]> BookStack Code Mirror - bookstack/blob - app/Entities/Tools/Markdown/HtmlToMarkdown.php
Fixed failing test after drawio default url change
[bookstack] / app / Entities / Tools / Markdown / HtmlToMarkdown.php
1 <?php
2
3 namespace BookStack\Entities\Tools\Markdown;
4
5 use League\HTMLToMarkdown\Converter\BlockquoteConverter;
6 use League\HTMLToMarkdown\Converter\CodeConverter;
7 use League\HTMLToMarkdown\Converter\CommentConverter;
8 use League\HTMLToMarkdown\Converter\DivConverter;
9 use League\HTMLToMarkdown\Converter\EmphasisConverter;
10 use League\HTMLToMarkdown\Converter\HardBreakConverter;
11 use League\HTMLToMarkdown\Converter\HeaderConverter;
12 use League\HTMLToMarkdown\Converter\HorizontalRuleConverter;
13 use League\HTMLToMarkdown\Converter\ImageConverter;
14 use League\HTMLToMarkdown\Converter\LinkConverter;
15 use League\HTMLToMarkdown\Converter\ListBlockConverter;
16 use League\HTMLToMarkdown\Converter\ListItemConverter;
17 use League\HTMLToMarkdown\Converter\PreformattedConverter;
18 use League\HTMLToMarkdown\Converter\TextConverter;
19 use League\HTMLToMarkdown\Environment;
20 use League\HTMLToMarkdown\HtmlConverter;
21
22 class HtmlToMarkdown
23 {
24     protected $html;
25
26     public function __construct(string $html)
27     {
28         $this->html = $html;
29     }
30
31     /**
32      * Run the conversion.
33      */
34     public function convert(): string
35     {
36         $converter = new HtmlConverter($this->getConverterEnvironment());
37         $html = $this->prepareHtml($this->html);
38
39         return $converter->convert($html);
40     }
41
42     /**
43      * Run any pre-processing to the HTML to clean it up manually before conversion.
44      */
45     protected function prepareHtml(string $html): string
46     {
47         // Carriage returns can cause whitespace issues in output
48         $html = str_replace("\r\n", "\n", $html);
49         // Attributes on the pre tag can cause issues with conversion
50         return preg_replace('/<pre .*?>/', '<pre>', $html);
51     }
52
53     /**
54      * Get the HTML to Markdown customized environment.
55      * Extends the default provided environment with some BookStack specific tweaks.
56      */
57     protected function getConverterEnvironment(): Environment
58     {
59         $environment = new Environment([
60             'header_style'            => 'atx', // Set to 'atx' to output H1 and H2 headers as # Header1 and ## Header2
61             'suppress_errors'         => true, // Set to false to show warnings when loading malformed HTML
62             '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.
63             'strip_placeholder_links' => false, // Set to true to remove <a> that doesn't have href.
64             'bold_style'              => '**', // DEPRECATED: Set to '__' if you prefer the underlined style
65             'italic_style'            => '*', // DEPRECATED: Set to '_' if you prefer the underlined style
66             'remove_nodes'            => '', // space-separated list of dom nodes that should be removed. example: 'meta style script'
67             'hard_break'              => false, // Set to true to turn <br> into `\n` instead of `  \n`
68             'list_item_style'         => '-', // Set the default character for each <li> in a <ul>. Can be '-', '*', or '+'
69             'preserve_comments'       => false, // Set to true to preserve comments, or set to an array of strings to preserve specific comments
70             'use_autolinks'           => false, // Set to true to use simple link syntax if possible. Will always use []() if set to false
71             'table_pipe_escape'       => '\|', // Replacement string for pipe characters inside markdown table cells
72             'table_caption_side'      => 'top', // Set to 'top' or 'bottom' to show <caption> content before or after table, null to suppress
73         ]);
74
75         $environment->addConverter(new BlockquoteConverter());
76         $environment->addConverter(new CodeConverter());
77         $environment->addConverter(new CommentConverter());
78         $environment->addConverter(new DivConverter());
79         $environment->addConverter(new EmphasisConverter());
80         $environment->addConverter(new HardBreakConverter());
81         $environment->addConverter(new HeaderConverter());
82         $environment->addConverter(new HorizontalRuleConverter());
83         $environment->addConverter(new ImageConverter());
84         $environment->addConverter(new LinkConverter());
85         $environment->addConverter(new ListBlockConverter());
86         $environment->addConverter(new ListItemConverter());
87         $environment->addConverter(new CustomParagraphConverter());
88         $environment->addConverter(new PreformattedConverter());
89         $environment->addConverter(new TextConverter());
90         $environment->addConverter(new CheckboxConverter());
91
92         return $environment;
93     }
94 }