All scripts in one repository

This commit is contained in:
mmk2410 2015-11-12 22:36:23 +01:00
parent a0a86492ee
commit 9f9304d6aa
61 changed files with 6668 additions and 681 deletions

View file

@ -0,0 +1,60 @@
<?php
namespace League\HTMLToMarkdown;
class Configuration
{
protected $config;
/**
* @param array $config
*/
public function __construct(array $config = array())
{
$this->config = $config;
}
/**
* @param array $config
*/
public function merge(array $config = array())
{
$this->config = array_replace_recursive($this->config, $config);
}
/**
* @param array $config
*/
public function replace(array $config = array())
{
$this->config = $config;
}
/**
* @param string $key
* @param mixed $value
*/
public function setOption($key, $value)
{
$this->config[$key] = $value;
}
/**
* @param string|null $key
* @param mixed|null $default
*
* @return mixed|null
*/
public function getOption($key = null, $default = null)
{
if ($key === null) {
return $this->config;
}
if (!isset($this->config[$key])) {
return $default;
}
return $this->config[$key];
}
}

View file

@ -0,0 +1,11 @@
<?php
namespace League\HTMLToMarkdown;
interface ConfigurationAwareInterface
{
/**
* @param Configuration $config
*/
public function setConfig(Configuration $config);
}

View file

@ -0,0 +1,44 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class BlockquoteConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
// Contents should have already been converted to Markdown by this point,
// so we just need to add '>' symbols to each line.
$markdown = '';
$quote_content = trim($element->getValue());
$lines = preg_split('/\r\n|\r|\n/', $quote_content);
$total_lines = count($lines);
foreach ($lines as $i => $line) {
$markdown .= '> ' . $line . "\n";
if ($i + 1 === $total_lines) {
$markdown .= "\n";
}
}
return $markdown;
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('blockquote');
}
}

View file

@ -0,0 +1,26 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class CommentConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
return '';
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('#comment');
}
}

View file

@ -0,0 +1,20 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
interface ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element);
/**
* @return string[]
*/
public function getSupportedTags();
}

View file

@ -0,0 +1,50 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\Configuration;
use League\HTMLToMarkdown\ConfigurationAwareInterface;
use League\HTMLToMarkdown\ElementInterface;
class DefaultConverter implements ConverterInterface, ConfigurationAwareInterface
{
const DEFAULT_CONVERTER = '_default';
/**
* @var Configuration
*/
protected $config;
/**
* @param Configuration $config
*/
public function setConfig(Configuration $config)
{
$this->config = $config;
}
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
// If strip_tags is false (the default), preserve tags that don't have Markdown equivalents,
// such as <span> nodes on their own. C14N() canonicalizes the node to a string.
// See: http://www.php.net/manual/en/domnode.c14n.php
if ($this->config->getOption('strip_tags', false)) {
return $element->getValue();
}
return html_entity_decode($element->getChildrenAsString());
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array(self::DEFAULT_CONVERTER);
}
}

View file

@ -0,0 +1,45 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\Configuration;
use League\HTMLToMarkdown\ConfigurationAwareInterface;
use League\HTMLToMarkdown\ElementInterface;
class DivConverter implements ConverterInterface, ConfigurationAwareInterface
{
/**
* @var Configuration
*/
protected $config;
/**
* @param Configuration $config
*/
public function setConfig(Configuration $config)
{
$this->config = $config;
}
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
if ($this->config->getOption('strip_tags', false)) {
return $element->getValue() . "\n\n";
}
return html_entity_decode($element->getChildrenAsString());
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('div');
}
}

View file

@ -0,0 +1,50 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\Configuration;
use League\HTMLToMarkdown\ConfigurationAwareInterface;
use League\HTMLToMarkdown\ElementInterface;
class EmphasisConverter implements ConverterInterface, ConfigurationAwareInterface
{
/**
* @var Configuration
*/
protected $config;
/**
* @param Configuration $config
*/
public function setConfig(Configuration $config)
{
$this->config = $config;
}
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
$tag = $element->getTagName();
$value = $element->getValue();
if ($tag === 'i' || $tag === 'em') {
$style = $this->config->getOption('italic_style');
} else {
$style = $this->config->getOption('bold_style');
}
return $style . $value . $style;
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('em', 'i', 'strong', 'b');
}
}

View file

@ -0,0 +1,26 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class HardBreakConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
return " \n";
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('br');
}
}

View file

@ -0,0 +1,78 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\Configuration;
use League\HTMLToMarkdown\ConfigurationAwareInterface;
use League\HTMLToMarkdown\ElementInterface;
class HeaderConverter implements ConverterInterface, ConfigurationAwareInterface
{
const STYLE_ATX = 'atx';
const STYLE_SETEXT = 'setext';
/**
* @var Configuration
*/
protected $config;
/**
* @param Configuration $config
*/
public function setConfig(Configuration $config)
{
$this->config = $config;
}
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
$level = (int) substr($element->getTagName(), 1, 1);
$style = $this->config->getOption('header_style', self::STYLE_SETEXT);
if (($level === 1 || $level === 2) && !$element->isDescendantOf('blockquote') && $style === self::STYLE_SETEXT) {
return $this->createSetextHeader($level, $element->getValue());
} else {
return $this->createAtxHeader($level, $element->getValue());
}
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('h1', 'h2', 'h3', 'h4', 'h5', 'h6');
}
/**
* @param int $level
* @param string $content
*
* @return string
*/
private function createSetextHeader($level, $content)
{
$length = (function_exists('mb_strlen')) ? mb_strlen($content, 'utf-8') : strlen($content);
$underline = ($level === 1) ? '=' : '-';
return $content . "\n" . str_repeat($underline, $length) . "\n\n";
}
/**
* @param int $level
* @param string $content
*
* @return string
*/
private function createAtxHeader($level, $content)
{
$prefix = str_repeat('#', $level) . ' ';
return $prefix . $content . "\n\n";
}
}

View file

@ -0,0 +1,26 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class HorizontalRuleConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
return "- - - - - -\n\n";
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('hr');
}
}

View file

@ -0,0 +1,37 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class ImageConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
$src = $element->getAttribute('src');
$alt = $element->getAttribute('alt');
$title = $element->getAttribute('title');
if ($title !== '') {
// No newlines added. <img> should be in a block-level element.
$markdown = '![' . $alt . '](' . $src . ' "' . $title . '")';
} else {
$markdown = '![' . $alt . '](' . $src . ')';
}
return $markdown;
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('img');
}
}

View file

@ -0,0 +1,42 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class LinkConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
$href = $element->getAttribute('href');
$title = $element->getAttribute('title');
$text = $element->getValue();
if ($title !== '') {
$markdown = '[' . $text . '](' . $href . ' "' . $title . '")';
} elseif ($href === $text) {
$markdown = '<' . $href . '>';
} else {
$markdown = '[' . $text . '](' . $href . ')';
}
if (!$href) {
$markdown = html_entity_decode($element->getChildrenAsString());
}
return $markdown;
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('a');
}
}

View file

@ -0,0 +1,26 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class ListBlockConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
return $element->getValue() . "\n";
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('ol', 'ul');
}
}

View file

@ -0,0 +1,37 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class ListItemConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
// If parent is an ol, use numbers, otherwise, use dashes
$list_type = $element->getParent()->getTagName();
$value = $element->getValue();
if ($list_type === 'ul') {
$markdown = '- ' . trim($value) . "\n";
} else {
$number = $element->getSiblingPosition();
$markdown = $number . '. ' . trim($value) . "\n";
}
return $markdown;
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('li');
}
}

View file

@ -0,0 +1,28 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class ParagraphConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
$value = $element->getValue();
return (trim($value)) ? rtrim($value) . "\n\n" : '';
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('p');
}
}

View file

@ -0,0 +1,73 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class PreformattedConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
// Store the content of the code block in an array, one entry for each line
$markdown = '';
$code_content = html_entity_decode($element->getChildrenAsString());
$code_content = str_replace(array('<code>', '</code>'), '', $code_content);
$code_content = str_replace(array('<pre>', '</pre>'), '', $code_content);
$lines = preg_split('/\r\n|\r|\n/', $code_content);
$total = count($lines);
// If there's more than one line of code, prepend each line with four spaces and no backticks.
if ($total > 1 || $element->getTagName() === 'pre') {
// Remove the first and last line if they're empty
$first_line = trim($lines[0]);
$last_line = trim($lines[$total - 1]);
$first_line = trim($first_line, '&#xD;'); //trim XML style carriage returns too
$last_line = trim($last_line, '&#xD;');
if (empty($first_line)) {
array_shift($lines);
}
if (empty($last_line)) {
array_pop($lines);
}
$count = 1;
foreach ($lines as $line) {
$line = str_replace('&#xD;', '', $line);
$markdown .= ' ' . $line;
// Add newlines, except final line of the code
if ($count !== $total) {
$markdown .= "\n";
}
$count++;
}
$markdown .= "\n";
} else {
// There's only one line of code. It's a code span, not a block. Just wrap it with backticks.
$markdown .= '`' . $lines[0] . '`';
}
if ($element->getTagName() === 'pre') {
$markdown = "\n" . $markdown . "\n";
}
return $markdown;
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('pre', 'code');
}
}

View file

@ -0,0 +1,38 @@
<?php
namespace League\HTMLToMarkdown\Converter;
use League\HTMLToMarkdown\ElementInterface;
class TextConverter implements ConverterInterface
{
/**
* @param ElementInterface $element
*
* @return string
*/
public function convert(ElementInterface $element)
{
$value = $element->getValue();
$markdown = preg_replace('~\s+~', ' ', $value);
$markdown = preg_replace('~^#~', '\\\\#', $markdown);
if ($markdown === ' ') {
$next = $element->getNext();
if (!$next || $next->isBlock()) {
$markdown = '';
}
}
return $markdown;
}
/**
* @return string[]
*/
public function getSupportedTags()
{
return array('#text');
}
}

View file

@ -0,0 +1,234 @@
<?php
namespace League\HTMLToMarkdown;
class Element implements ElementInterface
{
/**
* @var \DOMNode
*/
protected $node;
/**
* @var ElementInterface|null
*/
private $nextCached;
public function __construct(\DOMNode $node)
{
$this->node = $node;
}
/**
* @return bool
*/
public function isBlock()
{
switch ($this->getTagName()) {
case 'blockquote':
case 'body':
case 'code':
case 'div':
case 'h1':
case 'h2':
case 'h3':
case 'h4':
case 'h5':
case 'h6':
case 'hr':
case 'html':
case 'li':
case 'p':
case 'ol':
case 'ul':
return true;
default:
return false;
}
}
/**
* @return bool
*/
public function isText()
{
return $this->getTagName() === '#text';
}
/**
* @return bool
*/
public function isWhitespace()
{
return $this->getTagName() === '#text' && trim($this->getValue()) === '';
}
/**
* @return string
*/
public function getTagName()
{
return $this->node->nodeName;
}
/**
* @return string
*/
public function getValue()
{
return $this->node->nodeValue;
}
/**
* @return ElementInterface|null
*/
public function getParent()
{
return new static($this->node->parentNode) ?: null;
}
/**
* @return bool
*/
public function hasChildren()
{
return $this->node->hasChildNodes();
}
/**
* @return ElementInterface[]
*/
public function getChildren()
{
$ret = array();
/** @var \DOMNode $node */
foreach ($this->node->childNodes as $node) {
$ret[] = new static($node);
}
return $ret;
}
/**
* @return ElementInterface|null
*/
public function getNext()
{
if ($this->nextCached === null) {
$nextNode = $this->getNextNode($this->node);
if ($nextNode !== null) {
$this->nextCached = new static($nextNode);
}
}
return $this->nextCached;
}
/**
* @param \DomNode $node
*
* @return \DomNode|null
*/
private function getNextNode($node, $checkChildren = true)
{
if ($checkChildren && $node->firstChild) {
return $node->firstChild;
} elseif ($node->nextSibling) {
return $node->nextSibling;
} elseif ($node->parentNode) {
return $this->getNextNode($node->parentNode, false);
}
}
/**
* @param string[]|string $tagNames
*
* @return bool
*/
public function isDescendantOf($tagNames)
{
if (!is_array($tagNames)) {
$tagNames = array($tagNames);
}
for ($p = $this->node->parentNode; $p !== false; $p = $p->parentNode) {
if (is_null($p)) {
return false;
}
if (in_array($p->nodeName, $tagNames)) {
return true;
}
}
return false;
}
/**
* @param string $markdown
*/
public function setFinalMarkdown($markdown)
{
$markdown_node = $this->node->ownerDocument->createTextNode($markdown);
$this->node->parentNode->replaceChild($markdown_node, $this->node);
}
/**
* @return string
*/
public function getChildrenAsString()
{
return $this->node->C14N();
}
/**
* @return int
*/
public function getSiblingPosition()
{
$position = 0;
// Loop through all nodes and find the given $node
foreach ($this->getParent()->getChildren() as $current_node) {
if (!$current_node->isWhitespace()) {
$position++;
}
// TODO: Need a less-buggy way of comparing these
// Perhaps we can somehow ensure that we always have the exact same object and use === instead?
if ($this->equals($current_node)) {
break;
}
}
return $position;
}
/**
* @param string $name
*
* @return string
*/
public function getAttribute($name)
{
if ($this->node instanceof \DOMElement) {
return $this->node->getAttribute($name);
}
return '';
}
/**
* @param ElementInterface $element
*
* @return bool
*/
public function equals(ElementInterface $element)
{
if ($element instanceof self) {
return $element->node === $this->node;
}
return $element === $this;
}
}

View file

@ -0,0 +1,80 @@
<?php
namespace League\HTMLToMarkdown;
interface ElementInterface
{
/**
* @return bool
*/
public function isBlock();
/**
* @return bool
*/
public function isText();
/**
* @return bool
*/
public function isWhitespace();
/**
* @return string
*/
public function getTagName();
/**
* @return string
*/
public function getValue();
/**
* @return ElementInterface|null
*/
public function getParent();
/**
* @param string|string[] $tagNames
*
* @return bool
*/
public function isDescendantOf($tagNames);
/**
* @return bool
*/
public function hasChildren();
/**
* @return ElementInterface[]
*/
public function getChildren();
/**
* @return ElementInterface|null
*/
public function getNext();
/**
* @return int
*/
public function getSiblingPosition();
/**
* @return string
*/
public function getChildrenAsString();
/**
* @param string $markdown
*/
public function setFinalMarkdown($markdown);
/**
* @param string $name
*
* @return string
*/
public function getAttribute($name);
}

View file

@ -0,0 +1,102 @@
<?php
namespace League\HTMLToMarkdown;
use League\HTMLToMarkdown\Converter\BlockquoteConverter;
use League\HTMLToMarkdown\Converter\CommentConverter;
use League\HTMLToMarkdown\Converter\ConverterInterface;
use League\HTMLToMarkdown\Converter\DefaultConverter;
use League\HTMLToMarkdown\Converter\DivConverter;
use League\HTMLToMarkdown\Converter\EmphasisConverter;
use League\HTMLToMarkdown\Converter\HardBreakConverter;
use League\HTMLToMarkdown\Converter\HeaderConverter;
use League\HTMLToMarkdown\Converter\HorizontalRuleConverter;
use League\HTMLToMarkdown\Converter\ImageConverter;
use League\HTMLToMarkdown\Converter\LinkConverter;
use League\HTMLToMarkdown\Converter\ListBlockConverter;
use League\HTMLToMarkdown\Converter\ListItemConverter;
use League\HTMLToMarkdown\Converter\ParagraphConverter;
use League\HTMLToMarkdown\Converter\PreformattedConverter;
use League\HTMLToMarkdown\Converter\TextConverter;
final class Environment
{
/**
* @var Configuration
*/
protected $config;
/**
* @var ConverterInterface[]
*/
protected $converters = array();
public function __construct(array $config = array())
{
$this->config = new Configuration($config);
$this->addConverter(new DefaultConverter());
}
/**
* @return Configuration
*/
public function getConfig()
{
return $this->config;
}
/**
* @param ConverterInterface $converter
*/
public function addConverter(ConverterInterface $converter)
{
if ($converter instanceof ConfigurationAwareInterface) {
$converter->setConfig($this->config);
}
foreach ($converter->getSupportedTags() as $tag) {
$this->converters[$tag] = $converter;
}
}
/**
* @param string $tag
*
* @return ConverterInterface
*/
public function getConverterByTag($tag)
{
if (isset($this->converters[$tag])) {
return $this->converters[$tag];
}
return $this->converters[DefaultConverter::DEFAULT_CONVERTER];
}
/**
* @param array $config
*
* @return Environment
*/
public static function createDefaultEnvironment(array $config = array())
{
$environment = new static($config);
$environment->addConverter(new BlockquoteConverter());
$environment->addConverter(new CommentConverter());
$environment->addConverter(new DivConverter());
$environment->addConverter(new EmphasisConverter());
$environment->addConverter(new HardBreakConverter());
$environment->addConverter(new HeaderConverter());
$environment->addConverter(new HorizontalRuleConverter());
$environment->addConverter(new ImageConverter());
$environment->addConverter(new LinkConverter());
$environment->addConverter(new ListBlockConverter());
$environment->addConverter(new ListItemConverter());
$environment->addConverter(new ParagraphConverter());
$environment->addConverter(new PreformattedConverter());
$environment->addConverter(new TextConverter());
return $environment;
}
}

View file

@ -0,0 +1,194 @@
<?php
namespace League\HTMLToMarkdown;
/**
* Class HtmlConverter
*
* A helper class to convert HTML to Markdown.
*
* @author Colin O'Dell <colinodell@gmail.com>
* @author Nick Cernis <nick@cern.is>
*
* @link https://github.com/thephpleague/html-to-markdown/ Latest version on GitHub.
*
* @license http://www.opensource.org/licenses/mit-license.php MIT
*/
class HtmlConverter
{
/**
* @var Environment
*/
protected $environment;
/**
* Constructor
*
* @param array $options Configuration options
*/
public function __construct(array $options = array())
{
$defaults = array(
'header_style' => 'setext', // Set to 'atx' to output H1 and H2 headers as # Header1 and ## Header2
'suppress_errors' => true, // Set to false to show warnings when loading malformed HTML
'strip_tags' => false, // Set to true to strip tags that don't have markdown equivalents. N.B. Strips tags, not their content. Useful to clean MS Word HTML output.
'bold_style' => '**', // Set to '__' if you prefer the underlined style
'italic_style' => '*', // Set to '_' if you prefer the underlined style
'remove_nodes' => '', // space-separated list of dom nodes that should be removed. example: 'meta style script'
);
$this->environment = Environment::createDefaultEnvironment($defaults);
$this->environment->getConfig()->merge($options);
}
/**
* @return Environment
*/
public function getEnvironment()
{
return $this->environment;
}
/**
* @return Configuration
*/
public function getConfig()
{
return $this->environment->getConfig();
}
/**
* Convert
*
* Loads HTML and passes to getMarkdown()
*
* @param $html
*
* @return string The Markdown version of the html
*/
public function convert($html)
{
if (trim($html) === '') {
return '';
}
$document = $this->createDOMDocument($html);
// Work on the entire DOM tree (including head and body)
if (!($root = $document->getElementsByTagName('html')->item(0))) {
throw new \InvalidArgumentException('Invalid HTML was provided');
}
$rootElement = new Element($root);
$this->convertChildren($rootElement);
// Store the now-modified DOMDocument as a string
$markdown = $document->saveHTML();
$markdown = $this->sanitize($markdown);
return $markdown;
}
/**
* @param string $html
*
* @return \DOMDocument
*/
private function createDOMDocument($html)
{
$document = new \DOMDocument();
if ($this->getConfig()->getOption('suppress_errors')) {
// Suppress conversion errors (from http://bit.ly/pCCRSX)
libxml_use_internal_errors(true);
}
// Hack to load utf-8 HTML (from http://bit.ly/pVDyCt)
$document->loadHTML('<?xml encoding="UTF-8">' . $html);
$document->encoding = 'UTF-8';
if ($this->getConfig()->getOption('suppress_errors')) {
libxml_clear_errors();
}
return $document;
}
/**
* Convert Children
*
* Recursive function to drill into the DOM and convert each node into Markdown from the inside out.
*
* Finds children of each node and convert those to #text nodes containing their Markdown equivalent,
* starting with the innermost element and working up to the outermost element.
*
* @param ElementInterface $element
*/
private function convertChildren(ElementInterface $element)
{
// Don't convert HTML code inside <code> and <pre> blocks to Markdown - that should stay as HTML
if ($element->isDescendantOf(array('pre', 'code'))) {
return;
}
// If the node has children, convert those to Markdown first
if ($element->hasChildren()) {
foreach ($element->getChildren() as $child) {
$this->convertChildren($child);
}
}
// Now that child nodes have been converted, convert the original node
$markdown = $this->convertToMarkdown($element);
// Create a DOM text node containing the Markdown equivalent of the original node
// Replace the old $node e.g. '<h3>Title</h3>' with the new $markdown_node e.g. '### Title'
$element->setFinalMarkdown($markdown);
}
/**
* Convert to Markdown
*
* Converts an individual node into a #text node containing a string of its Markdown equivalent.
*
* Example: An <h3> node with text content of 'Title' becomes a text node with content of '### Title'
*
* @param ElementInterface $element
*
* @return string The converted HTML as Markdown
*/
protected function convertToMarkdown(ElementInterface $element)
{
$tag = $element->getTagName();
// Strip nodes named in remove_nodes
$tags_to_remove = explode(' ', $this->getConfig()->getOption('remove_nodes'));
if (in_array($tag, $tags_to_remove)) {
return false;
}
$converter = $this->environment->getConverterByTag($tag);
return $converter->convert($element);
}
/**
* @param string $markdown
*
* @return string
*/
protected function sanitize($markdown)
{
$markdown = html_entity_decode($markdown, ENT_QUOTES, 'UTF-8');
$markdown = html_entity_decode($markdown, ENT_QUOTES, 'UTF-8'); // Double decode to cover cases like &amp;nbsp; http://www.php.net/manual/en/function.htmlentities.php#99984
$markdown = preg_replace('/<!DOCTYPE [^>]+>/', '', $markdown); // Strip doctype declaration
$unwanted = array('<html>', '</html>', '<body>', '</body>', '<head>', '</head>', '<?xml encoding="UTF-8">', '&#xD;');
$markdown = str_replace($unwanted, '', $markdown); // Strip unwanted tags
$markdown = trim($markdown, "\n\r\0\x0B");
return $markdown;
}
}