| 1 | #!/usr/bin/env perl | ||
| 2 | |||
| 3 | use strict; | ||
| 4 | use utf8; | ||
| 5 | use warnings; | ||
| 6 | use FindBin qw( $Bin ); | ||
| 7 | use lib "$Bin/../lib"; | ||
| 8 | |||
| 9 | use Zuzu::Lexer; | ||
| 10 | use Zuzu::Parser; | ||
| 11 | |||
| 12 | my %WORD_OPERATOR_KW = map { $_ => 1 } qw( | ||
| 13 | and and? or or? xor xor? nand nand? nor nor? xnor xnor? onlyif onlyif? butnot butnot? not | ||
| 14 | eq ne gt ge lt le cmp eqi nei gti gei lti lei cmpi | ||
| 15 | mod abs sqrt floor ceil round int length uc lc typeof instanceof does can | ||
| 16 | union intersection subsetof supersetof equivalentof divides | ||
| 17 | in | ||
| 18 | ); | ||
| 19 | |||
| 20 | binmode STDOUT, ':encoding(UTF-8)'; | ||
| 21 | binmode STDERR, ':encoding(UTF-8)'; | ||
| 22 | |||
| 23 | my $source = read_source(@ARGV); | ||
| 24 | my ( $body, $parsed, $parse_error ) = highlight($source); | ||
| 25 | print render_html( $body, $parsed, $parse_error ); | ||
| 26 | |||
| 27 | sub read_source { | ||
| 28 | my (@args) = @_; | ||
| 29 | |||
| 30 | if ( @args > 1 ) { | ||
| 31 | die "usage: $0 [file.zzs]\n"; | ||
| 32 | } | ||
| 33 | |||
| 34 | if ( @args == 1 and $args[0] ne '-' ) { | ||
| 35 | my $path = $args[0]; | ||
| 36 | open my $fh, '<:encoding(UTF-8)', $path | ||
| 37 | or die "failed to read '$path': $!\n"; | ||
| 38 | local $/ = undef; | ||
| 39 | |||
| 40 | return <$fh>; | ||
| 41 | } | ||
| 42 | |||
| 43 | binmode STDIN, ':encoding(UTF-8)'; | ||
| 44 | local $/ = undef; | ||
| 45 | |||
| 46 | return <STDIN> // ''; | ||
| 47 | } | ||
| 48 | |||
| 49 | sub highlight { | ||
| 50 | my ($src) = @_; | ||
| 51 | my @out; | ||
| 52 | my @tokens = lex_tokens($src); | ||
| 53 | my @line_offsets = line_offsets($src); | ||
| 54 | my $cursor = 0; | ||
| 55 | |||
| 56 | for my $idx ( 0 .. $#tokens ) { | ||
| 57 | my $entry = $tokens[$idx]; | ||
| 58 | my $tok = $entry->{token}; | ||
| 59 | my $start = offset_for_token( \@line_offsets, $entry ); | ||
| 60 | if ( $start > $cursor ) { | ||
| 61 | push @out, esc_html( substr( $src, $cursor, $start - $cursor ) ); | ||
| 62 | } | ||
| 63 | |||
| 64 | my $lexeme = extract_lexeme( $src, $start, $tok ); | ||
| 65 | $lexeme = substr( $src, $start, 1 ) if $lexeme eq ''; | ||
| 66 | $cursor = $start + length($lexeme); | ||
| 67 | |||
| 68 | my $class = class_for_token( $tok, \@tokens, $idx ); | ||
| 69 | if ( defined $class ) { | ||
| 70 | push @out, span( $class, $lexeme ); | ||
| 71 | } | ||
| 72 | else { | ||
| 73 | push @out, esc_html($lexeme); | ||
| 74 | } | ||
| 75 | } | ||
| 76 | |||
| 77 | push @out, esc_html( substr( $src, $cursor ) ) if $cursor < length($src); | ||
| 78 | my ( $parsed, $parse_error ) = parse_check($src); | ||
| 79 | |||
| 80 | return ( join( '', @out ), $parsed, $parse_error ); | ||
| 81 | } | ||
| 82 | |||
| 83 | sub lex_tokens { | ||
| 84 | my ($src) = @_; | ||
| 85 | my $lexer = Zuzu::Lexer->new( src => $src, filename => '<highlight>' ); | ||
| 86 | my @tokens; | ||
| 87 | |||
| 88 | while (1) { | ||
| 89 | my $tok = $lexer->next_token; | ||
| 90 | last if $tok->type eq 'EOF'; | ||
| 91 | push @tokens, { | ||
| 92 | token => $tok, | ||
| 93 | line => $tok->line, | ||
| 94 | col => $tok->col, | ||
| 95 | }; | ||
| 96 | } | ||
| 97 | |||
| 98 | return @tokens; | ||
| 99 | } | ||
| 100 | |||
| 101 | sub parse_check { | ||
| 102 | my ($src) = @_; | ||
| 103 | my $parser = Zuzu::Parser->new; | ||
| 104 | my $ok = eval { | ||
| 105 | $parser->parse( $src, '<highlight>' ); | ||
| 106 | 1; | ||
| 107 | }; | ||
| 108 | if ($ok) { | ||
| 109 | return ( 1, '' ); | ||
| 110 | } | ||
| 111 | else { | ||
| 112 | my $err = $@; | ||
| 113 | $err = "$err"; | ||
| 114 | $err =~ s/\s+\z//; | ||
| 115 | |||
| 116 | ( 0, $err ); | ||
| 117 | } | ||
| 118 | } | ||
| 119 | |||
| 120 | sub line_offsets { | ||
| 121 | my ($src) = @_; | ||
| 122 | my @offsets = ( 0 ); | ||
| 123 | my $pos = 0; | ||
| 124 | |||
| 125 | while ( $src =~ /\n/g ) { | ||
| 126 | $pos = pos($src); | ||
| 127 | push @offsets, $pos; | ||
| 128 | } | ||
| 129 | |||
| 130 | return @offsets; | ||
| 131 | } | ||
| 132 | |||
| 133 | sub offset_for_token { | ||
| 134 | my ( $line_offsets, $entry ) = @_; | ||
| 135 | my $line_start = $line_offsets->[ $entry->{line} - 1 ] // 0; | ||
| 136 | |||
| 137 | return $line_start + $entry->{col} - 1; | ||
| 138 | } | ||
| 139 | |||
| 140 | sub extract_lexeme { | ||
| 141 | my ( $src, $start, $tok ) = @_; | ||
| 142 | my $type = $tok->type // ''; | ||
| 143 | my $rest = substr( $src, $start ); | ||
| 144 | |||
| 145 | if ( $type eq 'KW' ) { | ||
| 146 | return $tok->value // ''; | ||
| 147 | } | ||
| 148 | if ( $type eq 'IDENT' ) { | ||
| 149 | return $tok->value // ''; | ||
| 150 | } | ||
| 151 | if ( $type eq 'BOOL' ) { | ||
| 152 | return ( $rest =~ /\A(?:true|false|⊤|⊥)/u ) ? $& : ''; | ||
| 153 | } | ||
| 154 | if ( $type eq 'NULL' ) { | ||
| 155 | return ( $rest =~ /\Anull/u ) ? 'null' : ''; | ||
| 156 | } | ||
| 157 | if ( $type eq 'NUMBER' ) { | ||
| 158 | return ( | ||
| 159 | $rest =~ /\A(?:0x[0-9A-Fa-f]+|0b[01]+|0o[0-7]+|[0-9]+(?:\.[0-9]+)?(?:E[+-]?[0-9]+)?)/u | ||
| 160 | ) ? $& : ''; | ||
| 161 | } | ||
| 162 | if ( $type eq 'STRING' ) { | ||
| 163 | return ( $rest =~ /\A"""(?s:(?:\\.|(?!""").))*"""|\A"(?:\\.|[^"\\])*"?/u ) ? $& : ''; | ||
| 164 | } | ||
| 165 | if ( $type eq 'BINARY_STRING' ) { | ||
| 166 | return ( $rest =~ /\A'''(?s:(?:\\.|(?!''').))*'''|\A'(?:\\.|[^'\\])*'?/u ) ? $& : ''; | ||
| 167 | } | ||
| 168 | if ( $type eq 'TEMPLATE' ) { | ||
| 169 | return ( $rest =~ /\A```(?s:(?:\\.|(?!```).))*```|\A`(?:\\.|[^`\\])*`/u ) ? $& : ''; | ||
| 170 | } | ||
| 171 | if ( $type eq 'REGEXP' ) { | ||
| 172 | return extract_regexp_lexeme($rest); | ||
| 173 | } | ||
| 174 | if ( $type eq 'EMPTY_SET' ) { | ||
| 175 | return '∅'; | ||
| 176 | } | ||
| 177 | if ( $type eq 'OP' ) { | ||
| 178 | my $op = $tok->value // ''; | ||
| 179 | return $op if $op ne '' and index( $rest, $op ) == 0; | ||
| 180 | } | ||
| 181 | |||
| 182 | return ''; | ||
| 183 | } | ||
| 184 | |||
| 185 | sub extract_regexp_lexeme { | ||
| 186 | my ($rest) = @_; | ||
| 187 | return '' if substr( $rest, 0, 1 ) ne '/'; | ||
| 188 | |||
| 189 | my $len = length($rest); | ||
| 190 | my $pos = 1; | ||
| 191 | my $escaped = 0; | ||
| 192 | my $in_class = 0; | ||
| 193 | |||
| 194 | while ( $pos < $len ) { | ||
| 195 | my $c = substr( $rest, $pos, 1 ); | ||
| 196 | if ( !$escaped and !$in_class and $c eq '/' ) { | ||
| 197 | $pos++; | ||
| 198 | my $flags = ''; | ||
| 199 | while ( $pos < $len ) { | ||
| 200 | my $flag = substr( $rest, $pos, 1 ); | ||
| 201 | last if $flag ne 'i' and $flag ne 'g'; | ||
| 202 | last if index( $flags, $flag ) >= 0; | ||
| 203 | $flags .= $flag; | ||
| 204 | $pos++; | ||
| 205 | } | ||
| 206 | |||
| 207 | return substr( $rest, 0, $pos ); | ||
| 208 | } | ||
| 209 | if ( !$escaped and $c eq '[' ) { | ||
| 210 | $in_class = 1; | ||
| 211 | } | ||
| 212 | elsif ( !$escaped and $c eq ']' ) { | ||
| 213 | $in_class = 0; | ||
| 214 | } | ||
| 215 | $escaped = !$escaped && $c eq "\\" ? 1 : 0; | ||
| 216 | $pos++; | ||
| 217 | } | ||
| 218 | |||
| 219 | return ''; | ||
| 220 | } | ||
| 221 | |||
| 222 | sub class_for_token { | ||
| 223 | my ( $tok, $tokens, $idx ) = @_; | ||
| 224 | my $type = $tok->type // ''; | ||
| 225 | |||
| 226 | if ( $type eq 'KW' and ( $tok->value // '' ) eq 'default' ) { | ||
| 227 | my $next = next_significant_token( $tokens, $idx ); | ||
| 228 | return 'keyword' | ||
| 229 | if defined $next and $next->type eq 'OP' and ( $next->value // '' ) eq ':'; | ||
| 230 | |||
| 231 | return 'operator'; | ||
| 232 | } | ||
| 233 | if ( $type eq 'KW' and $WORD_OPERATOR_KW{ $tok->value // '' } ) { | ||
| 234 | return 'operator'; | ||
| 235 | } | ||
| 236 | return 'keyword' if $type eq 'KW'; | ||
| 237 | return 'boolean' if $type eq 'BOOL'; | ||
| 238 | return 'null' if $type eq 'NULL'; | ||
| 239 | return 'number' if $type eq 'NUMBER'; | ||
| 240 | return 'string' if $type eq 'STRING' or $type eq 'BINARY_STRING' or $type eq 'TEMPLATE'; | ||
| 241 | return 'regexp' if $type eq 'REGEXP'; | ||
| 242 | return 'operator' if $type eq 'OP' or $type eq 'EMPTY_SET'; | ||
| 243 | if ( $type eq 'IDENT' ) { | ||
| 244 | if ( $idx > 0 ) { | ||
| 245 | my $prev = $tokens->[ $idx - 1 ]{token}; | ||
| 246 | if ( $prev->type eq 'KW' ) { | ||
| 247 | my $kw = $prev->value // ''; | ||
| 248 | return 'ident-decl' if $kw =~ /\A(?:function|method|class|trait|catch)\z/u; | ||
| 249 | } | ||
| 250 | } | ||
| 251 | |||
| 252 | return 'ident'; | ||
| 253 | } | ||
| 254 | |||
| 255 | return undef; | ||
| 256 | } | ||
| 257 | |||
| 258 | sub next_significant_token { | ||
| 259 | my ( $tokens, $idx ) = @_; | ||
| 260 | |||
| 261 | for my $look ( $idx + 1 .. $#$tokens ) { | ||
| 262 | my $tok = $tokens->[$look]{token}; | ||
| 263 | my $type = $tok->type // ''; | ||
| 264 | return $tok if $type ne 'COMMENT' and $type ne 'POD'; | ||
| 265 | } | ||
| 266 | |||
| 267 | return undef; | ||
| 268 | } | ||
| 269 | |||
| 270 | sub span { | ||
| 271 | my ( $class, $text ) = @_; | ||
| 272 | |||
| 273 | return qq{<span class="$class">} . esc_html($text) . '</span>'; | ||
| 274 | } | ||
| 275 | |||
| 276 | sub esc_html { | ||
| 277 | my ($text) = @_; | ||
| 278 | |||
| 279 | $text =~ s/&/&/g; | ||
| 280 | $text =~ s/</</g; | ||
| 281 | $text =~ s/>/>/g; | ||
| 282 | |||
| 283 | return $text; | ||
| 284 | } | ||
| 285 | |||
| 286 | sub render_html { | ||
| 287 | my ( $body, $parsed, $parse_error ) = @_; | ||
| 288 | my $status_class = $parsed ? 'parse-ok' : 'parse-error'; | ||
| 289 | my $status_msg = $parsed ? 'Parse check: ok' : 'Parse check: failed'; | ||
| 290 | my $status_detail = $parsed ? '' : '<div class="parse-detail">' . esc_html($parse_error) . '</div>'; | ||
| 291 | my $head = <<'HTML'; | ||
| 292 | <!doctype html> | ||
| 293 | <html lang="en"> | ||
| 294 | <head> | ||
| 295 | <meta charset="utf-8"> | ||
| 296 | <meta name="viewport" content="width=device-width, initial-scale=1"> | ||
| 297 | <title>Zuzu Syntax Highlight</title> | ||
| 298 | <style> | ||
| 299 | :root { | ||
| 300 | color-scheme: light dark; | ||
| 301 | --bg: #F4F1E8; | ||
| 302 | --surface: #F8F5EE; | ||
| 303 | --border: #D6D0C4; | ||
| 304 | --text: #2E2E2E; | ||
| 305 | --comment: #7C7670; | ||
| 306 | --keyword: #8A9EFF; | ||
| 307 | --string: #B06E8A; | ||
| 308 | --number: #A77755; | ||
| 309 | --operator: #6C78C8; | ||
| 310 | --ident: #2E2E2E; | ||
| 311 | --boolean: #F2A7A7; | ||
| 312 | --null: #D98989; | ||
| 313 | --regexp: #7BA6B0; | ||
| 314 | } | ||
| 315 | |||
| 316 | @media (prefers-color-scheme: dark) { | ||
| 317 | :root { | ||
| 318 | --bg: #2E2E2E; | ||
| 319 | --surface: #363636; | ||
| 320 | --border: #4A4A4A; | ||
| 321 | --text: #F4F1E8; | ||
| 322 | --comment: #B9B2A5; | ||
| 323 | --keyword: #8A9EFF; | ||
| 324 | --string: #E0AFC4; | ||
| 325 | --number: #D6B28E; | ||
| 326 | --operator: #AAB6FF; | ||
| 327 | --ident: #F4F1E8; | ||
| 328 | --boolean: #F2A7A7; | ||
| 329 | --null: #FFC0C0; | ||
| 330 | --regexp: #9FC9D2; | ||
| 331 | } | ||
| 332 | } | ||
| 333 | |||
| 334 | body { | ||
| 335 | margin: 0; | ||
| 336 | padding: 2rem; | ||
| 337 | background: var(--bg); | ||
| 338 | color: var(--text); | ||
| 339 | font-family: Inter, ui-sans-serif, system-ui, -apple-system, | ||
| 340 | "Segoe UI", Roboto, Helvetica, Arial, sans-serif; | ||
| 341 | } | ||
| 342 | |||
| 343 | pre.zuzu-code { | ||
| 344 | margin: 0; | ||
| 345 | padding: 1.25rem 1.5rem; | ||
| 346 | background: var(--surface); | ||
| 347 | border: 1px solid var(--border); | ||
| 348 | border-radius: 10px; | ||
| 349 | overflow-x: auto; | ||
| 350 | line-height: 1.55; | ||
| 351 | font-size: 0.95rem; | ||
| 352 | font-family: "JetBrains Mono", "SFMono-Regular", Menlo, | ||
| 353 | Consolas, "Liberation Mono", monospace; | ||
| 354 | } | ||
| 355 | |||
| 356 | .comment { color: var(--comment); font-style: italic; } | ||
| 357 | .keyword { color: var(--keyword); font-weight: 600; } | ||
| 358 | .string { color: var(--string); } | ||
| 359 | .number { color: var(--number); } | ||
| 360 | .operator { color: var(--operator); } | ||
| 361 | .ident { color: var(--ident); } | ||
| 362 | .boolean { color: var(--boolean); font-weight: 600; } | ||
| 363 | .null { color: var(--null); font-weight: 600; } | ||
| 364 | .regexp { color: var(--regexp); } | ||
| 365 | .ident-decl { color: var(--keyword); font-weight: 700; } | ||
| 366 | .parse-status { margin-bottom: 0.8rem; font-size: 0.85rem; } | ||
| 367 | .parse-ok { color: #4C7A5C; } | ||
| 368 | .parse-error { color: #B85050; } | ||
| 369 | .parse-detail { font-family: monospace; margin-top: 0.25rem; white-space: pre-wrap; } | ||
| 370 | </style> | ||
| 371 | </head> | ||
| 372 | <body> | ||
| 373 | HTML | ||
| 374 | my $status = qq{<div class="parse-status $status_class">$status_msg$status_detail</div>\n}; | ||
| 375 | my $open_pre = <<'HTML'; | ||
| 376 | <pre class="zuzu-code"> | ||
| 377 | HTML | ||
| 378 | my $tail = <<'HTML'; | ||
| 379 | </pre> | ||
| 380 | </body> | ||
| 381 | </html> | ||
| 382 | HTML | ||
| 383 | |||
| 384 | return $head . $status . $open_pre . $body . $tail; | ||
| 385 | } |