3 namespace BookStack\Entities\Tools\Markdown;
5 use League\HTMLToMarkdown\Converter\BlockquoteConverter;
6 use League\HTMLToMarkdown\Converter\CodeConverter;
7 use League\HTMLToMarkdown\Converter\CommentConverter;
8 use League\HTMLToMarkdown\Converter\EmphasisConverter;
9 use League\HTMLToMarkdown\Converter\HardBreakConverter;
10 use League\HTMLToMarkdown\Converter\HeaderConverter;
11 use League\HTMLToMarkdown\Converter\HorizontalRuleConverter;
12 use League\HTMLToMarkdown\Converter\LinkConverter;
13 use League\HTMLToMarkdown\Converter\ListBlockConverter;
14 use League\HTMLToMarkdown\Converter\ListItemConverter;
15 use League\HTMLToMarkdown\Converter\PreformattedConverter;
16 use League\HTMLToMarkdown\Converter\TextConverter;
17 use League\HTMLToMarkdown\Environment;
18 use League\HTMLToMarkdown\HtmlConverter;
22 protected string $html;
24 public function __construct(string $html)
32 public function convert(): string
34 $converter = new HtmlConverter($this->getConverterEnvironment());
35 $html = $this->prepareHtml($this->html);
37 return $converter->convert($html);
41 * Run any pre-processing to the HTML to clean it up manually before conversion.
43 protected function prepareHtml(string $html): string
45 // Carriage returns can cause whitespace issues in output
46 $html = str_replace("\r\n", "\n", $html);
47 // Attributes on the pre tag can cause issues with conversion
48 return preg_replace('/<pre .*?>/', '<pre>', $html);
52 * Get the HTML to Markdown customized environment.
53 * Extends the default provided environment with some BookStack specific tweaks.
55 protected function getConverterEnvironment(): Environment
57 $environment = new Environment([
58 'header_style' => 'atx', // Set to 'atx' to output H1 and H2 headers as # Header1 and ## Header2
59 'suppress_errors' => true, // Set to false to show warnings when loading malformed HTML
60 '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.
61 'strip_placeholder_links' => false, // Set to true to remove <a> that doesn't have href.
62 'bold_style' => '**', // DEPRECATED: Set to '__' if you prefer the underlined style
63 'italic_style' => '*', // DEPRECATED: Set to '_' if you prefer the underlined style
64 'remove_nodes' => '', // space-separated list of dom nodes that should be removed. example: 'meta style script'
65 'hard_break' => false, // Set to true to turn <br> into `\n` instead of ` \n`
66 'list_item_style' => '-', // Set the default character for each <li> in a <ul>. Can be '-', '*', or '+'
67 'preserve_comments' => false, // Set to true to preserve comments, or set to an array of strings to preserve specific comments
68 'use_autolinks' => false, // Set to true to use simple link syntax if possible. Will always use []() if set to false
69 'table_pipe_escape' => '\|', // Replacement string for pipe characters inside markdown table cells
70 'table_caption_side' => 'top', // Set to 'top' or 'bottom' to show <caption> content before or after table, null to suppress
73 $environment->addConverter(new BlockquoteConverter());
74 $environment->addConverter(new CodeConverter());
75 $environment->addConverter(new CommentConverter());
76 $environment->addConverter(new CustomDivConverter());
77 $environment->addConverter(new EmphasisConverter());
78 $environment->addConverter(new HardBreakConverter());
79 $environment->addConverter(new HeaderConverter());
80 $environment->addConverter(new HorizontalRuleConverter());
81 $environment->addConverter(new CustomImageConverter());
82 $environment->addConverter(new LinkConverter());
83 $environment->addConverter(new ListBlockConverter());
84 $environment->addConverter(new ListItemConverter());
85 $environment->addConverter(new CustomParagraphConverter());
86 $environment->addConverter(new PreformattedConverter());
87 $environment->addConverter(new TextConverter());
88 $environment->addConverter(new CheckboxConverter());
89 $environment->addConverter(new SpacedTagFallbackConverter());