]> BookStack Code Mirror - bookstack/blob - app/Entities/Tools/Markdown/HtmlToMarkdown.php
Updated markdown export implementation
[bookstack] / app / Entities / Tools / Markdown / HtmlToMarkdown.php
1 <?php namespace BookStack\Entities\Tools\Markdown;
2
3 use League\HTMLToMarkdown\Converter\BlockquoteConverter;
4 use League\HTMLToMarkdown\Converter\CodeConverter;
5 use League\HTMLToMarkdown\Converter\CommentConverter;
6 use League\HTMLToMarkdown\Converter\DivConverter;
7 use League\HTMLToMarkdown\Converter\EmphasisConverter;
8 use League\HTMLToMarkdown\Converter\HardBreakConverter;
9 use League\HTMLToMarkdown\Converter\HeaderConverter;
10 use League\HTMLToMarkdown\Converter\HorizontalRuleConverter;
11 use League\HTMLToMarkdown\Converter\ImageConverter;
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;
19
20 class HtmlToMarkdown
21 {
22     protected $html;
23
24     public function __construct(string $html)
25     {
26         $this->html = $html;
27     }
28
29     /**
30      * Run the conversion
31      */
32     public function convert(): string
33     {
34         $converter = new HtmlConverter($this->getConverterEnvironment());
35         $html = $this->prepareHtml($this->html);
36         return $converter->convert($html);
37     }
38
39     /**
40      * Run any pre-processing to the HTML to clean it up manually before conversion.
41      */
42     protected function prepareHtml(string $html): string
43     {
44         // Carriage returns can cause whitespace issues in output
45         $html = str_replace("\r\n", "\n", $html);
46         // Attributes on the pre tag can cause issues with conversion
47         return preg_replace('/<pre .*?>/', '<pre>', $html);
48     }
49
50     /**
51      * Get the HTML to Markdown customized environment.
52      * Extends the default provided environment with some BookStack specific tweaks.
53      */
54     protected function getConverterEnvironment(): Environment
55     {
56         $environment = new Environment(['header_style' => 'atx']);
57
58         $environment->addConverter(new BlockquoteConverter());
59         $environment->addConverter(new CodeConverter());
60         $environment->addConverter(new CommentConverter());
61         $environment->addConverter(new DivConverter());
62         $environment->addConverter(new EmphasisConverter());
63         $environment->addConverter(new HardBreakConverter());
64         $environment->addConverter(new HeaderConverter());
65         $environment->addConverter(new HorizontalRuleConverter());
66         $environment->addConverter(new ImageConverter());
67         $environment->addConverter(new LinkConverter());
68         $environment->addConverter(new ListBlockConverter());
69         $environment->addConverter(new ListItemConverter());
70         $environment->addConverter(new CustomParagraphConverter());
71         $environment->addConverter(new PreformattedConverter());
72         $environment->addConverter(new TextConverter());
73
74         return $environment;
75     }
76 }