
|
|
sub parseShortName($){
|
|
my( $name) = @_;
|
|
$name = lc $name;
|
|
$name =~ tr/-/_/;
|
|
|
|
return $name;
|
|
}
|
|
|
|
my @emoji_variants = qw(
|
|
img-twitter-64
|
|
img-google-64
|
|
img-apple-64
|
|
img-apple-160
|
|
img-facebook-64
|
|
img-messenger-64
|
|
);
|
|
|
|
my %emoji_variants_used;
|
|
|
|
sub findEmojiImage($){
|
|
my($image)=@_;
|
|
for my $variant ( @emoji_variants ){
|
|
my $path = "emoji-data/$variant/$image";
|
|
if( -f $path ){
|
|
$emoji_variants_used{$variant} or $emoji_variants_used{$variant} =[];
|
|
if( @{$emoji_variants_used{$variant}} < 5 ){
|
|
push @{$emoji_variants_used{$variant}},$image;
|
|
}
|
|
return $path;
|
|
}
|
|
}
|
|
return;
|
|
}
|
|
|
|
sub getEmojiResId($$){
|
|
my($image,$name)=@_;
|
|
|
|
# コードポイントに合う画像ファイルがあるか調べる
|
|
my $image_path = findEmojiImage($image);
|
|
if( not $image_path ){
|
|
warn "$name : missing image. $image\n";
|
|
next;
|
|
}
|
|
|
|
# 画像ファイルをpng フォルダにコピーする
|
|
my $dst_name = "emj_". lc($image);
|
|
$dst_name =~ tr/-/_/;
|
|
my $dst_path = "png/$dst_name";
|
|
if( not -f $dst_path ){
|
|
copy( $image_path,$dst_path ) or die "$dst_path $!";
|
|
}
|
|
|
|
# 画像リソースの名前
|
|
my $res_name = $dst_name;
|
|
$res_name =~ s/\.png//;
|
|
|
|
return $res_name;
|
|
}
|
|
|
|
my @skin_tone_modifier = (
|
|
[ "1F3FB" , "_tone1" ],
|
|
[ "1F3FC" , "_tone2" ],
|
|
[ "1F3FD" , "_tone3" ],
|
|
[ "1F3FE" , "_tone4" ],
|
|
[ "1F3FF" , "_tone5" ],
|
|
);
|
|
|
|
my %res_map;
|
|
|
|
|
|
|
|
|
|
my $emoji_list = decode_json loadFile "./emoji-data/emoji.json";
|
|
for my $emoji ( @$emoji_list ){
|
|
|
|
# short_name のリストを作る
|
|
my @shortnames;
|
|
push @shortnames,map{ parseShortName($_) } $emoji->{"short_name"};
|
|
push @shortnames,map{ parseShortName($_) } @{ $emoji->{"short_names"} };
|
|
|
|
# 絵文字のコードポイント一覧を収集する
|
|
my @codepoints;
|
|
push @codepoints,map{ parseCodePoint($_) } $emoji->{unified};
|
|
push @codepoints,map{ parseCodePoint($_) } @{ $emoji->{variations} };
|
|
for my $k (qw(docomo au softbank google) ){
|
|
push @codepoints,map{ parseCodePoint($_) } $emoji->{$k};
|
|
}
|
|
|
|
my $name = $shortnames[0];
|
|
my $res_name = getEmojiResId($emoji->{"image"},$name);
|
|
my $res_info = $res_map{ $res_name };
|
|
$res_info or $res_info = $res_map{ $res_name } = { codepoint_map => {} , shortname_map => {} };
|
|
for ( @codepoints ){
|
|
$res_info->{codepoint_map}{ encodeCodePoint($_) } = $_;
|
|
}
|
|
|
|
for ( @shortnames ){
|
|
$res_info->{shortname_map}{ $_ } = $_;
|
|
}
|
|
|
|
# スキントーン
|
|
if( $emoji->{"skin_variations"} ){
|
|
for my $mod (@skin_tone_modifier){
|
|
my($mod_code,$mod_suffix)=@$mod;
|
|
my $mod_name = $name . $mod_suffix;
|
|
|
|
my $data = $emoji->{"skin_variations"}{$mod_code};
|
|
if( not $data ){
|
|
warn "$name : missing skin tone $mod_code $mod_suffix\n";
|
|
next;
|
|
}
|
|
|
|
$res_name = getEmojiResId($data->{"image"},$mod_name);
|
|
my $res_info = $res_map{ $res_name };
|
|
$res_info or $res_info = $res_map{ $res_name } = { codepoint_map => {} , shortname_map => {} };
|
|
|
|
for ( map{ parseCodePoint($_) } $data->{unified} ){
|
|
$res_info->{codepoint_map}{ encodeCodePoint($_) } = $_;
|
|
}
|
|
for ( map{ $_ . $mod_suffix } @shortnames ){
|
|
$res_info->{shortname_map}{ $_ } = $_;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
for my $variant ( @emoji_variants ){
|
|
next if not $emoji_variants_used{$variant};
|
|
warn "$variant ",join(',',@{$emoji_variants_used{$variant}})," ...\n";
|
|
}
|
|
|
|
my $out_file = "EmojiData201709.java";
|
|
open(my $fh, ">:encoding(utf8)",$out_file) or die "$out_file : $!";
|
|
|
|
my $line_num = 0;
|
|
my $func_num = 0;
|
|
sub addCode{
|
|
my($code)=@_;
|
|
# open new function
|
|
if( $line_num == 0 ){
|
|
++$func_num;
|
|
print $fh "\tprivate static void init$func_num(){\n";
|
|
}
|
|
# write code
|
|
print $fh "\t\t",$code,"\n";
|
|
|
|
# close function
|
|
if( ++ $line_num > 100 ){
|
|
print $fh "\t}\n";
|
|
$line_num = 0;
|
|
}
|
|
}
|
|
|
|
my $utf8 = Encode::find_encoding("utf8");
|
|
my $utf16 = Encode::find_encoding("UTF-16BE");
|
|
my $utf16_max_length = 0;
|
|
|
|
my %codepoint_duplicate;
|
|
my %shortname_duplicate;
|
|
|
|
for my $res_name ( sort keys %res_map ){
|
|
my $res_info = $res_map{$res_name};
|
|
|
|
my $java = "";
|
|
for my $codepoint_name( sort keys %{$res_info->{codepoint_map}} ){
|
|
my $codepoint_chars = $res_info->{codepoint_map}{$codepoint_name};
|
|
|
|
my $dup = $codepoint_duplicate{ $codepoint_name };
|
|
$dup or $dup = $codepoint_duplicate{ $codepoint_name } = [];
|
|
push @$dup,$res_name;
|
|
next if @$dup > 1;
|
|
# 一つのコードポイントに複数の絵文字グリフが割り当てられているケース
|
|
# - 過去のガラケー絵文字の「時計」が複数の時刻の時計にマッチする
|
|
# - 過去のガラケー絵文字の「占い」が水晶玉や魔法陣にマッチする
|
|
# など、現代では別に気にしなくてもよさそうなケースだった
|
|
|
|
# コードポイントのリストからperl内部表現の文字列にする
|
|
my $str = join '',map{ chr hex $_ } @$codepoint_chars;
|
|
# perl内部表現からUTF-16に変換する
|
|
my $str_utf16 = $utf16->encode( $str );
|
|
# $str_utf16 をJavaのエスケープ表現に直す
|
|
my @utf16_chars = unpack("n*",$str_utf16);
|
|
my $java_chars = join('',map{ sprintf qq(\\u%04x),$_} @utf16_chars );
|
|
$java .= qq(, "$java_chars");
|
|
|
|
my $char_count = 0+@utf16_chars;
|
|
if( $char_count > $utf16_max_length ){
|
|
$utf16_max_length = $char_count;
|
|
}
|
|
if( $char_count == 11 ){
|
|
warn "max length: $res_name\n";
|
|
}
|
|
}
|
|
addCode( qq{_addCodePoints( R.drawable.$res_name$java );});
|
|
}
|
|
for my $res_name ( sort keys %res_map ){
|
|
my $res_info = $res_map{$res_name};
|
|
|
|
my $java = "";
|
|
for my $short_name ( sort keys %{$res_info->{shortname_map}} ){
|
|
|
|
my $dup = $shortname_duplicate{ $short_name };
|
|
$dup or $dup = $shortname_duplicate{ $short_name } = [];
|
|
push @$dup,$res_name;
|
|
next if @$dup > 1;
|
|
# 一つのshort_name に複数の絵文字グリフが割り当てられているケース
|
|
# family, man_woman_boy だけだった
|
|
# emj_1f468_200d_1f469_200d_1f466, emj_1f46a の２パターンがある
|
|
# 図柄は全く同じに見えるので気にしなくてもよさそうだった
|
|
|
|
|
|
$java .= qq(, "$short_name");
|
|
}
|
|
addCode( qq{_addShortNames( R.drawable.$res_name$java );});
|
|
}
|
|
|
|
while(my($k,$ra)=each %codepoint_duplicate){
|
|
next if @$ra <= 1;
|
|
warn "codepoint duplicate $k : ",join(',',@$ra),"\n";
|
|
}
|
|
|
|
while(my($k,$ra)=each %shortname_duplicate){
|
|
next if @$ra <= 1;
|
|
warn "shortname duplicate $k : ",join(',',@$ra),"\n";
|
|
}
|
|
|
|
# close function
|
|
if( $line_num > 0 ){
|
|
print $fh "\t}\n";
|
|
}
|
|
|
|
# write function to call init**()
|
|
|
|
print $fh "\tstatic final int utf16_max_length=$utf16_max_length;\n\n";
|
|
print $fh "\tstatic void initAll(){\n";
|
|
for(my $i=1;$i <= $func_num;++$i){
|
|
print $fh "\t\tinit$i();\n";
|
|
}
|
|
print $fh "\t}\n";
|
|
|
|
close($fh) or die "$out_file : $!";
|