[13983] | 1 | package HTML::TokeParser::Simple;
|
---|
| 2 |
|
---|
| 3 | use strict;
|
---|
| 4 | use HTML::TokeParser;
|
---|
| 5 | use HTML::TokeParser::Simple::Token;
|
---|
| 6 | use HTML::TokeParser::Simple::Token::Tag;
|
---|
| 7 | use HTML::TokeParser::Simple::Token::Tag::Start;
|
---|
| 8 | use HTML::TokeParser::Simple::Token::Tag::End;
|
---|
| 9 | use HTML::TokeParser::Simple::Token::Text;
|
---|
| 10 | use HTML::TokeParser::Simple::Token::Comment;
|
---|
| 11 | use HTML::TokeParser::Simple::Token::Declaration;
|
---|
| 12 | use HTML::TokeParser::Simple::Token::ProcessInstruction;
|
---|
| 13 |
|
---|
| 14 | use vars qw/ @ISA $VERSION $REVISION /;
|
---|
| 15 |
|
---|
| 16 | $REVISION = '$Id: Simple.pm 13983 2007-03-15 01:32:44Z lh92 $';
|
---|
| 17 | $VERSION = '3.15';
|
---|
| 18 | @ISA = qw/ HTML::TokeParser /;
|
---|
| 19 |
|
---|
| 20 | # constructors
|
---|
| 21 |
|
---|
| 22 | my %FACTORY_CLASSES = (
|
---|
| 23 | S => 'HTML::TokeParser::Simple::Token::Tag::Start',
|
---|
| 24 | E => 'HTML::TokeParser::Simple::Token::Tag::End',
|
---|
| 25 | T => 'HTML::TokeParser::Simple::Token::Text',
|
---|
| 26 | C => 'HTML::TokeParser::Simple::Token::Comment',
|
---|
| 27 | D => 'HTML::TokeParser::Simple::Token::Declaration',
|
---|
| 28 | PI => 'HTML::TokeParser::Simple::Token::ProcessInstruction',
|
---|
| 29 | );
|
---|
| 30 |
|
---|
| 31 | sub _croak {
|
---|
| 32 | my ($proto, $message) = @_;
|
---|
| 33 | require Carp;
|
---|
| 34 | Carp::croak($message);
|
---|
| 35 | }
|
---|
| 36 |
|
---|
| 37 | sub new {
|
---|
| 38 | my ($class, @args) = @_;
|
---|
| 39 | return 1 == @args
|
---|
| 40 | ? $class->SUPER::new(@args)
|
---|
| 41 | : $class->_init(@args);
|
---|
| 42 | }
|
---|
| 43 |
|
---|
| 44 | sub _init {
|
---|
| 45 | my ($class, $source_type, $source) = @_;
|
---|
| 46 | my %sources = (
|
---|
| 47 | file => sub { $source },
|
---|
| 48 | handle => sub { $source },
|
---|
| 49 | string => sub { \$source },
|
---|
| 50 | url => sub {
|
---|
| 51 | eval "require LWP::Simple";
|
---|
| 52 | $class->_croak("Cannot load LWP::Simple: $@") if $@;
|
---|
| 53 | my $content = LWP::Simple::get($source);
|
---|
| 54 | $class->_croak("Could not fetch content from ($source)")
|
---|
| 55 | unless defined $content;
|
---|
| 56 | return \$content;
|
---|
| 57 | },
|
---|
| 58 | );
|
---|
| 59 | unless (exists $sources{$source_type}) {
|
---|
| 60 | $class->_croak("Unknown source type ($source_type)");
|
---|
| 61 | }
|
---|
| 62 | return $class->new($sources{$source_type}->());
|
---|
| 63 | }
|
---|
| 64 |
|
---|
| 65 | sub get_token {
|
---|
| 66 | my $self = shift;
|
---|
| 67 | my @args = @_;
|
---|
| 68 | my $token = $self->SUPER::get_token( @args );
|
---|
| 69 | return unless defined $token;
|
---|
| 70 | if (my $factory_class = $FACTORY_CLASSES{$token->[0]}) {
|
---|
| 71 | return $factory_class->new($token);
|
---|
| 72 | }
|
---|
| 73 | else {
|
---|
| 74 | # this should never happen
|
---|
| 75 | $self->_croak("Cannot determine token class for token (@$token)");
|
---|
| 76 | }
|
---|
| 77 | }
|
---|
| 78 |
|
---|
| 79 | sub get_tag {
|
---|
| 80 | my $self = shift;
|
---|
| 81 | my @args = @_;
|
---|
| 82 | my $token = $self->SUPER::get_tag( @args );
|
---|
| 83 | return unless defined $token;
|
---|
| 84 | return $token->[0] =~ /^\//
|
---|
| 85 | ? HTML::TokeParser::Simple::Token::Tag::End->new($token)
|
---|
| 86 | : HTML::TokeParser::Simple::Token::Tag::Start->new($token);
|
---|
| 87 | }
|
---|
| 88 |
|
---|
| 89 | sub peek {
|
---|
| 90 | my ($self, $count) = @_;
|
---|
| 91 | $count ||= 1;
|
---|
| 92 |
|
---|
| 93 | unless ($count =~ /^\d+$/) {
|
---|
| 94 | $self->_croak("Argument to peek() must be a positive integer, not ($count)");
|
---|
| 95 | }
|
---|
| 96 |
|
---|
| 97 | my $items = 0;
|
---|
| 98 | my $html = '';
|
---|
| 99 | my @tokens;
|
---|
| 100 | while ( $items++ < $count && defined ( my $token = $self->get_token ) ) {
|
---|
| 101 | $html .= $token->as_is;
|
---|
| 102 | push @tokens, $token;
|
---|
| 103 | }
|
---|
| 104 | $self->unget_token(@tokens);
|
---|
| 105 | return $html;
|
---|
| 106 | }
|
---|
| 107 |
|
---|
| 108 | 1;
|
---|
| 109 |
|
---|
| 110 | __END__
|
---|
| 111 |
|
---|
| 112 | =head1 NAME
|
---|
| 113 |
|
---|
| 114 | HTML::TokeParser::Simple - Easy to use C<HTML::TokeParser> interface
|
---|
| 115 |
|
---|
| 116 | =head1 SYNOPSIS
|
---|
| 117 |
|
---|
| 118 | use HTML::TokeParser::Simple;
|
---|
| 119 | my $p = HTML::TokeParser::Simple->new( $somefile );
|
---|
| 120 |
|
---|
| 121 | while ( my $token = $p->get_token ) {
|
---|
| 122 | # This prints all text in an HTML doc (i.e., it strips the HTML)
|
---|
| 123 | next unless $token->is_text;
|
---|
| 124 | print $token->as_is;
|
---|
| 125 | }
|
---|
| 126 |
|
---|
| 127 |
|
---|
| 128 | =head1 DESCRIPTION
|
---|
| 129 |
|
---|
| 130 | C<HTML::TokeParser> is an excellent module that's often used for parsing HTML.
|
---|
| 131 | However, the tokens returned are not exactly intuitive to parse:
|
---|
| 132 |
|
---|
| 133 | ["S", $tag, $attr, $attrseq, $text]
|
---|
| 134 | ["E", $tag, $text]
|
---|
| 135 | ["T", $text, $is_data]
|
---|
| 136 | ["C", $text]
|
---|
| 137 | ["D", $text]
|
---|
| 138 | ["PI", $token0, $text]
|
---|
| 139 |
|
---|
| 140 | To simplify this, C<HTML::TokeParser::Simple> allows the user ask more
|
---|
| 141 | intuitive (read: more self-documenting) questions about the tokens returned.
|
---|
| 142 |
|
---|
| 143 | You can also rebuild some tags on the fly. Frequently, the attributes
|
---|
| 144 | associated with start tags need to be altered, added to, or deleted. This
|
---|
| 145 | functionality is built in.
|
---|
| 146 |
|
---|
| 147 | Since this is a subclass of C<HTML::TokeParser>, all C<HTML::TokeParser>
|
---|
| 148 | methods are available. To truly appreciate the power of this module, please
|
---|
| 149 | read the documentation for C<HTML::TokeParser> and C<HTML::Parser>.
|
---|
| 150 |
|
---|
| 151 | =head1 CONTRUCTORS
|
---|
| 152 |
|
---|
| 153 | =head2 C<new($source)>
|
---|
| 154 |
|
---|
| 155 | The constructor for C<HTML::TokeParser::Simple> can be used just like
|
---|
| 156 | C<HTML::TokeParser>'s constructor:
|
---|
| 157 |
|
---|
| 158 | my $parser = HTML::TokeParser::Simple->new($filename);
|
---|
| 159 | # or
|
---|
| 160 | my $parser = HTML::TokeParser::Simple->new($filehandle);
|
---|
| 161 | # or
|
---|
| 162 | my $parser = HTML::TokeParser::Simple->new(\$html_string);
|
---|
| 163 |
|
---|
| 164 | =head2 C<new($source_type, $source)>
|
---|
| 165 |
|
---|
| 166 | If you wish to be more explicit, there is a new style of
|
---|
| 167 | constructor avaiable.
|
---|
| 168 |
|
---|
| 169 | my $parser = HTML::TokeParser::Simple->new(file => $filename);
|
---|
| 170 | # or
|
---|
| 171 | my $parser = HTML::TokeParser::Simple->new(handle => $filehandle);
|
---|
| 172 | # or
|
---|
| 173 | my $parser = HTML::TokeParser::Simple->new(string => $html_string);
|
---|
| 174 |
|
---|
| 175 | Note that you do not have to provide a reference for the string if using the
|
---|
| 176 | string constructor.
|
---|
| 177 |
|
---|
| 178 | As a convenience, you can also attempt to fetch the HTML directly from a URL.
|
---|
| 179 |
|
---|
| 180 | my $parser = HTML::TokeParser::Simple->new(url => 'http://some.url');
|
---|
| 181 |
|
---|
| 182 | This method relies on C<LWP::Simple>. If this module is not found or the page
|
---|
| 183 | cannot be fetched, the constructor will C<croak()>.
|
---|
| 184 |
|
---|
| 185 | =head1 PARSER METHODS
|
---|
| 186 |
|
---|
| 187 | =head2 get_token
|
---|
| 188 |
|
---|
| 189 | This method will return the next token that C<HTML::TokeParser::get_token()>
|
---|
| 190 | method would return. However, it will be blessed into a class appropriate
|
---|
| 191 | which represents the token type.
|
---|
| 192 |
|
---|
| 193 | =head2 get_tag
|
---|
| 194 |
|
---|
| 195 | This method will return the next token that C<HTML::TokeParser::get_tag()>
|
---|
| 196 | method would return. However, it will be blessed into either the
|
---|
| 197 | L<HTML::TokeParser::Simple::Token::Tag::Start> or
|
---|
| 198 | L<HTML::TokeParser::Simple::Token::Tag::End> class.
|
---|
| 199 |
|
---|
| 200 | =head2 peek
|
---|
| 201 |
|
---|
| 202 | As of version C<3.14>, you can now C<peek()> at the upcomings tokens without
|
---|
| 203 | affecting the state of the parser. By default, C<peek()> will return the text
|
---|
| 204 | of the next token, but specifying an integer C<$count> will return the text of
|
---|
| 205 | the next C<$count> tokens.
|
---|
| 206 |
|
---|
| 207 | This is useful when you're trying to debug where you are in a document.
|
---|
| 208 |
|
---|
| 209 | warn $parser->peek(3); # show the next 3 tokens
|
---|
| 210 |
|
---|
| 211 | =head1 ACCESSORS
|
---|
| 212 |
|
---|
| 213 | The following methods may be called on the token object which is returned,
|
---|
| 214 | not on the parser object.
|
---|
| 215 |
|
---|
| 216 | =head2 Boolean Accessors
|
---|
| 217 |
|
---|
| 218 | These accessors return true or false.
|
---|
| 219 |
|
---|
| 220 | =over 4
|
---|
| 221 |
|
---|
| 222 | =item * C<is_tag([$tag])>
|
---|
| 223 |
|
---|
| 224 | Use this to determine if you have any tag. An optional "tag type" may be
|
---|
| 225 | passed. This will allow you to match if it's a I<particular> tag. The
|
---|
| 226 | supplied tag is case-insensitive.
|
---|
| 227 |
|
---|
| 228 | if ( $token->is_tag ) { ... }
|
---|
| 229 |
|
---|
| 230 | Optionally, you may pass a regular expression as an argument.
|
---|
| 231 |
|
---|
| 232 | =item * C<is_start_tag([$tag])>
|
---|
| 233 |
|
---|
| 234 | Use this to determine if you have a start tag. An optional "tag type" may be
|
---|
| 235 | passed. This will allow you to match if it's a I<particular> start tag. The
|
---|
| 236 | supplied tag is case-insensitive.
|
---|
| 237 |
|
---|
| 238 | if ( $token->is_start_tag ) { ... }
|
---|
| 239 | if ( $token->is_start_tag( 'font' ) ) { ... }
|
---|
| 240 |
|
---|
| 241 | Optionally, you may pass a regular expression as an argument. To match all
|
---|
| 242 | header (h1, h2, ... h6) tags:
|
---|
| 243 |
|
---|
| 244 | if ( $token->is_start_tag( qr/^h[123456]$/ ) ) { ... }
|
---|
| 245 |
|
---|
| 246 | =item * C<is_end_tag([$tag])>
|
---|
| 247 |
|
---|
| 248 | Use this to determine if you have an end tag. An optional "tag type" may be
|
---|
| 249 | passed. This will allow you to match if it's a I<particular> end tag. The
|
---|
| 250 | supplied tag is case-insensitive.
|
---|
| 251 |
|
---|
| 252 | When testing for an end tag, the forward slash on the tag is optional.
|
---|
| 253 |
|
---|
| 254 | while ( $token = $p->get_token ) {
|
---|
| 255 | if ( $token->is_end_tag( 'form' ) ) { ... }
|
---|
| 256 | }
|
---|
| 257 |
|
---|
| 258 | Or:
|
---|
| 259 |
|
---|
| 260 | while ( $token = $p->get_token ) {
|
---|
| 261 | if ( $token->is_end_tag( '/form' ) ) { ... }
|
---|
| 262 | }
|
---|
| 263 |
|
---|
| 264 | Optionally, you may pass a regular expression as an argument.
|
---|
| 265 |
|
---|
| 266 | =item * C<is_text()>
|
---|
| 267 |
|
---|
| 268 | Use this to determine if you have text. Note that this is I<not> to be
|
---|
| 269 | confused with the C<return_text> (I<deprecated>) method described below!
|
---|
| 270 | C<is_text> will identify text that the user typically sees display in the Web
|
---|
| 271 | browser.
|
---|
| 272 |
|
---|
| 273 | =item * C<is_comment()>
|
---|
| 274 |
|
---|
| 275 | Are you still reading this? Nobody reads POD. Don't you know you're supposed
|
---|
| 276 | to go to CLPM, ask a question that's answered in the POD and get flamed? It's
|
---|
| 277 | a rite of passage.
|
---|
| 278 |
|
---|
| 279 | Really.
|
---|
| 280 |
|
---|
| 281 | C<is_comment> is used to identify comments. See the HTML::Parser documentation
|
---|
| 282 | for more information about comments. There's more than you might think.
|
---|
| 283 |
|
---|
| 284 | =item * C<is_declaration()>
|
---|
| 285 |
|
---|
| 286 | This will match the DTD at the top of your HTML. (You I<do> use DTD's, don't
|
---|
| 287 | you?)
|
---|
| 288 |
|
---|
| 289 | =item * C<is_process_instruction()>
|
---|
| 290 |
|
---|
| 291 | Process Instructions are from XML. This is very handy if you need to parse out
|
---|
| 292 | PHP and similar things with a parser.
|
---|
| 293 |
|
---|
| 294 | Currently, there appear to be some problems with process instructions. You can
|
---|
| 295 | override C<HTML::TokeParser::Simple::Token::ProcessInstruction> if you need to.
|
---|
| 296 |
|
---|
| 297 | =item * C<is_pi()>
|
---|
| 298 |
|
---|
| 299 | This is a shorthand for C<is_process_instruction()>.
|
---|
| 300 |
|
---|
| 301 | =back
|
---|
| 302 |
|
---|
| 303 | =head2 Data Accessors
|
---|
| 304 |
|
---|
| 305 | Some of these were originally C<return_> methods, but that name was not only
|
---|
| 306 | unwieldy, but also went against reasonable conventions. The C<get_> methods
|
---|
| 307 | listed below still have C<return_> methods available for backwards
|
---|
| 308 | compatibility reasons, but they merely call their C<get_> counterpart. For
|
---|
| 309 | example, calling C<return_tag()> actually calls C<get_tag()> internally.
|
---|
| 310 |
|
---|
| 311 | =over 4
|
---|
| 312 |
|
---|
| 313 | =item * C<get_tag()>
|
---|
| 314 |
|
---|
| 315 | Do you have a start tag or end tag? This will return the type (lower case).
|
---|
| 316 | Note that this is I<not> the same as the C<get_tag()> method on the actual
|
---|
| 317 | parser object.
|
---|
| 318 |
|
---|
| 319 | =item * C<get_attr([$attribute])>
|
---|
| 320 |
|
---|
| 321 | If you have a start tag, this will return a hash ref with the attribute names
|
---|
| 322 | as keys and the values as the values.
|
---|
| 323 |
|
---|
| 324 | If you pass in an attribute name, it will return the value for just that
|
---|
| 325 | attribute.
|
---|
| 326 |
|
---|
| 327 | Returns false if the token is not a start tag.
|
---|
| 328 |
|
---|
| 329 | =item * C<get_attrseq()>
|
---|
| 330 |
|
---|
| 331 | For a start tag, this is an array reference with the sequence of the
|
---|
| 332 | attributes, if any.
|
---|
| 333 |
|
---|
| 334 | Returns false if the token is not a start tag.
|
---|
| 335 |
|
---|
| 336 | =item * C<return_text()>
|
---|
| 337 |
|
---|
| 338 | This method has been heavily deprecated (for a couple of years) in favor of
|
---|
| 339 | C<as_is>. Programmers were getting confused over the difference between
|
---|
| 340 | C<is_text>, C<return_text>, and some parser methods such as
|
---|
| 341 | C<HTML::TokeParser::get_text> and friends.
|
---|
| 342 |
|
---|
| 343 | Using this method still succeeds, but will now carp and B<will be removed>
|
---|
| 344 | in the next major release of this module.
|
---|
| 345 |
|
---|
| 346 | =item * C<as_is()>
|
---|
| 347 |
|
---|
| 348 | This is the exact text of whatever the token is representing.
|
---|
| 349 |
|
---|
| 350 | =item * C<get_token0()>
|
---|
| 351 |
|
---|
| 352 | For processing instructions, this will return the token found immediately after
|
---|
| 353 | the opening tag. Example: For <?php, "php" will be the start of the returned
|
---|
| 354 | string.
|
---|
| 355 |
|
---|
| 356 | Note that process instruction handling appears to be incomplete in
|
---|
| 357 | C<HTML::TokeParser>.
|
---|
| 358 |
|
---|
| 359 | Returns false if the token is not a process instruction.
|
---|
| 360 |
|
---|
| 361 | =back
|
---|
| 362 |
|
---|
| 363 | =head1 MUTATORS
|
---|
| 364 |
|
---|
| 365 | The C<delete_attr()> and C<set_attr()> methods allow the programmer to rewrite
|
---|
| 366 | start tag attributes on the fly. It should be noted that bad HTML will be
|
---|
| 367 | "corrected" by this. Specifically, the new tag will have all attributes
|
---|
| 368 | lower-cased with the values properly quoted.
|
---|
| 369 |
|
---|
| 370 | Self-closing tags (e.g. E<lt>hr /E<gt>) are also handled correctly. Some older
|
---|
| 371 | browsers require a space prior to the final slash in a self-closed tag. If
|
---|
| 372 | such a space is detected in the original HTML, it will be preserved.
|
---|
| 373 |
|
---|
| 374 | Calling a mutator on an token type that does not support that property is a
|
---|
| 375 | no-op. For example:
|
---|
| 376 |
|
---|
| 377 | if ($token->is_comment) {
|
---|
| 378 | $token->set_attr(foo => 'bar'); # does nothing
|
---|
| 379 | }
|
---|
| 380 |
|
---|
| 381 | =over 4
|
---|
| 382 |
|
---|
| 383 | =item * C<delete_attr($name)>
|
---|
| 384 |
|
---|
| 385 | This method attempts to delete the attribute specified. It will silently fail
|
---|
| 386 | if called on anything other than a start tag. The argument is
|
---|
| 387 | case-insensitive, but must otherwise be an exact match of the attribute you are
|
---|
| 388 | attempting to delete. If the attribute is not found, the method will return
|
---|
| 389 | without changing the tag.
|
---|
| 390 |
|
---|
| 391 | # <body bgcolor="#FFFFFF">
|
---|
| 392 | $token->delete_attr('bgcolor');
|
---|
| 393 | print $token->as_is;
|
---|
| 394 | # <body>
|
---|
| 395 |
|
---|
| 396 | After this method is called, if successful, the C<as_is()>, C<get_attr()>
|
---|
| 397 | and C<get_attrseq()> methods will all return updated results.
|
---|
| 398 |
|
---|
| 399 | =item * C<set_attr($name,$value)>
|
---|
| 400 |
|
---|
| 401 | This method will set the value of an attribute. If the attribute is not found,
|
---|
| 402 | then C<get_attrseq()> will have the new attribute listed at the end.
|
---|
| 403 |
|
---|
| 404 | # <p>
|
---|
| 405 | $token->set_attr(class => 'some_class');
|
---|
| 406 | print $token->as_is;
|
---|
| 407 | # <p class="some_class">
|
---|
| 408 |
|
---|
| 409 | # <body bgcolor="#FFFFFF">
|
---|
| 410 | $token->set_attr('bgcolor','red');
|
---|
| 411 | print $token->as_is;
|
---|
| 412 | # <body bgcolor="red">
|
---|
| 413 |
|
---|
| 414 | After this method is called, if successful, the C<as_is()>, C<get_attr()>
|
---|
| 415 | and C<get_attrseq()> methods will all return updated results.
|
---|
| 416 |
|
---|
| 417 | =item * C<set_attr($hashref)>
|
---|
| 418 |
|
---|
| 419 | Under the premise that C<set_> methods should accept what their corresponding
|
---|
| 420 | C<get_> methods emit, the following works:
|
---|
| 421 |
|
---|
| 422 | $tag->set_attr($tag->get_attr);
|
---|
| 423 |
|
---|
| 424 | Theoretically that's a no-op and for purposes of rendering HTML, it should be.
|
---|
| 425 | However, internally this calls C<$tag-E<gt>rewrite_tag>, so see that method to
|
---|
| 426 | understand how this may affect you.
|
---|
| 427 |
|
---|
| 428 | Of course, this is useless if you want to actually change the attributes, so you
|
---|
| 429 | can do this:
|
---|
| 430 |
|
---|
| 431 | my $attrs = {
|
---|
| 432 | class => 'headline',
|
---|
| 433 | valign => 'top'
|
---|
| 434 | };
|
---|
| 435 | $token->set_attr($attrs)
|
---|
| 436 | if $token->is_start_tag('td') && $token->get_attr('class') eq 'stories';
|
---|
| 437 |
|
---|
| 438 | =item * C<rewrite_tag()>
|
---|
| 439 |
|
---|
| 440 | This method rewrites the tag. The tag name and the name of all attributes will
|
---|
| 441 | be lower-cased. Values that are not quoted with double quotes will be. This
|
---|
| 442 | may be called on both start or end tags. Note that both C<set_attr()> and
|
---|
| 443 | C<delete_attr()> call this method prior to returning.
|
---|
| 444 |
|
---|
| 445 | If called on a token that is not a tag, it simply returns. Regardless of how
|
---|
| 446 | it is called, it returns the token.
|
---|
| 447 |
|
---|
| 448 | # <body alink=#0000ff BGCOLOR=#ffffff class='none'>
|
---|
| 449 | $token->rewrite_tag;
|
---|
| 450 | print $token->as_is;
|
---|
| 451 | # <body alink="#0000ff" bgcolor="#ffffff" class="none">
|
---|
| 452 |
|
---|
| 453 | A quick cleanup of sloppy HTML is now the following:
|
---|
| 454 |
|
---|
| 455 | my $parser = HTML::TokeParser::Simple->new( string => $ugly_html );
|
---|
| 456 | while (my $token = $parser->get_token) {
|
---|
| 457 | $token->rewrite_tag;
|
---|
| 458 | print $token->as_is;
|
---|
| 459 | }
|
---|
| 460 |
|
---|
| 461 | =back
|
---|
| 462 |
|
---|
| 463 | =head1 PARSER VERSUS TOKENS
|
---|
| 464 |
|
---|
| 465 | The parser returns tokens that are blessed into appropriate classes. Some
|
---|
| 466 | people get confused and try to call parser methods on tokens and token methods
|
---|
| 467 | on the parser. To prevent this, C<HTML::TokeParser::Simple> versions 1.4 and
|
---|
| 468 | above now bless all tokens into appropriate token classes. Please keep this in
|
---|
| 469 | mind while using this module (and many thanks to PodMaster
|
---|
| 470 | L<http://www.perlmonks.org/index.pl?node_id=107642> for pointing out this issue
|
---|
| 471 | to me.)
|
---|
| 472 |
|
---|
| 473 | =head1 EXAMPLES
|
---|
| 474 |
|
---|
| 475 | =head2 Finding comments
|
---|
| 476 |
|
---|
| 477 | For some strange reason, your Pointy-Haired Boss (PHB) is convinced that the
|
---|
| 478 | graphics department is making fun of him by embedding rude things about him in
|
---|
| 479 | HTML comments. You need to get all HTML comments from the HTML.
|
---|
| 480 |
|
---|
| 481 | use strict;
|
---|
| 482 | use HTML::TokeParser::Simple;
|
---|
| 483 |
|
---|
| 484 | my @html_docs = glob( "*.html" );
|
---|
| 485 |
|
---|
| 486 | open PHB, "> phbreport.txt" or die "Cannot open phbreport for writing: $!";
|
---|
| 487 |
|
---|
| 488 | foreach my $doc ( @html_docs ) {
|
---|
| 489 | print "Processing $doc\n";
|
---|
| 490 | my $p = HTML::TokeParser::Simple->new( file => $doc );
|
---|
| 491 | while ( my $token = $p->get_token ) {
|
---|
| 492 | next unless $token->is_comment;
|
---|
| 493 | print PHB $token->as_is, "\n";
|
---|
| 494 | }
|
---|
| 495 | }
|
---|
| 496 |
|
---|
| 497 | close PHB;
|
---|
| 498 |
|
---|
| 499 | =head2 Stripping Comments
|
---|
| 500 |
|
---|
| 501 | Uh oh. Turns out that your PHB was right for a change. Many of the comments
|
---|
| 502 | in the HTML weren't very polite. Since your entire graphics department was
|
---|
| 503 | just fired, it falls on you need to strip those comments from the HTML.
|
---|
| 504 |
|
---|
| 505 | use strict;
|
---|
| 506 | use HTML::TokeParser::Simple;
|
---|
| 507 |
|
---|
| 508 | my $new_folder = 'no_comment/';
|
---|
| 509 | my @html_docs = glob( "*.html" );
|
---|
| 510 |
|
---|
| 511 | foreach my $doc ( @html_docs ) {
|
---|
| 512 | print "Processing $doc\n";
|
---|
| 513 | my $new_file = "$new_folder$doc";
|
---|
| 514 |
|
---|
| 515 | open PHB, "> $new_file" or die "Cannot open $new_file for writing: $!";
|
---|
| 516 |
|
---|
| 517 | my $p = HTML::TokeParser::Simple->new( $file => doc );
|
---|
| 518 | while ( my $token = $p->get_token ) {
|
---|
| 519 | next if $token->is_comment;
|
---|
| 520 | print PHB $token->as_is;
|
---|
| 521 | }
|
---|
| 522 | close PHB;
|
---|
| 523 | }
|
---|
| 524 |
|
---|
| 525 | =head2 Changing form tags
|
---|
| 526 |
|
---|
| 527 | Your company was foo.com and now is bar.com. Unfortunately, whoever wrote your
|
---|
| 528 | HTML decided to hardcode "http://www.foo.com/" into the C<action> attribute of
|
---|
| 529 | the form tags. You need to change it to "http://www.bar.com/".
|
---|
| 530 |
|
---|
| 531 | use strict;
|
---|
| 532 | use HTML::TokeParser::Simple;
|
---|
| 533 |
|
---|
| 534 | my $new_folder = 'new_html/';
|
---|
| 535 | my @html_docs = glob( "*.html" );
|
---|
| 536 |
|
---|
| 537 | foreach my $doc ( @html_docs ) {
|
---|
| 538 | print "Processing $doc\n";
|
---|
| 539 | my $new_file = "$new_folder$doc";
|
---|
| 540 |
|
---|
| 541 | open FILE, "> $new_file" or die "Cannot open $new_file for writing: $!";
|
---|
| 542 |
|
---|
| 543 | my $p = HTML::TokeParser::Simple->new( file => $doc );
|
---|
| 544 | while ( my $token = $p->get_token ) {
|
---|
| 545 | if ( $token->is_start_tag('form') ) {
|
---|
| 546 | my $action = $token->get_attr(action);
|
---|
| 547 | $action =~ s/www\.foo\.com/www.bar.com/;
|
---|
| 548 | $token->set_attr('action', $action);
|
---|
| 549 | }
|
---|
| 550 | print FILE $token->as_is;
|
---|
| 551 | }
|
---|
| 552 | close FILE;
|
---|
| 553 | }
|
---|
| 554 |
|
---|
| 555 | =head1 CAVEATS
|
---|
| 556 |
|
---|
| 557 | For compatability reasons with C<HTML::TokeParser>, methods that return
|
---|
| 558 | references are violating encapsulation and altering the references directly
|
---|
| 559 | B<will> alter the state of the object. Subsequent calls to C<rewrite_tag()>
|
---|
| 560 | can thus have unexpected results. Do not alter these references directly
|
---|
| 561 | unless you are following behavior described in these docs. In the future,
|
---|
| 562 | certain methods such as C<get_attr>, C<get_attrseq> and others may return a
|
---|
| 563 | copy of the reference rather than the original reference. This behavior has
|
---|
| 564 | not yet been changed in order to maintain compatability with previous versions
|
---|
| 565 | of this module. At the present time, your author is not aware of anyone taking
|
---|
| 566 | advantage of this "feature," but it's better to be safe than sorry.
|
---|
| 567 |
|
---|
| 568 | Use of C<$HTML::Parser::VERSION> which is less than 3.25 may result in
|
---|
| 569 | incorrect behavior as older versions do not always handle XHTML correctly. It
|
---|
| 570 | is the programmer's responsibility to verify that the behavior of this code
|
---|
| 571 | matches the programmer's needs.
|
---|
| 572 |
|
---|
| 573 | Note that C<HTML::Parser> processes text in 512 byte chunks. This sometimes
|
---|
| 574 | will cause strange behavior and cause text to be broken into more than one
|
---|
| 575 | token. You can suppress this behavior with the following command:
|
---|
| 576 |
|
---|
| 577 | $p->unbroken_text( [$bool] );
|
---|
| 578 |
|
---|
| 579 | See the C<HTML::Parser> documentation and
|
---|
| 580 | http://www.perlmonks.org/index.pl?node_id=230667 for more information.
|
---|
| 581 |
|
---|
| 582 | =head1 BUGS
|
---|
| 583 |
|
---|
| 584 | There are no known bugs, but that's no guarantee.
|
---|
| 585 |
|
---|
| 586 | Address bug reports and comments to: E<lt>[email protected]<gt>. When
|
---|
| 587 | sending bug reports, please provide the version of C<HTML::Parser>,
|
---|
| 588 | C<HTML::TokeParser>, C<HTML::TokeParser::Simple>, the version of Perl, and the
|
---|
| 589 | version of the operating system you are using.
|
---|
| 590 |
|
---|
| 591 | Reverse the name to email the author.
|
---|
| 592 |
|
---|
| 593 | =head1 SUBCLASSING
|
---|
| 594 |
|
---|
| 595 | You may wish to change the behavior of this module. You probably do not want
|
---|
| 596 | to subclass C<HTML::TokeParser::Simple>. Instead, you'll want to subclass one
|
---|
| 597 | of the token classes. C<HTML::TokeParser::Simple::Token> is the base class for
|
---|
| 598 | all tokens. Global behavioral changes should go there. Otherwise, see the
|
---|
| 599 | appropriate token class for the behavior you wish to alter.
|
---|
| 600 |
|
---|
| 601 | =head1 SEE ALSO
|
---|
| 602 |
|
---|
| 603 | L<HTML::TokeParser::Simple::Token>
|
---|
| 604 |
|
---|
| 605 | L<HTML::TokeParser::Simple::Token::Tag>
|
---|
| 606 |
|
---|
| 607 | L<HTML::TokeParser::Simple::Token::Text>
|
---|
| 608 |
|
---|
| 609 | L<HTML::TokeParser::Simple::Token::Comment>
|
---|
| 610 |
|
---|
| 611 | L<HTML::TokeParser::Simple::Token::Declaration>
|
---|
| 612 |
|
---|
| 613 | L<HTML::TokeParser::Simple::Token::ProcessInstruction>
|
---|
| 614 |
|
---|
| 615 | =head1 COPYRIGHT
|
---|
| 616 |
|
---|
| 617 | Copyright (c) 2004 by Curtis "Ovid" Poe. All rights reserved. This program is
|
---|
| 618 | free software; you may redistribute it and/or modify it under the same terms as
|
---|
| 619 | Perl itself
|
---|
| 620 |
|
---|
| 621 | =head1 AUTHOR
|
---|
| 622 |
|
---|
| 623 | Curtis "Ovid" Poe E<lt>[email protected]<gt>
|
---|
| 624 |
|
---|
| 625 | Reverse the name to email the author.
|
---|
| 626 |
|
---|
| 627 | =cut
|
---|