pierre/wallabag
1
0
Fork 0
forked from wallabag/wallabag
Code Issues Pull Requests Packages Projects Releases 1 Wiki Activity

Remove htmlawed and use graby instead

Browse Source 
Instead of using htmlawed (which is already used in graby) use graby directly (which require some refacto on graby side).
Still needs some tests.
This commit is contained in:
Jeremy Benoist
2017-05-16 23:11:20 +02:00
parent cf05a1ae34
commit 0d6cfb884c
3 changed files with 14 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
composer.json
View File

@ -61,7 +61,6 @@
"wallabag/tcpdf": "^6.2", "wallabag/tcpdf": "^6.2",
"simplepie/simplepie": "~1.3.1", "simplepie/simplepie": "~1.3.1",
"willdurand/hateoas-bundle": "~1.0", "willdurand/hateoas-bundle": "~1.0",
"htmlawed/htmlawed": "~1.1.19",
"liip/theme-bundle": "~1.1", "liip/theme-bundle": "~1.1",
"lexik/form-filter-bundle": "~5.0", "lexik/form-filter-bundle": "~5.0",
"j0k3r/graby": "dev-extractor", "j0k3r/graby": "dev-extractor",

19
src/Wallabag/CoreBundle/Helper/ContentProxy.php
View File

@ -47,24 +47,7 @@ class ContentProxy
{ {
// ensure content is a bit cleaned up // ensure content is a bit cleaned up
if (!empty($content['html'])) { if (!empty($content['html'])) {
$extractor = $this->graby->getExtractor(); $content['html'] = $this->graby->cleanupHtml($content['html'], $url);
$contentExtracted = $extractor->process($content['html'], $url);
if ($contentExtracted) {
$contentBlock = $extractor->getContent();
$contentBlock->normalize();
$content['html'] = trim($contentBlock->innerHTML);
}
$content['html'] = htmLawed($content['html'], [
'safe' => 1,
// which means: do not remove iframe elements
'elements' => '*+iframe',
'deny_attribute' => 'style',
'comment' => 1,
'cdata' => 1,
]);
} }
// do we have to fetch the content or the provided one is ok? // do we have to fetch the content or the provided one is ok?

26
tests/Wallabag/CoreBundle/Helper/ContentProxyTest.php
View File

@ -210,16 +210,18 @@ class ContentProxyTest extends \PHPUnit_Framework_TestCase
$tagger->expects($this->once()) $tagger->expects($this->once())
->method('tag'); ->method('tag');
$graby = $this->getMockBuilder('Graby\Graby')->getMock(); $proxy = new ContentProxy((new Graby()), $tagger, $this->getLogger(), $this->fetchingErrorMessage);
$entry = $proxy->updateEntry(
$proxy = new ContentProxy($graby, $tagger, $this->getLogger(), $this->fetchingErrorMessage); new Entry(new User()),
$entry = $proxy->updateEntry(new Entry(new User()), 'http://0.0.0.0', [ 'http://0.0.0.0',
'html' => str_repeat('this is my content', 325), [
'title' => 'this is my title', 'html' => str_repeat('this is my content', 325),
'url' => 'http://1.1.1.1', 'title' => 'this is my title',
'content_type' => 'text/html', 'url' => 'http://1.1.1.1',
'language' => 'fr', 'content_type' => 'text/html',
]); 'language' => 'fr',
]
);
$this->assertEquals('http://1.1.1.1', $entry->getUrl()); $this->assertEquals('http://1.1.1.1', $entry->getUrl());
$this->assertEquals('this is my title', $entry->getTitle()); $this->assertEquals('this is my title', $entry->getTitle());
@ -277,9 +279,7 @@ class ContentProxyTest extends \PHPUnit_Framework_TestCase
$tagger->expects($this->once()) $tagger->expects($this->once())
->method('tag'); ->method('tag');
$graby = new Graby(); $proxy = new ContentProxy((new Graby()), $tagger, $this->getLogger(), $this->fetchingErrorMessage);
$proxy = new ContentProxy($graby, $tagger, $this->getTagRepositoryMock(), $this->getLogger(), $this->fetchingErrorMessage);
$entry = $proxy->updateEntry( $entry = $proxy->updateEntry(
new Entry(new User()), new Entry(new User()),
'http://1.1.1.1', 'http://1.1.1.1',