From fd8a5ff4c49fc1361f9928ea4d33e6d24d43a3a5 Mon Sep 17 00:00:00 2001 From: Mario Date: Sat, 5 Jun 2021 08:47:24 +0000 Subject: composer update league/html-to-markdown --- .../league/html-to-markdown/src/HtmlConverter.php | 116 ++++++++++----------- 1 file changed, 54 insertions(+), 62 deletions(-) (limited to 'vendor/league/html-to-markdown/src/HtmlConverter.php') diff --git a/vendor/league/html-to-markdown/src/HtmlConverter.php b/vendor/league/html-to-markdown/src/HtmlConverter.php index 6f98e97b4..7cd543b34 100644 --- a/vendor/league/html-to-markdown/src/HtmlConverter.php +++ b/vendor/league/html-to-markdown/src/HtmlConverter.php @@ -1,10 +1,10 @@ @@ -16,25 +16,24 @@ namespace League\HTMLToMarkdown; */ class HtmlConverter implements HtmlConverterInterface { - /** - * @var Environment - */ + /** @var Environment */ protected $environment; /** * Constructor * - * @param Environment|array $options Environment object or configuration options + * @param Environment|array $options Environment object or configuration options */ - public function __construct($options = array()) + public function __construct($options = []) { if ($options instanceof Environment) { $this->environment = $options; - } elseif (is_array($options)) { - $defaults = array( + } elseif (\is_array($options)) { + $defaults = [ '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. + 'strip_placeholder_links' => false, // Set to true to remove that doesn't have href. 'bold_style' => '**', // DEPRECATED: Set to '__' if you prefer the underlined style 'italic_style' => '*', // DEPRECATED: Set to '_' if you prefer the underlined style 'remove_nodes' => '', // space-separated list of dom nodes that should be removed. example: 'meta style script' @@ -42,7 +41,9 @@ class HtmlConverter implements HtmlConverterInterface 'list_item_style' => '-', // Set the default character for each
  • in a
      . Can be '-', '*', or '+' 'preserve_comments' => false, // Set to true to preserve comments, or set to an array of strings to preserve specific comments 'use_autolinks' => true, // Set to true to use simple link syntax if possible. Will always use []() if set to false - ); + 'table_pipe_escape' => '\|', // Replacement string for pipe characters inside markdown table cells + 'table_caption_side' => 'top', // Set to 'top' or 'bottom' to show content before or after table, null to suppress + ]; $this->environment = Environment::createDefaultEnvironment($defaults); @@ -50,18 +51,12 @@ class HtmlConverter implements HtmlConverterInterface } } - /** - * @return Environment - */ - public function getEnvironment() + public function getEnvironment(): Environment { return $this->environment; } - /** - * @return Configuration - */ - public function getConfig() + public function getConfig(): Configuration { return $this->environment->getConfig(); } @@ -71,11 +66,9 @@ class HtmlConverter implements HtmlConverterInterface * * @see HtmlConverter::convert * - * @param string $html - * * @return string The Markdown version of the html */ - public function __invoke($html) + public function __invoke(string $html): string { return $this->convert($html); } @@ -85,22 +78,20 @@ class HtmlConverter implements HtmlConverterInterface * * Loads HTML and passes to getMarkdown() * - * @param string $html - * - * @throws \InvalidArgumentException - * * @return string The Markdown version of the html + * + * @throws \InvalidArgumentException|\RuntimeException */ - public function convert($html) + public function convert(string $html): string { - if (trim($html) === '') { + if (\trim($html) === '') { return ''; } $document = $this->createDOMDocument($html); // Work on the entire DOM tree (including head and body) - if (!($root = $document->getElementsByTagName('html')->item(0))) { + if (! ($root = $document->getElementsByTagName('html')->item(0))) { throw new \InvalidArgumentException('Invalid HTML was provided'); } @@ -110,21 +101,20 @@ class HtmlConverter implements HtmlConverterInterface // Store the now-modified DOMDocument as a string $markdown = $document->saveHTML(); + if ($markdown === false) { + throw new \RuntimeException('Unknown error occurred during HTML to Markdown conversion'); + } + return $this->sanitize($markdown); } - /** - * @param string $html - * - * @return \DOMDocument - */ - private function createDOMDocument($html) + private function createDOMDocument(string $html): \DOMDocument { $document = new \DOMDocument(); if ($this->getConfig()->getOption('suppress_errors')) { // Suppress conversion errors (from http://bit.ly/pCCRSX) - libxml_use_internal_errors(true); + \libxml_use_internal_errors(true); } // Hack to load utf-8 HTML (from http://bit.ly/pVDyCt) @@ -132,7 +122,7 @@ class HtmlConverter implements HtmlConverterInterface $document->encoding = 'UTF-8'; if ($this->getConfig()->getOption('suppress_errors')) { - libxml_clear_errors(); + \libxml_clear_errors(); } return $document; @@ -145,17 +135,21 @@ class HtmlConverter implements HtmlConverterInterface * * 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) + private function convertChildren(ElementInterface $element): void { // Don't convert HTML code inside and
       blocks to Markdown - that should stay as HTML
               // except if the current node is a code tag, which needs to be converted by the CodeConverter.
      -        if ($element->isDescendantOf(array('pre', 'code')) && $element->getTagName() !== 'code') {
      +        if ($element->isDescendantOf(['pre', 'code']) && $element->getTagName() !== 'code') {
                   return;
               }
       
      +        // Give converter a chance to inspect/modify the DOM before children are converted
      +        $converter = $this->environment->getConverterByTag($element->getTagName());
      +        if ($converter instanceof PreConverterInterface) {
      +            $converter->preConvert($element);
      +        }
      +
               // If the node has children, convert those to Markdown first
               if ($element->hasChildren()) {
                   foreach ($element->getChildren() as $child) {
      @@ -179,18 +173,16 @@ class HtmlConverter implements HtmlConverterInterface
            *
            * Example: An 

      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) + protected function convertToMarkdown(ElementInterface $element): string { $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; + $tagsToRemove = \explode(' ', $this->getConfig()->getOption('remove_nodes') ?? ''); + if (\in_array($tag, $tagsToRemove, true)) { + return ''; } $converter = $this->environment->getConverterByTag($tag); @@ -198,38 +190,34 @@ class HtmlConverter implements HtmlConverterInterface return $converter->convert($element); } - /** - * @param string $markdown - * - * @return string - */ - protected function sanitize($markdown) + protected function sanitize(string $markdown): string { - $markdown = html_entity_decode($markdown, ENT_QUOTES, 'UTF-8'); - $markdown = preg_replace('/]+>/', '', $markdown); // Strip doctype declaration - $markdown = trim($markdown); // Remove blank spaces at the beggining of the html + $markdown = \html_entity_decode($markdown, ENT_QUOTES, 'UTF-8'); + $markdown = \preg_replace('/]+>/', '', $markdown); // Strip doctype declaration + \assert($markdown !== null); + $markdown = \trim($markdown); // Remove blank spaces at the beggining of the html /* * Removing unwanted tags. Tags should be added to the array in the order they are expected. * XML, html and body opening tags should be in that order. Same case with closing tags */ - $unwanted = array('', '', '', '', '', '', '', ' '); + $unwanted = ['', '', '', '', '', '', '', ' ']; foreach ($unwanted as $tag) { - if (strpos($tag, '/') === false) { + if (\strpos($tag, '/') === false) { // Opening tags - if (strpos($markdown, $tag) === 0) { - $markdown = substr($markdown, strlen($tag)); + if (\strpos($markdown, $tag) === 0) { + $markdown = \substr($markdown, \strlen($tag)); } } else { // Closing tags - if (strpos($markdown, $tag) === strlen($markdown) - strlen($tag)) { - $markdown = substr($markdown, 0, -strlen($tag)); + if (\strpos($markdown, $tag) === \strlen($markdown) - \strlen($tag)) { + $markdown = \substr($markdown, 0, -\strlen($tag)); } } } - return trim($markdown, "\n\r\0\x0B"); + return \trim($markdown, "\n\r\0\x0B"); } /** @@ -239,6 +227,10 @@ class HtmlConverter implements HtmlConverterInterface * An example being: * * HtmlConverter::setOptions(['strip_tags' => true])->convert('

      test

      '); + * + * @param array $options + * + * @return $this */ public function setOptions(array $options) { -- cgit v1.2.3