package HTML::Lint::Error; use warnings; use strict; use base 'Exporter'; our @EXPORT = (); our @EXPORT_OK = qw( STRUCTURE HELPER FLUFF ); our %EXPORT_TAGS = ( types => [@EXPORT_OK] ); our %errors; =head1 NAME HTML::Lint::Error - Error object for the Lint functionality =head1 SYNOPSIS See L for all the gory details. =head1 EXPORTS None. It's all object-based. =head1 METHODS Almost everything is an accessor. =head1 Error types: C, C, C Each error has a type. Note that these roughly, but not exactly, go from most severe to least severe. =over 4 =item * C For problems that relate to the structural validity of the code. Examples: Unclosed tags, incorrect values for attributes, and repeated attributes. =item * C Helpers are notes that will help you with your HTML, or that will help the browser render the code better or faster. Example: Missing HEIGHT and WIDTH attributes in an IMG tag. =item * C Fluff is for items that don't hurt your page, but don't help it either. This is usually something like an unknown attribute on a tag. =back =cut use constant STRUCTURE => 1; use constant HELPER => 2; use constant FLUFF => 3; =head2 new() Create an object. It's not very exciting. =cut sub new { my $class = shift; my $file = shift; my $line = shift; my $column = shift; my $errcode = shift; my @errparms = @_; # Add an element that says what tag caused the error (B, TR, etc) # so that we can match 'em up down the road. my $self = { _file => $file, _line => $line, _column => $column, _errcode => $errcode, _errtext => undef, _type => undef, }; bless $self, $class; $self->_expand_error( $errcode, @errparms ); return $self; } sub _expand_error { my $self = shift; my $errcode = shift; my $specs = $errors{$errcode}; my $str; if ( $specs ) { ($str, $self->{_type}) = @{$specs}; } else { $str = "Unknown code: $errcode"; } while ( @_ ) { my $var = shift; my $val = shift; $str =~ s/\$\{$var\}/$val/g; } $self->{_errtext} = $str; } =head2 is_type( $type1 [, $type2 ] ) Tells if any of I<$type1>, I<$type2>... match the error's type. Returns the type that matched. if ( $err->is_type( HTML::Lint::Error::STRUCTURE ) ) {.... =cut sub is_type { my $self = shift; for my $matcher ( @_ ) { return $matcher if $matcher eq $self->type; } return; } =head2 where() Returns a formatted string that describes where in the file the error has occurred. For example, (14:23) for line 14, column 23. The terrible thing about this function is that it's both a plain ol' formatting function as in my $str = where( 14, 23 ); AND it's an object method, as in: my $str = $error->where(); I don't know what I was thinking when I set it up this way, but it's bad practice. =cut sub where { my $line; my $col; if ( not ref $_[0] ) { $line = shift; $col = shift; } else { my $self = shift; $line = $self->line; $col = $self->column; } $col ||= 0; return sprintf( '(%s:%s)', $line, $col + 1 ); } =head2 as_string() Returns a nicely-formatted string for printing out to stdout or some similar user thing. =cut sub as_string { my $self = shift; return sprintf( '%s %s %s', $self->file, $self->where, $self->errtext ); } =head2 file() Returns the filename of the error, as set by the caller. =head2 line() Returns the line number of the error. =head2 column() Returns the column number, starting from 0 =head2 errcode() Returns the HTML::Lint error code. Don't rely on this, because it will probably go away. =head2 errtext() Descriptive text of the error =head2 type() Type of the error =cut sub file { my $self = shift; return $self->{_file} || '' } sub line { my $self = shift; return $self->{_line} || '' } sub column { my $self = shift; return $self->{_column} || '' } sub errcode { my $self = shift; return $self->{_errcode} || '' } sub errtext { my $self = shift; return $self->{_errtext} || '' } sub type { my $self = shift; return $self->{_type} || '' } =head1 TODO None, other than incorporating more errors, as driven by HTML::Lint. =head1 LICENSE This code may be distributed under the same terms as Perl itself. Please note that these modules are not products of or supported by the employers of the various contributors to the code. =head1 AUTHOR Andy Lester, C =cut # Errors that are commented out have not yet been implemented. # Generic element stuff %errors = ( 'elem-unknown' => ['Unknown element <${tag}>', STRUCTURE], 'elem-unopened' => [' with no opening <${tag}>', STRUCTURE], 'elem-unclosed' => ['<${tag}> at ${where} is never closed', STRUCTURE], 'elem-empty-but-closed' => ['<${tag}> is not a container -- is not allowed', STRUCTURE], 'elem-img-sizes-missing' => [' tag has no HEIGHT and WIDTH attributes.', HELPER], 'elem-img-alt-missing' => [' does not have ALT text defined', HELPER], 'elem-nonrepeatable' => ['<${tag}> is not repeatable, but already appeared at ${where}', STRUCTURE], 'doc-tag-required' => ['<${tag}> tag is required', STRUCTURE], 'attr-repeated' => ['${attr} attribute in <${tag}> is repeated', STRUCTURE], 'attr-unknown' => ['Unknown attribute "${attr}" for tag <${tag}>', FLUFF], 'text-use-entity' => ['Invalid character ${char} should be written as ${entity}', STRUCTURE], ); 1; # happy __DATA__ Errors that haven't been done yet. #elem-head-only <${tag}> can only appear in the element #elem-non-head-element <${tag}> cannot appear in the element #elem-obsolete <${tag}> is obsolete #elem-nested-element <${tag}> cannot be nested -- one is already opened at ${where} #elem-wrong-context Illegal context for <${tag}> -- must appear in <${othertag}> tag. #elem-heading-in-anchor should be inside <${tag}>, not <${tag}> inside #elem-head-missing No element found #elem-head-missing-title No in <HEAD> element #elem-img-sizes-incorrect <IMG> tag's HEIGHT and WIDTH attributes are incorrect. They should be ${correct}. #attr-missing <${tag}> is missing a "${attr}" attribute #comment-unclosed Unclosed comment #comment-markup Markup embedded in a comment can confuse some browsers #text-literal-metacharacter Metacharacter $char should be represented as "$otherchar" #text-title-length The HTML spec recommends that that <TITLE> be no more than 64 characters #text-markup Tag <${tag}> found in the <TITLE>, which will not be rendered properly. #elem-physical-markup <${tag}> is physical font markup. Use logical (such as <${othertag}>) instead. #elem-leading-whitespace <${tag}> should not have whitespace between "<" and "${tag}>" #'must-follow' => [ ENABLED, MC_ERROR, '<$argv[0]> must immediately follow <$argv[1]>', ], # 'empty-container' => [ ENABLED, MC_WARNING, 'empty container element <$argv[0]>.', ], # 'directory-index' => [ ENABLED, MC_WARNING, 'directory $argv[0] does not have an index file ($argv[1])', ], # 'attribute-delimiter' => [ ENABLED, MC_WARNING, 'use of \' for attribute value delimiter is not supported by all browsers (attribute $argv[0] of tag $argv[1])', ], # 'container-whitespace' => [ DISABLED, MC_WARNING, '$argv[0] whitespace in content of container element $argv[1]', ], # 'bad-text-context' => [ ENABLED, MC_ERROR, 'illegal context, <$argv[0]>, for text; should be in $argv[1].', ], # 'attribute-format' => [ ENABLED, MC_ERROR, 'illegal value for $argv[0] attribute of $argv[1] ($argv[2])', ], # 'quote-attribute-value' => [ ENABLED, MC_ERROR, 'value for attribute $argv[0] ($argv[1]) of element $argv[2] should be quoted (i.e. $argv[0]="$argv[1]")', ], # 'meta-in-pre' => [ ENABLED, MC_ERROR, 'you should use "$argv[0]" in place of "$argv[1]", even in a PRE element.', ], # 'implied-element' => [ ENABLED, MC_WARNING, 'saw <$argv[0]> element, but no <$argv[1]> element', ], # 'button-usemap' => [ ENABLED, MC_ERROR, 'illegal to associate an image map with IMG inside a BUTTON', ],