mirrors/htmlpurifier
0
0
Fork 0
mirror of https://github.com/ezyang/htmlpurifier.git synced 2024-12-23 00:41:52 +00:00
Code Releases Activity

Rename MarkupFragment.php to Token.php, change internal class names and rewire the classes. We also started adding more dependence on the Lexer and Generator in unrelated tests.

Browse Source 
git-svn-id: http://htmlpurifier.org/svnroot/htmlpurifier/trunk@63 48356398-32a2-884e-a903-53898d9a118a
This commit is contained in:
Edward Z. Yang 2006-07-21 11:27:54 +00:00
parent 8bde230c99
commit 23dba8b55e
9 changed files with 323 additions and 372 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
HTML_Generator.php
View File

@ -12,18 +12,18 @@ class HTML_Generator
} }
function generateFromToken($token) { function generateFromToken($token) {
if (is_a($token, 'MF_StartTag')) { if ($token->type == 'start') {
$attr = $this->generateAttributes($token->attributes); $attr = $this->generateAttributes($token->attributes);
return '<' . $token->name . ($attr ? ' ' : '') . $attr . '>'; return '<' . $token->name . ($attr ? ' ' : '') . $attr . '>';
} elseif (is_a($token, 'MF_EndTag')) { } elseif ($token->type == 'end') {
return '</' . $token->name . '>'; return '</' . $token->name . '>';
} elseif (is_a($token, 'MF_EmptyTag')) { } elseif ($token->type == 'empty') {
$attr = $this->generateAttributes($token->attributes); $attr = $this->generateAttributes($token->attributes);
return '<' . $token->name . ($attr ? ' ' : '') . $attr . ' />'; return '<' . $token->name . ($attr ? ' ' : '') . $attr . ' />';
} elseif (is_a($token, 'MF_Text')) { } elseif ($token->type == 'text') {
return htmlentities($token->data, ENT_COMPAT, 'UTF-8'); return htmlentities($token->data, ENT_COMPAT, 'UTF-8');
} else { } else {

30
HTML_Lexer.php
View File

@ -121,7 +121,7 @@ class HTML_Lexer
if (!$inside_tag && $position_next_lt !== false) { if (!$inside_tag && $position_next_lt !== false) {
// We are not inside tag and there still is another tag to parse // We are not inside tag and there still is another tag to parse
$array[] = new $array[] = new
MF_Text( HTMLPurifier_Token_Text(
html_entity_decode( html_entity_decode(
substr( substr(
$string, $cursor, $position_next_lt - $cursor $string, $cursor, $position_next_lt - $cursor
@ -138,7 +138,7 @@ class HTML_Lexer
if ($cursor === strlen($string)) break; if ($cursor === strlen($string)) break;
// Create Text of rest of string // Create Text of rest of string
$array[] = new $array[] = new
MF_Text( HTMLPurifier_Token_Text(
html_entity_decode( html_entity_decode(
substr( substr(
$string, $cursor $string, $cursor
@ -158,7 +158,7 @@ class HTML_Lexer
substr($segment,strlen($segment)-2,2) == '--' substr($segment,strlen($segment)-2,2) == '--'
) { ) {
$array[] = new $array[] = new
MF_Comment( HTMLPurifier_Token_Comment(
substr( substr(
$segment, 3, strlen($segment) - 5 $segment, 3, strlen($segment) - 5
) )
@ -172,7 +172,7 @@ class HTML_Lexer
$is_end_tag = (strpos($segment,'/') === 0); $is_end_tag = (strpos($segment,'/') === 0);
if ($is_end_tag) { if ($is_end_tag) {
$type = substr($segment, 1); $type = substr($segment, 1);
$array[] = new MF_EndTag($type); $array[] = new HTMLPurifier_Token_End($type);
$inside_tag = false; $inside_tag = false;
$cursor = $position_next_gt + 1; $cursor = $position_next_gt + 1;
continue; continue;
@ -191,9 +191,9 @@ class HTML_Lexer
$position_first_space = $this->nextWhiteSpace($segment); $position_first_space = $this->nextWhiteSpace($segment);
if ($position_first_space === false) { if ($position_first_space === false) {
if ($is_self_closing) { if ($is_self_closing) {
$array[] = new MF_EmptyTag($segment); $array[] = new HTMLPurifier_Token_Empty($segment);
} else { } else {
$array[] = new MF_StartTag($segment, array()); $array[] = new HTMLPurifier_Token_Start($segment);
} }
$inside_tag = false; $inside_tag = false;
$cursor = $position_next_gt + 1; $cursor = $position_next_gt + 1;
@ -210,16 +210,16 @@ class HTML_Lexer
); );
$attributes = $this->tokenizeAttributeString($attribute_string); $attributes = $this->tokenizeAttributeString($attribute_string);
if ($is_self_closing) { if ($is_self_closing) {
$array[] = new MF_EmptyTag($type, $attributes); $array[] = new HTMLPurifier_Token_Empty($type, $attributes);
} else { } else {
$array[] = new MF_StartTag($type, $attributes); $array[] = new HTMLPurifier_Token_Start($type, $attributes);
} }
$cursor = $position_next_gt + 1; $cursor = $position_next_gt + 1;
$inside_tag = false; $inside_tag = false;
continue; continue;
} else { } else {
$array[] = new $array[] = new
MF_Text( HTMLPurifier_Token_Text(
'<' . '<' .
html_entity_decode( html_entity_decode(
substr($string, $cursor), substr($string, $cursor),
@ -362,9 +362,9 @@ class HTML_Lexer_Sax extends HTML_Lexer
function openHandler(&$parser, $name, $attrs, $closed) { function openHandler(&$parser, $name, $attrs, $closed) {
if ($closed) { if ($closed) {
$this->tokens[] = new MF_EmptyTag($name, $attrs); $this->tokens[] = new HTMLPurifier_Token_Empty($name, $attrs);
} else { } else {
$this->tokens[] = new MF_StartTag($name, $attrs); $this->tokens[] = new HTMLPurifier_Token_Start($name, $attrs);
} }
return true; return true;
} }
@ -373,21 +373,21 @@ class HTML_Lexer_Sax extends HTML_Lexer
// HTMLSax3 seems to always send empty tags an extra close tag // HTMLSax3 seems to always send empty tags an extra close tag
// check and ignore if you see it: // check and ignore if you see it:
// [TESTME] to make sure it doesn't overreach // [TESTME] to make sure it doesn't overreach
if (is_a($this->tokens[count($this->tokens)-1], 'MF_EmptyTag')) { if ($this->tokens[count($this->tokens)-1]->type == 'empty') {
return true; return true;
} }
$this->tokens[] = new MF_EndTag($name); $this->tokens[] = new HTMLPurifier_Token_End($name);
return true; return true;
} }
function dataHandler(&$parser, $data) { function dataHandler(&$parser, $data) {
$this->tokens[] = new MF_Text($data); $this->tokens[] = new HTMLPurifier_Token_Text($data);
return true; return true;
} }
function escapeHandler(&$parser, $data) { function escapeHandler(&$parser, $data) {
if (strpos($data, '-') === 0) { if (strpos($data, '-') === 0) {
$this->tokens[] = new MF_Comment($data); $this->tokens[] = new HTMLPurifier_Token_Comment($data);
} }
return true; return true;
} }

65
MarkupFragment.php
View File

@ -1,65 +0,0 @@
<?php
// MF = Markup Fragment
// all objects here are immutable
class MF {}
class MF_Tag extends MF // abstract
{
var $name;
function MF_Tag($name) {
$this->name = strtolower($name); // for some reason, the SAX parser
// uses uppercase. Investigate?
}
}
class MF_TagWithAttributes extends MF_Tag // abstract
{
var $attributes = array();
function MF_TagWithAttributes($name, $attributes = array()) {
$this->MF_Tag($name);
$this->attributes = $attributes;
}
}
class MF_StartTag extends MF_TagWithAttributes
{
var $type = 'start';
}
class MF_EmptyTag extends MF_TagWithAttributes
{
var $type = 'empty';
}
class MF_EndTag extends MF_Tag
{
var $type = 'end';
}
class MF_Text extends MF
{
var $name = '#PCDATA';
var $type = 'text';
var $data;
var $is_whitespace = false;
function MF_Text($data) {
$this->data = $data;
if (trim($data, " \n\r\t") === '') $this->is_whitespace = true;
}
function append($mf_text) {
return new MF_Text($this->data . $mf_text->data);
}
}
class MF_Comment extends MF
{
var $data;
var $type = 'comment';
function MF_Comment($data) {
$this->data = $data;
}
}
?>

38
PureHTMLDefinition.php
View File

@ -162,10 +162,10 @@ class PureHTMLDefinition
if (empty($this->info)) $this->loadData(); if (empty($this->info)) $this->loadData();
$result = array(); $result = array();
foreach($tokens as $token) { foreach($tokens as $token) {
if (is_subclass_of($token, 'MF_Tag')) { if (!empty( $token->is_tag )) {
if (!isset($this->info[$token->name])) { if (!isset($this->info[$token->name])) {
// invalid tag, generate HTML and insert in // invalid tag, generate HTML and insert in
$token = new MF_Text( $token = new HTMLPurifier_Token_Text(
$this->generator->generateFromToken($token) $this->generator->generateFromToken($token)
); );
} }
@ -186,7 +186,7 @@ class PureHTMLDefinition
$result = array(); $result = array();
$current_nesting = array(); $current_nesting = array();
foreach ($tokens as $token) { foreach ($tokens as $token) {
if (!is_subclass_of($token, 'MF_Tag')) { if (empty( $token->is_tag )) {
$result[] = $token; $result[] = $token;
continue; continue;
} }
@ -196,7 +196,8 @@ class PureHTMLDefinition
if ($info->child_def->type == 'empty' && if ($info->child_def->type == 'empty' &&
$token->type == 'start' ) { $token->type == 'start' ) {
$result[] = new MF_EmptyTag($token->name, $token->attributes); $result[] = new HTMLPurifier_Token_Empty($token->name,
$token->attributes);
continue; continue;
} }
@ -204,8 +205,9 @@ class PureHTMLDefinition
if ($info->child_def->type != 'empty' && if ($info->child_def->type != 'empty' &&
$token->type == 'empty' ) { $token->type == 'empty' ) {
$result[] = new MF_StartTag($token->name, $token->attributes); $result[] = new HTMLPurifier_Token_Start($token->name,
$result[] = new MF_EndTag($token->name); $token->attributes);
$result[] = new HTMLPurifier_Token_End($token->name);
continue; continue;
} }
@ -228,7 +230,7 @@ class PureHTMLDefinition
if ($current_parent->name == 'p' && if ($current_parent->name == 'p' &&
isset($this->info_closes_p[$token->name]) isset($this->info_closes_p[$token->name])
) { ) {
$result[] = new MF_EndTag('p'); $result[] = new HTMLPurifier_Token_End('p');
$result[] = $token; $result[] = $token;
$current_nesting[] = $token; $current_nesting[] = $token;
continue; continue;
@ -238,7 +240,7 @@ class PureHTMLDefinition
if ($current_parent->name == 'li' && if ($current_parent->name == 'li' &&
$token->name == 'li' $token->name == 'li'
) { ) {
$result[] = new MF_EndTag('li'); $result[] = new HTMLPurifier_Token_End('li');
$result[] = $token; $result[] = $token;
$current_nesting[] = $token; $current_nesting[] = $token;
continue; continue;
@ -263,7 +265,7 @@ class PureHTMLDefinition
// make sure that we have something open // make sure that we have something open
if (empty($current_nesting)) { if (empty($current_nesting)) {
$result[] = new MF_Text( $result[] = new HTMLPurifier_Token_Text(
$this->generator->generateFromToken($token) $this->generator->generateFromToken($token)
); );
continue; continue;
@ -298,7 +300,7 @@ class PureHTMLDefinition
// we still didn't find the tag, so translate to text // we still didn't find the tag, so translate to text
if ($skipped_tags === false) { if ($skipped_tags === false) {
$result[] = new MF_Text( $result[] = new HTMLPurifier_Token_Text(
$this->generator->generateFromToken($token) $this->generator->generateFromToken($token)
); );
continue; continue;
@ -308,7 +310,7 @@ class PureHTMLDefinition
// note that skipped tags contains the element we need closed // note that skipped tags contains the element we need closed
$size = count($skipped_tags); $size = count($skipped_tags);
for ($i = $size - 1; $i >= 0; $i--) { for ($i = $size - 1; $i >= 0; $i--) {
$result[] = new MF_EndTag($skipped_tags[$i]->name); $result[] = new HTMLPurifier_Token_End($skipped_tags[$i]->name);
} }
// done! // done!
@ -320,7 +322,8 @@ class PureHTMLDefinition
if (!empty($current_nesting)) { if (!empty($current_nesting)) {
$size = count($current_nesting); $size = count($current_nesting);
for ($i = $size - 1; $i >= 0; $i--) { for ($i = $size - 1; $i >= 0; $i--) {
$result[] = new MF_EndTag($current_nesting[$i]->name); $result[] =
new HTMLPurifier_Token_End($current_nesting[$i]->name);
} }
} }
@ -331,8 +334,8 @@ class PureHTMLDefinition
if (empty($this->info)) $this->loadData(); if (empty($this->info)) $this->loadData();
// insert implicit "parent" node, will be removed at end // insert implicit "parent" node, will be removed at end
array_unshift($tokens, new MF_StartTag('div')); array_unshift($tokens, new HTMLPurifier_Token_Start('div'));
$tokens[] = new MF_EndTag('div'); $tokens[] = new HTMLPurifier_Token_End('div');
for ($i = 0, $size = count($tokens) ; $i < $size; ) { for ($i = 0, $size = count($tokens) ; $i < $size; ) {
@ -553,7 +556,7 @@ class HTMLDTD_ChildDef_Required extends HTMLDTD_ChildDef_Simple
if (!isset($this->elements[$token->name])) { if (!isset($this->elements[$token->name])) {
$is_deleting = true; $is_deleting = true;
if ($pcdata_allowed) { if ($pcdata_allowed) {
$result[] = new MF_Text( $result[] = new HTMLPurifier_Token_Text(
$this->gen->generateFromToken($token) $this->gen->generateFromToken($token)
); );
} }
@ -563,7 +566,10 @@ class HTMLDTD_ChildDef_Required extends HTMLDTD_ChildDef_Simple
if (!$is_deleting) { if (!$is_deleting) {
$result[] = $token; $result[] = $token;
} elseif ($pcdata_allowed) { } elseif ($pcdata_allowed) {
$result[] = new MF_Text($this->gen->generateFromToken($token)); $result[] =
new HTMLPurifier_Token_Text(
$this->gen->generateFromToken( $token )
);
} else { } else {
// drop silently // drop silently
} }

66
Token.php Normal file
View File

@ -0,0 +1,66 @@
<?php
// all objects here are immutable
class HTMLPurifier_Token {} // abstract
class HTMLPurifier_Token_Tag extends HTMLPurifier_Token // abstract
{
var $is_tag = true;
var $name;
function HTMLPurifier_Token_Tag($name) {
$this->name = strtolower($name); // for some reason, the SAX parser
// uses uppercase. Investigate?
}
}
// a rich tag has attributes
class HTMLPurifier_Token_RichTag extends HTMLPurifier_Token_Tag // abstract
{
var $attributes = array();
function HTMLPurifier_Token_RichTag($name, $attributes = array()) {
$this->HTMLPurifier_Token_Tag($name);
$this->attributes = $attributes;
}
}
class HTMLPurifier_Token_Start extends HTMLPurifier_Token_RichTag
{
var $type = 'start';
}
class HTMLPurifier_Token_Empty extends HTMLPurifier_Token_RichTag
{
var $type = 'empty';
}
class HTMLPurifier_Token_End extends HTMLPurifier_Token_Tag
{
var $type = 'end';
}
class HTMLPurifier_Token_Text extends HTMLPurifier_Token
{
var $name = '#PCDATA';
var $type = 'text';
var $data;
var $is_whitespace = false;
function HTMLPurifier_Token_Text($data) {
$this->data = $data;
if (trim($data, " \n\r\t") === '') $this->is_whitespace = true;
}
function append($text) {
return new HTMLPurifier_Token_Text($this->data . $text->data);
}
}
class HTMLPurifier_Token_Comment extends HTMLPurifier_Token
{
var $data;
var $type = 'comment';
function HTMLPurifier_Token_Comment($data) {
$this->data = $data;
}
}
?>

6
tester.php
View File

@ -6,7 +6,7 @@ require_once 'XML/HTMLSax3.php'; // optional PEAR class
require_once 'HTML_Purifier.php'; require_once 'HTML_Purifier.php';
require_once 'HTML_Lexer.php'; require_once 'HTML_Lexer.php';
require_once 'MarkupFragment.php'; require_once 'Token.php';
require_once 'PureHTMLDefinition.php'; require_once 'PureHTMLDefinition.php';
require_once 'HTML_Generator.php'; require_once 'HTML_Generator.php';
@ -15,11 +15,11 @@ $test = new GroupTest('HTML_Purifier');
chdir('tests/'); chdir('tests/');
$test->addTestFile('HTML_Purifier.php'); $test->addTestFile('HTML_Purifier.php');
$test->addTestFile('HTML_Lexer.php'); $test->addTestFile('HTML_Lexer.php');
//$test->addTestFile('MarkupFragment.php'); //$test->addTestFile('Token.php');
$test->addTestFile('PureHTMLDefinition.php'); $test->addTestFile('PureHTMLDefinition.php');
$test->addTestFile('HTML_Generator.php'); $test->addTestFile('HTML_Generator.php');
chdir('../'); chdir('../');
$test->run(new HtmlReporter()); $test->run( new HtmlReporter() );
?> ?>

22
tests/HTML_Generator.php
View File

@ -15,22 +15,26 @@ class Test_HTML_Generator extends UnitTestCase
$inputs = array(); $inputs = array();
$expect = array(); $expect = array();
$inputs[0] = new MF_Text('Foobar.<>'); $inputs[0] = new HTMLPurifier_Token_Text('Foobar.<>');
$expect[0] = 'Foobar.&lt;&gt;'; $expect[0] = 'Foobar.&lt;&gt;';
$inputs[1] = new MF_StartTag('a', array('href' => 'dyn?a=foo&b=bar')); $inputs[1] = new HTMLPurifier_Token_Start('a',
array('href' => 'dyn?a=foo&b=bar')
);
$expect[1] = '<a href="dyn?a=foo&amp;b=bar">'; $expect[1] = '<a href="dyn?a=foo&amp;b=bar">';
$inputs[2] = new MF_EndTag('b'); $inputs[2] = new HTMLPurifier_Token_End('b');
$expect[2] = '</b>'; $expect[2] = '</b>';
$inputs[3] = new MF_EmptyTag('br', array('style' => 'font-family:"Courier New";')); $inputs[3] = new HTMLPurifier_Token_Empty('br',
array('style' => 'font-family:"Courier New";')
);
$expect[3] = '<br style="font-family:&quot;Courier New&quot;;" />'; $expect[3] = '<br style="font-family:&quot;Courier New&quot;;" />';
$inputs[4] = new MF_StartTag('asdf'); $inputs[4] = new HTMLPurifier_Token_Start('asdf');
$expect[4] = '<asdf>'; $expect[4] = '<asdf>';
$inputs[5] = new MF_EmptyTag('br'); $inputs[5] = new HTMLPurifier_Token_Empty('br');
$expect[5] = '<br />'; $expect[5] = '<br />';
foreach ($inputs as $i => $input) { foreach ($inputs as $i => $input) {
@ -69,9 +73,9 @@ class Test_HTML_Generator extends UnitTestCase
function test_generateFromTokens() { function test_generateFromTokens() {
$tokens = array( $tokens = array(
new MF_StartTag('b'), new HTMLPurifier_Token_Start('b'),
new MF_Text('Foobar!'), new HTMLPurifier_Token_Text('Foobar!'),
new MF_EndTag('b') new HTMLPurifier_Token_End('b')
); );
$expect = '<b>Foobar!</b>'; $expect = '<b>Foobar!</b>';
$this->assertEqual($expect, $this->gen->generateFromTokens($tokens)); $this->assertEqual($expect, $this->gen->generateFromTokens($tokens));

80
tests/HTML_Lexer.php
View File

@ -46,83 +46,83 @@ class Test_HTML_Lexer extends UnitTestCase
$input[1] = 'This is regular text.'; $input[1] = 'This is regular text.';
$expect[1] = array( $expect[1] = array(
new MF_Text('This is regular text.') new HTMLPurifier_Token_Text('This is regular text.')
); );
$input[2] = 'This is <b>bold</b> text'; $input[2] = 'This is <b>bold</b> text';
$expect[2] = array( $expect[2] = array(
new MF_Text('This is ') new HTMLPurifier_Token_Text('This is ')
,new MF_StartTag('b', array()) ,new HTMLPurifier_Token_Start('b', array())
,new MF_Text('bold') ,new HTMLPurifier_Token_Text('bold')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
,new MF_Text(' text') ,new HTMLPurifier_Token_Text(' text')
); );
$input[3] = '<DIV>Totally rad dude. <b>asdf</b></div>'; $input[3] = '<DIV>Totally rad dude. <b>asdf</b></div>';
$expect[3] = array( $expect[3] = array(
new MF_StartTag('DIV', array()) new HTMLPurifier_Token_Start('DIV', array())
,new MF_Text('Totally rad dude. ') ,new HTMLPurifier_Token_Text('Totally rad dude. ')
,new MF_StartTag('b', array()) ,new HTMLPurifier_Token_Start('b', array())
,new MF_Text('asdf') ,new HTMLPurifier_Token_Text('asdf')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
,new MF_EndTag('div') ,new HTMLPurifier_Token_End('div')
); );
$input[4] = '<asdf></asdf><d></d><poOloka><poolasdf><ds></asdf></ASDF>'; $input[4] = '<asdf></asdf><d></d><poOloka><poolasdf><ds></asdf></ASDF>';
$expect[4] = array( $expect[4] = array(
new MF_StartTag('asdf') new HTMLPurifier_Token_Start('asdf')
,new MF_EndTag('asdf') ,new HTMLPurifier_Token_End('asdf')
,new MF_StartTag('d') ,new HTMLPurifier_Token_Start('d')
,new MF_EndTag('d') ,new HTMLPurifier_Token_End('d')
,new MF_StartTag('poOloka') ,new HTMLPurifier_Token_Start('poOloka')
,new MF_StartTag('poolasdf') ,new HTMLPurifier_Token_Start('poolasdf')
,new MF_StartTag('ds') ,new HTMLPurifier_Token_Start('ds')
,new MF_EndTag('asdf') ,new HTMLPurifier_Token_End('asdf')
,new MF_EndTag('ASDF') ,new HTMLPurifier_Token_End('ASDF')
); );
$input[5] = '<a'."\t".'href="foobar.php"'."\n".'title="foo!">Link to <b id="asdf">foobar</b></a>'; $input[5] = '<a'."\t".'href="foobar.php"'."\n".'title="foo!">Link to <b id="asdf">foobar</b></a>';
$expect[5] = array( $expect[5] = array(
new MF_StartTag('a',array('href'=>'foobar.php','title'=>'foo!')) new HTMLPurifier_Token_Start('a',array('href'=>'foobar.php','title'=>'foo!'))
,new MF_Text('Link to ') ,new HTMLPurifier_Token_Text('Link to ')
,new MF_StartTag('b',array('id'=>'asdf')) ,new HTMLPurifier_Token_Start('b',array('id'=>'asdf'))
,new MF_Text('foobar') ,new HTMLPurifier_Token_Text('foobar')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
,new MF_EndTag('a') ,new HTMLPurifier_Token_End('a')
); );
$input[6] = '<br />'; $input[6] = '<br />';
$expect[6] = array( $expect[6] = array(
new MF_EmptyTag('br') new HTMLPurifier_Token_Empty('br')
); );
// [INVALID] [RECOVERABLE] // [INVALID] [RECOVERABLE]
$input[7] = '<!-- Comment --> <!-- not so well formed --->'; $input[7] = '<!-- Comment --> <!-- not so well formed --->';
$expect[7] = array( $expect[7] = array(
new MF_Comment(' Comment ') new HTMLPurifier_Token_Comment(' Comment ')
,new MF_Text(' ') ,new HTMLPurifier_Token_Text(' ')
,new MF_Comment(' not so well formed -') ,new HTMLPurifier_Token_Comment(' not so well formed -')
); );
$sax_expect[7] = false; // we need to figure out proper comment output $sax_expect[7] = false; // we need to figure out proper comment output
// [INVALID] // [INVALID]
$input[8] = '<a href=""'; $input[8] = '<a href=""';
$expect[8] = array( $expect[8] = array(
new MF_Text('<a href=""') new HTMLPurifier_Token_Text('<a href=""')
); );
// SAX parses it into a tag // SAX parses it into a tag
$sax_expect[8] = array( $sax_expect[8] = array(
new MF_StartTag('a', array('href'=>'')) new HTMLPurifier_Token_Start('a', array('href'=>''))
); );
$input[9] = '&lt;b&gt;'; $input[9] = '&lt;b&gt;';
$expect[9] = array( $expect[9] = array(
new MF_Text('<b>') new HTMLPurifier_Token_Text('<b>')
); );
$sax_expect[9] = array( $sax_expect[9] = array(
new MF_Text('<') new HTMLPurifier_Token_Text('<')
,new MF_Text('b') ,new HTMLPurifier_Token_Text('b')
,new MF_Text('>') ,new HTMLPurifier_Token_Text('>')
); );
// note that SAX can clump text nodes together. We won't be // note that SAX can clump text nodes together. We won't be
// too picky though // too picky though
@ -130,16 +130,16 @@ class Test_HTML_Lexer extends UnitTestCase
// [INVALID] // [INVALID]
$input[10] = '<a "=>'; $input[10] = '<a "=>';
$expect[10] = array( $expect[10] = array(
new MF_StartTag('a', array('"' => '')) new HTMLPurifier_Token_Start('a', array('"' => ''))
); );
// [INVALID] [RECOVERABLE] // [INVALID] [RECOVERABLE]
$input[11] = '"'; $input[11] = '"';
$expect[11] = array( new MF_Text('"') ); $expect[11] = array( new HTMLPurifier_Token_Text('"') );
// compare with this valid one: // compare with this valid one:
$input[12] = '&quot;'; $input[12] = '&quot;';
$expect[12] = array( new MF_Text('"') ); $expect[12] = array( new HTMLPurifier_Token_Text('"') );
$sax_expect[12] = false; $sax_expect[12] = false;
// SAX chokes on this? We do have entity parsing on, so it should work! // SAX chokes on this? We do have entity parsing on, so it should work!

378
tests/PureHTMLDefinition.php
View File

@ -3,14 +3,25 @@
class Test_HTMLDTD_ChildDef extends UnitTestCase class Test_HTMLDTD_ChildDef extends UnitTestCase
{ {
var $lex;
var $gen;
function Test_HTMLDTD_ChildDef() {
$this->lex = new HTML_Lexer();
$this->gen = new HTML_Generator();
parent::UnitTestCase();
}
function assertSeries($inputs, $expect, $def) { function assertSeries($inputs, $expect, $def) {
foreach ($inputs as $i => $input) { foreach ($inputs as $i => $input) {
$result = $def->validateChildren($input); $tokens = $this->lex->tokenizeHTML($input);
$result = $def->validateChildren($tokens);
if (is_bool($expect[$i])) { if (is_bool($expect[$i])) {
$this->assertIdentical($expect[$i], $result); $this->assertIdentical($expect[$i], $result);
} else { } else {
$this->assertEqual($expect[$i], $result); $result_html = $this->gen->generateFromTokens($result);
paintIf($result, $result != $expect[$i]); $this->assertEqual($expect[$i], $result_html);
paintIf($result_html, $result_html != $expect[$i]);
} }
} }
} }
@ -21,32 +32,20 @@ class Test_HTMLDTD_ChildDef extends UnitTestCase
$def = new HTMLDTD_ChildDef( $def = new HTMLDTD_ChildDef(
'(caption?, (col*|colgroup*), thead?, tfoot?, (tbody+|tr+))'); '(caption?, (col*|colgroup*), thead?, tfoot?, (tbody+|tr+))');
$inputs[0] = array(); $inputs[0] = '';
$expect[0] = false; $expect[0] = false;
// we really don't care what's inside, because if it turns out // we really don't care what's inside, because if it turns out
// this tr is illegal, we'll end up re-evaluating the parent node // this tr is illegal, we'll end up re-evaluating the parent node
// anyway. // anyway.
$inputs[1] = array( $inputs[1] = '<tr></tr>';
new MF_StartTag('tr') ,new MF_EndTag('tr')
);
$expect[1] = true; $expect[1] = true;
$inputs[2] = array( $inputs[2] = '<caption></caption><col></col><thead></thead>' .
new MF_StartTag('caption') ,new MF_EndTag('caption') '<tfoot></tfoot><tbody></tbody>';
,new MF_StartTag('col') ,new MF_EndTag('col')
,new MF_StartTag('thead') ,new MF_EndTag('thead')
,new MF_StartTag('tfoot') ,new MF_EndTag('tfoot')
,new MF_StartTag('tbody') ,new MF_EndTag('tbody')
);
$expect[2] = true; $expect[2] = true;
$inputs[3] = array( $inputs[3] = '<col></col><col></col><col></col><tr></tr>';
new MF_StartTag('col') ,new MF_EndTag('col')
,new MF_StartTag('col') ,new MF_EndTag('col')
,new MF_StartTag('col') ,new MF_EndTag('col')
,new MF_StartTag('tr') ,new MF_EndTag('tr')
);
$expect[3] = true; $expect[3] = true;
$this->assertSeries($inputs, $expect, $def); $this->assertSeries($inputs, $expect, $def);
@ -81,63 +80,22 @@ class Test_HTMLDTD_ChildDef extends UnitTestCase
$inputs[0] = array(); $inputs[0] = array();
$expect[0] = false; $expect[0] = false;
$inputs[1] = array( $inputs[1] = '<dt>Term</dt>Text in an illegal location'.
new MF_StartTag('dt') '<dd>Definition</dd><b>Illegal tag</b>';
,new MF_Text('Term')
,new MF_EndTag('dt')
,new MF_Text('Text in an illegal location') $expect[1] = '<dt>Term</dt><dd>Definition</dd>';
,new MF_StartTag('dd') $inputs[2] = 'How do you do!';
,new MF_Text('Definition')
,new MF_EndTag('dd')
,new MF_StartTag('b') // test tag removal too
,new MF_EndTag('b')
);
$expect[1] = array(
new MF_StartTag('dt')
,new MF_Text('Term')
,new MF_EndTag('dt')
,new MF_StartTag('dd')
,new MF_Text('Definition')
,new MF_EndTag('dd')
);
$inputs[2] = array(new MF_Text('How do you do!'));
$expect[2] = false; $expect[2] = false;
// whitespace shouldn't trigger it // whitespace shouldn't trigger it
$inputs[3] = array( $inputs[3] = "\n<dd>Definition</dd> ";
new MF_Text("\n")
,new MF_StartTag('dd')
,new MF_Text('Definition')
,new MF_EndTag('dd')
,new MF_Text(' ')
);
$expect[3] = true; $expect[3] = true;
$inputs[4] = array( $inputs[4] ='<dd>Definition</dd> <b></b> ';
new MF_StartTag('dd') $expect[4] = '<dd>Definition</dd> ';
,new MF_Text('Definition')
,new MF_EndTag('dd') $inputs[5] = "\t ";
,new MF_Text(' ')
,new MF_StartTag('b')
,new MF_EndTag('b')
,new MF_Text(' ')
);
$expect[4] = array(
new MF_StartTag('dd')
,new MF_Text('Definition')
,new MF_EndTag('dd')
,new MF_Text(' ')
,new MF_Text(' ')
);
$inputs[5] = array(
new MF_Text(' ')
,new MF_Text("\t")
);
$expect[5] = false; $expect[5] = false;
$this->assertSeries($inputs, $expect, $def); $this->assertSeries($inputs, $expect, $def);
@ -146,41 +104,23 @@ class Test_HTMLDTD_ChildDef extends UnitTestCase
function test_required_pcdata_allowed() { function test_required_pcdata_allowed() {
$def = new HTMLDTD_ChildDef_Required('#PCDATA | b'); $def = new HTMLDTD_ChildDef_Required('#PCDATA | b');
$input = array(
new MF_StartTag('b') $inputs[0] = '<b>Bold text</b><img />';
,new MF_Text('Bold text') $expect[0] = '<b>Bold text</b>&lt;img /&gt;';
,new MF_EndTag('b')
,new MF_EmptyTag('img') // illegal tag $this->assertSeries($inputs, $expect, $def);
);
$expect = array(
new MF_StartTag('b')
,new MF_Text('Bold text')
,new MF_EndTag('b')
,new MF_Text('<img />')
);
$this->assertEqual($expect, $def->validateChildren($input));
} }
function test_optional() { function test_optional() {
$def = new HTMLDTD_ChildDef_Optional('b | i'); $def = new HTMLDTD_ChildDef_Optional('b | i');
$input = array(
new MF_StartTag('b')
,new MF_Text('Bold text')
,new MF_EndTag('b')
,new MF_EmptyTag('img') // illegal tag
);
$expect = array(
new MF_StartTag('b')
,new MF_Text('Bold text')
,new MF_EndTag('b')
);
$this->assertEqual($expect, $def->validateChildren($input));
$input = array( $inputs[0] = '<b>Bold text</b><img />';
new MF_Text('Not allowed text') $expect[0] = '<b>Bold text</b>';
);
$expect = array(); $inputs[1] = 'Not allowed text';
$this->assertEqual($expect, $def->validateChildren($input)); $expect[1] = '';
$this->assertSeries($inputs, $expect, $def);
} }
} }
@ -188,13 +128,13 @@ class Test_HTMLDTD_ChildDef extends UnitTestCase
class Test_PureHTMLDefinition extends UnitTestCase class Test_PureHTMLDefinition extends UnitTestCase
{ {
var $def, $lexer; var $def, $lex;
function Test_PureHTMLDefinition() { function Test_PureHTMLDefinition() {
$this->UnitTestCase(); $this->UnitTestCase();
$this->def = new PureHTMLDefinition(); $this->def = new PureHTMLDefinition();
$this->def->loadData(); $this->def->loadData();
$this->lexer = new HTML_Lexer(); $this->lex = new HTML_Lexer();
} }
function test_removeForeignElements() { function test_removeForeignElements() {
@ -206,35 +146,35 @@ class Test_PureHTMLDefinition extends UnitTestCase
$expect[0] = $inputs[0]; $expect[0] = $inputs[0];
$inputs[1] = array( $inputs[1] = array(
new MF_Text('This is ') new HTMLPurifier_Token_Text('This is ')
,new MF_StartTag('b', array()) ,new HTMLPurifier_Token_Start('b', array())
,new MF_Text('bold') ,new HTMLPurifier_Token_Text('bold')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
,new MF_Text(' text') ,new HTMLPurifier_Token_Text(' text')
); );
$expect[1] = $inputs[1]; $expect[1] = $inputs[1];
$inputs[2] = array( $inputs[2] = array(
new MF_StartTag('asdf') new HTMLPurifier_Token_Start('asdf')
,new MF_EndTag('asdf') ,new HTMLPurifier_Token_End('asdf')
,new MF_StartTag('d', array('href' => 'bang!')) ,new HTMLPurifier_Token_Start('d', array('href' => 'bang!'))
,new MF_EndTag('d') ,new HTMLPurifier_Token_End('d')
,new MF_StartTag('pooloka') ,new HTMLPurifier_Token_Start('pooloka')
,new MF_StartTag('poolasdf') ,new HTMLPurifier_Token_Start('poolasdf')
,new MF_StartTag('ds', array('moogle' => '&')) ,new HTMLPurifier_Token_Start('ds', array('moogle' => '&'))
,new MF_EndTag('asdf') ,new HTMLPurifier_Token_End('asdf')
,new MF_EndTag('asdf') ,new HTMLPurifier_Token_End('asdf')
); );
$expect[2] = array( $expect[2] = array(
new MF_Text('<asdf>') new HTMLPurifier_Token_Text('<asdf>')
,new MF_Text('</asdf>') ,new HTMLPurifier_Token_Text('</asdf>')
,new MF_Text('<d href="bang!">') ,new HTMLPurifier_Token_Text('<d href="bang!">')
,new MF_Text('</d>') ,new HTMLPurifier_Token_Text('</d>')
,new MF_Text('<pooloka>') ,new HTMLPurifier_Token_Text('<pooloka>')
,new MF_Text('<poolasdf>') ,new HTMLPurifier_Token_Text('<poolasdf>')
,new MF_Text('<ds moogle="&amp;">') ,new HTMLPurifier_Token_Text('<ds moogle="&amp;">')
,new MF_Text('</asdf>') ,new HTMLPurifier_Token_Text('</asdf>')
,new MF_Text('</asdf>') ,new HTMLPurifier_Token_Text('</asdf>')
); );
foreach ($inputs as $i => $input) { foreach ($inputs as $i => $input) {
@ -254,113 +194,113 @@ class Test_PureHTMLDefinition extends UnitTestCase
$expect[0] = $inputs[0]; $expect[0] = $inputs[0];
$inputs[1] = array( $inputs[1] = array(
new MF_Text('This is ') new HTMLPurifier_Token_Text('This is ')
,new MF_StartTag('b') ,new HTMLPurifier_Token_Start('b')
,new MF_Text('bold') ,new HTMLPurifier_Token_Text('bold')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
,new MF_Text(' text') ,new HTMLPurifier_Token_Text(' text')
,new MF_EmptyTag('br') ,new HTMLPurifier_Token_Empty('br')
); );
$expect[1] = $inputs[1]; $expect[1] = $inputs[1];
$inputs[2] = array( $inputs[2] = array(
new MF_StartTag('b') new HTMLPurifier_Token_Start('b')
,new MF_Text('Unclosed tag, gasp!') ,new HTMLPurifier_Token_Text('Unclosed tag, gasp!')
); );
$expect[2] = array( $expect[2] = array(
new MF_StartTag('b') new HTMLPurifier_Token_Start('b')
,new MF_Text('Unclosed tag, gasp!') ,new HTMLPurifier_Token_Text('Unclosed tag, gasp!')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
); );
$inputs[3] = array( $inputs[3] = array(
new MF_StartTag('b') new HTMLPurifier_Token_Start('b')
,new MF_StartTag('i') ,new HTMLPurifier_Token_Start('i')
,new MF_Text('The b is closed, but the i is not') ,new HTMLPurifier_Token_Text('The b is closed, but the i is not')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
); );
$expect[3] = array( $expect[3] = array(
new MF_StartTag('b') new HTMLPurifier_Token_Start('b')
,new MF_StartTag('i') ,new HTMLPurifier_Token_Start('i')
,new MF_Text('The b is closed, but the i is not') ,new HTMLPurifier_Token_Text('The b is closed, but the i is not')
,new MF_EndTag('i') ,new HTMLPurifier_Token_End('i')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
); );
$inputs[4] = array( $inputs[4] = array(
new MF_Text('Hey, recycle unused end tags!') new HTMLPurifier_Token_Text('Hey, recycle unused end tags!')
,new MF_EndTag('b') ,new HTMLPurifier_Token_End('b')
); );
$expect[4] = array( $expect[4] = array(
new MF_Text('Hey, recycle unused end tags!') new HTMLPurifier_Token_Text('Hey, recycle unused end tags!')
,new MF_Text('</b>') ,new HTMLPurifier_Token_Text('</b>')
); );
$inputs[5] = array(new MF_StartTag('br', array('style' => 'clear:both;'))); $inputs[5] = array(new HTMLPurifier_Token_Start('br', array('style' => 'clear:both;')));
$expect[5] = array(new MF_EmptyTag('br', array('style' => 'clear:both;'))); $expect[5] = array(new HTMLPurifier_Token_Empty('br', array('style' => 'clear:both;')));
$inputs[6] = array(new MF_EmptyTag('div', array('style' => 'clear:both;'))); $inputs[6] = array(new HTMLPurifier_Token_Empty('div', array('style' => 'clear:both;')));
$expect[6] = array( $expect[6] = array(
new MF_StartTag('div', array('style' => 'clear:both;')) new HTMLPurifier_Token_Start('div', array('style' => 'clear:both;'))
,new MF_EndTag('div') ,new HTMLPurifier_Token_End('div')
); );
// test automatic paragraph closing // test automatic paragraph closing
$inputs[7] = array( $inputs[7] = array(
new MF_StartTag('p') new HTMLPurifier_Token_Start('p')
,new MF_Text('Paragraph 1') ,new HTMLPurifier_Token_Text('Paragraph 1')
,new MF_StartTag('p') ,new HTMLPurifier_Token_Start('p')
,new MF_Text('Paragraph 2') ,new HTMLPurifier_Token_Text('Paragraph 2')
); );
$expect[7] = array( $expect[7] = array(
new MF_StartTag('p') new HTMLPurifier_Token_Start('p')
,new MF_Text('Paragraph 1') ,new HTMLPurifier_Token_Text('Paragraph 1')
,new MF_EndTag('p') ,new HTMLPurifier_Token_End('p')
,new MF_StartTag('p') ,new HTMLPurifier_Token_Start('p')
,new MF_Text('Paragraph 2') ,new HTMLPurifier_Token_Text('Paragraph 2')
,new MF_EndTag('p') ,new HTMLPurifier_Token_End('p')
); );
$inputs[8] = array( $inputs[8] = array(
new MF_StartTag('div') new HTMLPurifier_Token_Start('div')
,new MF_StartTag('p') ,new HTMLPurifier_Token_Start('p')
,new MF_Text('Paragraph 1 in a div') ,new HTMLPurifier_Token_Text('Paragraph 1 in a div')
,new MF_EndTag('div') ,new HTMLPurifier_Token_End('div')
); );
$expect[8] = array( $expect[8] = array(
new MF_StartTag('div') new HTMLPurifier_Token_Start('div')
,new MF_StartTag('p') ,new HTMLPurifier_Token_Start('p')
,new MF_Text('Paragraph 1 in a div') ,new HTMLPurifier_Token_Text('Paragraph 1 in a div')
,new MF_EndTag('p') ,new HTMLPurifier_Token_End('p')
,new MF_EndTag('div') ,new HTMLPurifier_Token_End('div')
); );
// automatic list closing // automatic list closing
$inputs[9] = array( $inputs[9] = array(
new MF_StartTag('ol') new HTMLPurifier_Token_Start('ol')
,new MF_StartTag('li') ,new HTMLPurifier_Token_Start('li')
,new MF_Text('Item 1') ,new HTMLPurifier_Token_Text('Item 1')
,new MF_StartTag('li') ,new HTMLPurifier_Token_Start('li')
,new MF_Text('Item 2') ,new HTMLPurifier_Token_Text('Item 2')
,new MF_EndTag('ol') ,new HTMLPurifier_Token_End('ol')
); );
$expect[9] = array( $expect[9] = array(
new MF_StartTag('ol') new HTMLPurifier_Token_Start('ol')
,new MF_StartTag('li') ,new HTMLPurifier_Token_Start('li')
,new MF_Text('Item 1') ,new HTMLPurifier_Token_Text('Item 1')
,new MF_EndTag('li') ,new HTMLPurifier_Token_End('li')
,new MF_StartTag('li') ,new HTMLPurifier_Token_Start('li')
,new MF_Text('Item 2') ,new HTMLPurifier_Token_Text('Item 2')
,new MF_EndTag('li') ,new HTMLPurifier_Token_End('li')
,new MF_EndTag('ol') ,new HTMLPurifier_Token_End('ol')
); );
foreach ($inputs as $i => $input) { foreach ($inputs as $i => $input) {
@ -379,62 +319,62 @@ class Test_PureHTMLDefinition extends UnitTestCase
// legal inline nesting // legal inline nesting
$inputs[0] = array( $inputs[0] = array(
new MF_StartTag('b'), new HTMLPurifier_Token_Start('b'),
new MF_Text('Bold text'), new HTMLPurifier_Token_Text('Bold text'),
new MF_EndTag('b'), new HTMLPurifier_Token_End('b'),
); );
$expect[0] = $inputs[0]; $expect[0] = $inputs[0];
// legal inline and block // legal inline and block
// as the parent element is considered FLOW // as the parent element is considered FLOW
$inputs[1] = array( $inputs[1] = array(
new MF_StartTag('a', array('href' => 'http://www.example.com/')), new HTMLPurifier_Token_Start('a', array('href' => 'http://www.example.com/')),
new MF_Text('Linky'), new HTMLPurifier_Token_Text('Linky'),
new MF_EndTag('a'), new HTMLPurifier_Token_End('a'),
new MF_StartTag('div'), new HTMLPurifier_Token_Start('div'),
new MF_Text('Block element'), new HTMLPurifier_Token_Text('Block element'),
new MF_EndTag('div'), new HTMLPurifier_Token_End('div'),
); );
$expect[1] = $inputs[1]; $expect[1] = $inputs[1];
// illegal block in inline, element -> text // illegal block in inline, element -> text
$inputs[2] = array( $inputs[2] = array(
new MF_StartTag('b'), new HTMLPurifier_Token_Start('b'),
new MF_StartTag('div'), new HTMLPurifier_Token_Start('div'),
new MF_Text('Illegal Div'), new HTMLPurifier_Token_Text('Illegal Div'),
new MF_EndTag('div'), new HTMLPurifier_Token_End('div'),
new MF_EndTag('b'), new HTMLPurifier_Token_End('b'),
); );
$expect[2] = array( $expect[2] = array(
new MF_StartTag('b'), new HTMLPurifier_Token_Start('b'),
new MF_Text('<div>'), new HTMLPurifier_Token_Text('<div>'),
new MF_Text('Illegal Div'), new HTMLPurifier_Token_Text('Illegal Div'),
new MF_Text('</div>'), new HTMLPurifier_Token_Text('</div>'),
new MF_EndTag('b'), new HTMLPurifier_Token_End('b'),
); );
// test of empty set that's required, resulting in removal of node // test of empty set that's required, resulting in removal of node
$inputs[3] = array( $inputs[3] = array(
new MF_StartTag('ul'), new HTMLPurifier_Token_Start('ul'),
new MF_EndTag('ul') new HTMLPurifier_Token_End('ul')
); );
$expect[3] = array(); $expect[3] = array();
// test illegal text which gets removed // test illegal text which gets removed
$inputs[4] = array( $inputs[4] = array(
new MF_StartTag('ul'), new HTMLPurifier_Token_Start('ul'),
new MF_Text('Illegal Text'), new HTMLPurifier_Token_Text('Illegal Text'),
new MF_StartTag('li'), new HTMLPurifier_Token_Start('li'),
new MF_Text('Legal item'), new HTMLPurifier_Token_Text('Legal item'),
new MF_EndTag('li'), new HTMLPurifier_Token_End('li'),
new MF_EndTag('ul') new HTMLPurifier_Token_End('ul')
); );
$expect[4] = array( $expect[4] = array(
new MF_StartTag('ul'), new HTMLPurifier_Token_Start('ul'),
new MF_StartTag('li'), new HTMLPurifier_Token_Start('li'),
new MF_Text('Legal item'), new HTMLPurifier_Token_Text('Legal item'),
new MF_EndTag('li'), new HTMLPurifier_Token_End('li'),
new MF_EndTag('ul') new HTMLPurifier_Token_End('ul')
); );
foreach ($inputs as $i => $input) { foreach ($inputs as $i => $input) {