Skip to content

New algorithm for plain text values #168

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Mar 29, 2018
Merged

New algorithm for plain text values #168

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e1de3ea
Introduce new way of getting plain text values from HTML elements
Zegnat Mar 26, 2018
7dbe03d
Fix three tests that failed with the new algorithm
Zegnat Mar 26, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Next Next commit
Introduce new way of getting plain text values from HTML elements
  • Loading branch information
@Zegnat
Zegnat committed Mar 26, 2018
commit e1de3ea2c0661bc35d8c6a3344111d95e0b0d267
134 changes: 41 additions & 93 deletions Mf2/Parser.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -438,96 +438,44 @@ private function resolveChildUrls(DOMElement $el) {
}
}

public function textContent(DOMElement $el) {
$excludeTags = array('noframe', 'noscript', 'script', 'style', 'frames', 'frameset');

if (isset($el->tagName) and in_array(strtolower($el->tagName), $excludeTags)) {
return '';
}

$this->resolveChildUrls($el);

$clonedEl = $el->cloneNode(true);

foreach ($this->xpath->query('.//img', $clonedEl) as $imgEl) {
$newNode = $this->doc->createTextNode($imgEl->getAttribute($imgEl->hasAttribute('alt') ? 'alt' : 'src'));
$imgEl->parentNode->replaceChild($newNode, $imgEl);
}

foreach ($excludeTags as $tagName) {
foreach ($this->xpath->query(".//{$tagName}", $clonedEl) as $elToRemove) {
$elToRemove->parentNode->removeChild($elToRemove);
}
}

return $this->innerText($clonedEl);
/**
* The following two methods implements plain text parsing.
* @see https://wiki.zegnat.net/media/textparsing.html
**/
public function textContent(DOMElement $element)
{
return preg_replace(
'/(^[\t\n\f\r ]+| +(?=\n)|(?<=\n) +| +(?= )|[\t\n\f\r ]+$)/',
'',
$this->elementToString($element)
);
}

/**
* This method attempts to return a better 'innerText' representation than DOMNode::textContent
*
* @param DOMElement|DOMText $el
* @param bool $implied when parsing for implied name for h-*, rules may be slightly different
* @see: https://github.com/glennjones/microformat-shiv/blob/dev/lib/text.js
*/
public function innerText($el, $implied=false) {
$out = '';

$blockLevelTags = array('h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'p', 'hr', 'pre', 'table',
'address', 'article', 'aside', 'blockquote', 'caption', 'col', 'colgroup', 'dd', 'div',
'dt', 'dir', 'fieldset', 'figcaption', 'figure', 'footer', 'form', 'header', 'hgroup', 'hr',
'li', 'map', 'menu', 'nav', 'optgroup', 'option', 'section', 'tbody', 'testarea',
'tfoot', 'th', 'thead', 'tr', 'td', 'ul', 'ol', 'dl', 'details');

$excludeTags = array('noframe', 'noscript', 'script', 'style', 'frames', 'frameset');

// PHP DOMDocument doesn’t correctly handle whitespace around elements it doesn’t recognise.
$unsupportedTags = array('data');

if (isset($el->tagName)) {
if (in_array(strtolower($el->tagName), $excludeTags)) {
return $out;
} else if ($el->tagName == 'img') {
if ($el->hasAttribute('alt')) {
return $el->getAttribute('alt');
} else if (!$implied && $el->hasAttribute('src')) {
return $this->resolveUrl($el->getAttribute('src'));
}
} else if ($el->tagName == 'area' and $el->hasAttribute('alt')) {
return $el->getAttribute('alt');
} else if ($el->tagName == 'abbr' and $el->hasAttribute('title')) {
return $el->getAttribute('title');
}
}

// if node is a text node get its text
if (isset($el->nodeType) && $el->nodeType === 3) {
$out .= $el->textContent;
}

// get the text of the child nodes
if ($el->childNodes && $el->childNodes->length > 0) {
for ($j = 0; $j < $el->childNodes->length; $j++) {
$text = $this->innerText($el->childNodes->item($j), $implied);
if (!is_null($text)) {
$out .= $text;
}
}
}

if (isset($el->tagName)) {
// if its a block level tag add an additional space at the end
if (in_array(strtolower($el->tagName), $blockLevelTags)) {
$out .= ' ';
} elseif ($implied and in_array(strtolower($el->tagName), $unsupportedTags)) {
$out .= ' ';
} else if (strtolower($el->tagName) == 'br') {
// else if its a br, replace with newline
$out .= "\n";
}
}

return ($out === '') ? NULL : $out;
private function elementToString(DOMElement $input)
{
$output = '';
foreach ($input->childNodes as $child) {
if ($child->nodeType === XML_TEXT_NODE) {
$output .= str_replace(array("\t", "\n", "\r") , ' ', $child->textContent);
} else if ($child->nodeType === XML_ELEMENT_NODE) {
$tagName = strtoupper($child->tagName);
if (in_array($tagName, array('SCRIPT', 'STYLE'))) {
continue;
} else if ($tagName === 'IMG') {
if ($child->hasAttribute('alt')) {
$output .= ' ' . trim($child->getAttribute('alt'), "\t\n\f\r ") . ' ';
} else if ($child->hasAttribute('src')) {
$output .= ' ' . $this->resolveUrl(trim($child->getAttribute('src'), "\t\n\f\r ")) . ' ';
}
} else if ($tagName === 'BR') {
$output .= "\n";
} else if ($tagName === 'P') {
$output .= "\n" . $this->elementToString($child);
} else {
$output .= $this->elementToString($child);
}
}
}
return $output;
}

/**
Expand Down Expand Up @@ -643,7 +591,7 @@ public function parseP(\DOMElement $p) {
} elseif (in_array($p->tagName, array('data', 'input')) and $p->hasAttribute('value')) {
$pValue = $p->getAttribute('value');
} else {
$pValue = unicodeTrim($this->innerText($p));
$pValue = $this->textContent($p);
}

return $pValue;
Expand Down Expand Up @@ -680,7 +628,7 @@ public function parseU(\DOMElement $u) {
} elseif (in_array($u->tagName, array('data', 'input')) and $u->hasAttribute('value')) {
return $u->getAttribute('value');
} else {
return unicodeTrim($this->textContent($u));
return $this->textContent($u);
}
}

Expand Down Expand Up @@ -911,7 +859,7 @@ public function parseE(\DOMElement $e) {

$return = array(
'html' => unicodeTrim($html),
'value' => unicodeTrim($this->innerText($e)),
'value' => $this->textContent($e),
);

if($this->lang) {
Expand Down Expand Up @@ -1118,7 +1066,7 @@ public function parseH(\DOMElement $e, $is_backcompat = false, $has_nested_mf =
}
}

throw new Exception($this->innerText($e, true));
throw new Exception($this->textContent($e, true));
} catch (Exception $exc) {
$return['name'][] = unicodeTrim($exc->getMessage());
}
Expand Down
76 changes: 76 additions & 0 deletions tests/Mf2/PlainTextTest.php
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,76 @@
<?php

namespace Mf2\Parser\Test;

class PlainTextTest extends \PHPUnit_Framework_TestCase {
/**
* @dataProvider aaronpkExpectations
*/
public function testAaronpkExpectations($input, $pName, $eValue, $eHtml) {
$parser = new \Mf2\Parser($input);
$output = $parser->parse();
$entryProperties = $output['items'][0]['properties'];
$this->assertEquals($pName, $entryProperties['name'][0]);
$this->assertEquals($eValue, $entryProperties['content'][0]['value']);
$this->assertEquals($eHtml, $entryProperties['content'][0]['html']);
}

public function aaronpkExpectations() {
return array(
1 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\"><p>Hello World</p></div>\n</div>",
"Hello World",
"Hello World",
"<p>Hello World</p>"
),
2 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\"><p>Hello<br>World</p></div>\n</div>",
"Hello\nWorld",
"Hello\nWorld",
"<p>Hello<br>World</p>"
),
3 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\"><p>Hello<br>\nWorld</p></div>\n</div>",
"Hello\nWorld",
"Hello\nWorld",
"<p>Hello<br>\nWorld</p>"
),
4 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\">\n <p>Hello World</p>\n </div>\n</div>",
"Hello World",
"Hello World",
"<p>Hello World</p>"
),
5 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\">Hello\nWorld</div>\n</div>",
"Hello World",
"Hello World",
"Hello\nWorld"
),
6 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\"><p>Hello</p><p>World</p></div>\n</div>",
"Hello\nWorld",
"Hello\nWorld",
"<p>Hello</p><p>World</p>"
),
7 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\">Hello<br>\n World</div>\n</div>",
"Hello\nWorld",
"Hello\nWorld",
"Hello<br>\n World",
),
8 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\"><br>Hello<br>World<br></div>\n</div>",
"Hello\nWorld",
"Hello\nWorld",
"<br>Hello<br>World<br>"
),
9 => array(
"<div class=\"h-entry\">\n <div class=\"e-content p-name\">\n <p>One</p>\n <p>Two</p>\n <p>Three</p>\n </div>\n</div>",
"One\nTwo\nThree",
"One\nTwo\nThree",
"<p>One</p>\n <p>Two</p>\n <p>Three</p>"
)
);
}
}