mirror/php-htmlpurifier
1
0
Fork 0
mirror of https://github.com/ezyang/htmlpurifier.git synced 2025-07-31 03:10:09 +02:00
Code Issues Releases Wiki Activity

Scratch PI or JASP from spec. Add extra comments about MarkupFragment's behavior to incoming data. More Todos for our home-brew lexer, and also fix empty tag support in the SAX parer adapter. Add test framework for marking SAX to be tested with the regular result, or a specialized result.

Browse Source 
git-svn-id: http://htmlpurifier.org/svnroot/html_purifier/trunk@22 48356398-32a2-884e-a903-53898d9a118a
This commit is contained in:
Edward Z. Yang
2006-04-15 02:19:27 +00:00
parent 181d544b61
commit dad395ac45
4 changed files with 77 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files

78
tests/HTML_Lexer.php
View File

@@ -1,5 +1,9 @@
<?php
/* TODO
* Benchmark the SAX parser with my homemade one
*/
class TestCase_HTML_Lexer extends UnitTestCase
{
@@ -23,16 +27,20 @@ class TestCase_HTML_Lexer extends UnitTestCase
function test_tokenizeHTML() {
$input[] = '';
$expect[] = array();
$input = array();
$expect = array();
$sax_expect = array();
$input[] = 'This is regular text.';
$expect[] = array(
$input[0] = '';
$expect[0] = array();
$input[1] = 'This is regular text.';
$expect[1] = array(
new MF_Text('This is regular text.')
);
$input[] = 'This is <b>bold</b> text';
$expect[] = array(
$input[2] = 'This is <b>bold</b> text';
$expect[2] = array(
new MF_Text('This is ')
,new MF_StartTag('b', array())
,new MF_Text('bold')
@@ -40,8 +48,8 @@ class TestCase_HTML_Lexer extends UnitTestCase
,new MF_Text(' text')
);
$input[] = '<DIV>Totally rad dude. <b>asdf</b></div>';
$expect[] = array(
$input[3] = '<DIV>Totally rad dude. <b>asdf</b></div>';
$expect[3] = array(
new MF_StartTag('DIV', array())
,new MF_Text('Totally rad dude. ')
,new MF_StartTag('b', array())
@@ -50,8 +58,8 @@ class TestCase_HTML_Lexer extends UnitTestCase
,new MF_EndTag('div')
);
$input[] = '<asdf></asdf><d></d><poOloka><poolasdf><ds></asdf></ASDF>';
$expect[] = array(
$input[4] = '<asdf></asdf><d></d><poOloka><poolasdf><ds></asdf></ASDF>';
$expect[4] = array(
new MF_StartTag('asdf')
,new MF_EndTag('asdf')
,new MF_StartTag('d')
@@ -63,8 +71,8 @@ class TestCase_HTML_Lexer extends UnitTestCase
,new MF_EndTag('ASDF')
);
$input[] = '<a'."\t".'href="foobar.php"'."\n".'title="foo!">Link to <b id="asdf">foobar</b></a>';
$expect[] = array(
$input[5] = '<a'."\t".'href="foobar.php"'."\n".'title="foo!">Link to <b id="asdf">foobar</b></a>';
$expect[5] = array(
new MF_StartTag('a',array('href'=>'foobar.php','title'=>'foo!'))
,new MF_Text('Link to ')
,new MF_StartTag('b',array('id'=>'asdf'))
@@ -73,32 +81,56 @@ class TestCase_HTML_Lexer extends UnitTestCase
,new MF_EndTag('a')
);
$input[] = '<br />';
$expect[] = array(
$input[6] = '<br />';
$expect[6] = array(
new MF_EmptyTag('br')
);
$input[] = '<!-- Comment --> <!-- not so well formed --->';
$expect[] = array(
// [INVALID] [RECOVERABLE]
$input[7] = '<!-- Comment --> <!-- not so well formed --->';
$expect[7] = array(
new MF_Comment(' Comment ')
,new MF_Text(' ')
,new MF_Comment(' not so well formed -')
);
$sax_expect[7] = false; // we need to figure out proper comment output
$input[] = '<a href=""';
$expect[] = array(
// [INVALID]
$input[8] = '<a href=""';
$expect[8] = array(
new MF_Text('<a href=""')
);
// SAX parses it into a tag
$sax_expect[8] = array(
new MF_StartTag('a', array('href'=>''))
);
$size = count($input);
for($i = 0; $i < $size; $i++) {
$input[9] = '&lt;b&gt;';
$expect[9] = array(
new MF_Text('&lt;b&gt;')
);
// however, we may want to change both styles
// into parsed: '<b>'. SAX has an option for this
foreach($input as $i => $discard) {
$result = $this->HTML_Lexer->tokenizeHTML($input[$i]);
$this->assertEqual($expect[$i], $result);
paintIf($result, $expect[$i] != $result);
// since I didn't write the parser, I can't define its behavior
// however, make sure that the class runs without any errors
$exp_result = $this->HTML_Lexer_Sax->tokenizeHTML($input[$i]);
// assert unless I say otherwise
$sax_result = $this->HTML_Lexer_Sax->tokenizeHTML($input[$i]);
if (!isset($sax_expect[$i])) {
// by default, assert with normal result
$this->assertEqual($expect[$i], $sax_result);
paintIf($sax_result, $expect[$i] != $sax_result);
} elseif ($sax_expect[$i] === false) {
// assertions were turned off, optionally dump
// paintIf($sax_expect, $i == NUMBER);
} else {
// match with a custom SAX result array
$this->assertEqual($sax_expect[$i], $sax_result);
paintIf($sax_result, $sax_expect[$i] != $sax_result);
}
}
}