xref: /plugin/siteexport/inc/pdfgenerator.php (revision a8c17ab5b37308343f86651acb8c4a1b3f36f0ae) 
1<?php
2
3if (!defined('DOKU_PLUGIN')) die('meh');
4
5if (!empty($_REQUEST['pdfExport']) && intval($_REQUEST['pdfExport']) == 1 && file_exists(DOKU_PLUGIN . 'dw2pdf/mpdf/mpdf.php')) {
6
7    require_once(DOKU_PLUGIN . 'siteexport/inc/mpdf.php');
8    class siteexport_pdfgenerator
9    {
10        private $functions;
11
12        public function __construct($functions = null)
13        {
14            $this->functions = $functions;
15        }
16
17        public function createPDFFromFile($filename, &$NAME) {
18
19            if (!preg_match("/" . $this->functions->settings->fileType . "$/", $NAME)) {
20                $this->functions->debug->message("Filetype " . $this->functions->settings->fileType . " did not match filename '$NAME'", null, 4);
21                return false;
22            }
23
24            $mpdf = new siteexportPDF($this->functions->debug);
25
26            if (!$mpdf) {
27                $this->functions->debug->message("Could not instantiate MPDF", null, 4);
28                return false;
29            }
30
31            $html = @file_get_contents($filename);
32
33            if (!strstr($html, "<html")) {
34                $this->functions->debug->message("Filecontent had no HTML starting tag", null, 4);
35                return false;
36            }
37
38            // Save HTML too
39            $this->functions->debug->message("Arranging HTML", null, 2);
40            $this->arrangeHtml($html, 'bl,acronym');
41            $this->functions->debug->message("Done arranging HTML:", $html, 1);
42
43            $mpdf->debug = false;
44            $mpdf->list_indent_first_level = 1; // Indents the first level of lists.
45
46            $mpdf->usepre = false;
47            $mpdf->margin_bottom_collapse = true;
48            $mpdf->SetDisplayMode('fullpage');
49            $mpdf->restoreBlockPageBreaks = true;
50            $mpdf->img_dpi = 300;
51
52            $mpdf->setBasePath(empty($this->functions->settings->depth) ? './' : $this->functions->settings->depth);
53
54            $mpdf->ignore_invalid_utf8 = true;
55            $mpdf->mirrorMargins = $this->functions->getConf('useOddEven'); // don't mirror margins
56            $mpdf->setAutoTopMargin = 'pad';
57            $mpdf->setAutoBottomMargin = 'pad';
58
59            $mpdf->WriteHTML($html);
60            $mpdf->Output($filename, "F");
61
62            return $html;
63        }
64
65        private function arrangeHtml(&$html, $norendertags = '')
66        {
67            global $conf;
68
69            // add bookmark links
70            $html = preg_replace_callback("/<h(\d)(.*?)>(.*?)<\/h\\1>/s", array($this, '__pdfHeaderCallback'), $html);
71            $html = preg_replace_callback("/<\/div>\s*?<h({$conf['plugin']['siteexport']['PDFHeaderPagebreak']})(.*?)>/s", array($this, '__pdfHeaderCallbackPagebreak'), $html);
72            $html = preg_replace("/(<img.*?mediacenter.*?\/>)/", "<table style=\"width:100%; border: 0px solid #000;\"><tr><td style=\"text-align: center\">$1</td></tr></table>", $html);
73
74            // Remove p arround img and table
75            $html = preg_replace("/<p[^>]*?>(\s*?<img[^>]*?\/?>\s*?)<\/p>/s", "$1", $html);
76            $html = preg_replace("/<p[^>]*?>(\s*?<table.*?<\/table>\s*?)<\/p>/s", "$1", $html);
77            $html = preg_replace_callback("/<pre(.*?)>(.*?)<\/pre>/s", array($this, '__pdfPreCodeCallback'), $html);
78            $html = preg_replace_callback("/<a href=\"mailto:(.*?)\".*?>(.*?)<\/a>/s", array($this, '__pdfMailtoCallback'), $html);
79            /**/
80
81            $standardReplacer = array(
82            // insert a pagebreak for support of WRAP and PAGEBREAK plugins
83                                    '<br style="page-break-after:always;">' => '<pagebreak />',
84                                    '<div class="wrap_pagebreak"></div>' => '<pagebreak />',
85                                    '<sup>' => '<sup class="sup">',
86                                    '<sub>' => '<sub class="sub">',
87                                    '<code>' => '<code class="code">'
88            );
89            $html = str_replace(array_keys($standardReplacer), array_values($standardReplacer), $html);
90
91            // thanks to Jared Ong
92            // Customized to strip all span tags so that the wiki <code> SQL would display properly
93            $norender = explode(',', $norendertags);
94            $html = $this->strip_only($html, $norender); //array('span','acronym'));
95            $html = $this->strip_htmlencodedchars($html);
96            // Customized to strip all span tags so that the wiki <code> SQL would display properly
97        }
98
99        private function __pdfMailtoCallback($DATA) {
100            if ($DATA[1] == $DATA[2]) {
101                $DATA[2] = $this->deobfuscate($DATA[2]);
102            }
103            $DATA[1] = $this->deobfuscate($DATA[1]);
104            return "<a href=\"mailto:{$DATA[1]}\">{$DATA[2]}</a>";
105        }
106
107        private function __pdfPreCodeCallback($DATA) {
108
109            $code = nl2br($DATA[2]);
110            $code = preg_replace_callback("/(^|<br \/>)(\s+)(\S)/s", array($this, '__pdfPreWhitespacesCallback'), $code);
111
112            return "\n<pre" . $DATA[1] . ">\n" . $code . "\n</pre>\n";
113        }
114
115        private function __pdfPreWhitespacesCallback($DATA) {
116            return $DATA[1] . "\n" . str_repeat("&nbsp;", strlen($DATA[2])-($DATA[2]{0} == "\n" ? 1 : 0)) . $DATA[3];
117        }
118
119        private function __pdfHeaderCallback($DATA) {
120            $contentText = htmlspecialchars_decode(preg_replace("/<\/?.*?>/s", '', $DATA[3]), ENT_NOQUOTES); // 2014-07-23 Do not encode again. or &auml; -> &amp;auml;
121            return '<h' . $DATA[1] . $DATA[2] . '><tocentry content="' . $contentText . '" level="' . ($DATA[1]-1) . '" /><bookmark content="' . $contentText . '" level="' . ($DATA[1]-1) . '" />' . $DATA[3] . '</h' . $DATA[1] . '>';
122        }
123
124        private function __pdfHeaderCallbackPagebreak($DATA) {
125            return '</div>' . "\r\n" . '<pagebreak />' . "\r\n\r\n" . '<h' . $DATA[1] . $DATA[2] . '>';
126        }
127        // thanks to Jared Ong
128        // Custom function for help in stripping span tags
129        private function strip_only($str, $tags) {
130            if (!is_array($tags)) {
131                $tags = (strpos($str, '>') !== false ? explode('>', str_replace('<', '', $tags)) : array($tags));
132                if (end($tags) == '') array_pop($tags);
133            }
134
135            foreach ($tags as $tag) $str = preg_replace('#</?' . $tag . '[^>]*>#is', '', $str);
136            return $str;
137        }
138        // Custom function for help in stripping span tags
139
140        // Custom function for help in replacing &#039; &quot; &gt; &lt; &amp;
141        private function strip_htmlencodedchars($str) {
142            $str = str_replace('&#039;', '\'', $str);
143            return $str;
144        }
145        // Custom function for help in replacing &#039; &quot; &gt; &lt; &amp;
146
147        /**
148         * return an de-obfuscated email address in line with $conf['mailguard'] setting
149         */
150        private function deobfuscate($email) {
151            global $conf;
152
153            switch ($conf['mailguard']) {
154                case 'visible' :
155                    return /** @scrutinizer ignore-call */ strtr($email, array(' [at] ' => '@', ' [dot] ' => '.', ' [dash] ' => '-'));
156
157                case 'hex' :
158                    $encode = '';
159                    $len = strlen($email);
160                    for ($x = 0; $x < $len; $x += 6) {
161                        $encode .= chr((int)hexdec($email{$x+3} . $email{($x+4)}));
162                    }
163                    return $encode;
164
165                case 'none' :
166                default :
167                    return $email;
168            }
169        }
170    }
171}
172