How to generate table of contents using dompdf? - php

Question

I am using the dompdf library of php to generate PDF report from an HTML template. In that html template there is a section table of contents. When genrating PDF i need to update the page number of table of contents. Does anyone know how I can achieve this in dompdf library of php?
Thanks in advance.

Answer 1

I have achieved this in Drupal and guess it works on other php opensource and frameworks as well.
I have kept this code inside script tag
$GLOBALS['entity_page'][] = $pdf->get_page_number();
in template that stores the page number. Template is with extension tpl.php
Now in the module
after other codes for export I have added......
$canvas = $dompdf->get_canvas();
$font = Font_Metrics::get_font("helvetica", "normal");
$canvas->page_text(520, 805, "Page {PAGE_NUM}", $font, 9, array(0.4, 0.4, 0.4));
foreach ($GLOBALS['entity_page'] as $key => $val) {
$GLOBALS["entity_val"] = 0;
$GLOBALS["entity_y"] = 110;
$canvas->page_script('if($PAGE_NUM == 3 && $PAGE_NUM < 4){
$font = Font_Metrics::get_font("helvetica", "normal");
$x = 380;
$y = $GLOBALS["entity_y"];
$pdf->text($x, $y, "-------------------------".$GLOBALS["entity_page"][$GLOBALS["entity_val"]]."", $font, 12, array(0, 0, 0, 0.8));
$GLOBALS["entity_y"] = $GLOBALS["entity_y"] + 33;
$GLOBALS["entity_val"] = $GLOBALS["entity_val"] + 1;
}');
}
$pdf->text this part adds the page numbers with constant increment in y axis position. Other global variables entity_y and entity_val are used to store the values.

Answer 2

Generating a Table of Contents from HTML (with h1,h2,h3), I did the following:
First give every header an unique ID (because we are using PrinceXML) and is a good pratice.
Then create an OL/LI structure of it, although that piece of code can contain bugs.
$matches = null;
$smatches = null;
$had_headers = array();
preg_match_all('/<h[0-9].*?>.*?<\/h[0-9]>/i', $content, $matches);
if (!empty($matches[0]) && count($matches[0]) > 0)
foreach ($matches[0] as $headertag) {
preg_match('/>(.*?)<\/(h[0-9])>/i', $headertag, $smatches);
if (!empty($smatches[1]) && count($smatches[1]) > 0) {
$headerid = strip_tags($headertag);
$headerid = trim(strtolower(preg_replace('/[^a-z0-9]/i', '', $headerid)));
$smatches[2] = strtolower($smatches[2]);
$header_depth = intval(trim(str_ireplace('h', '', $smatches[2])));
while (in_array($headerid, $had_headers)) {
$headerid .= '1';
}
$had_headers[] = $headerid;
$content = str_replace($headertag, '<'. $smatches[2] . ' id="' . htmlentities($headerid) . '">' . $smatches[1] . '</' . $smatches[2] . '>', $content);
}
}
$matches = null;
$smatches = null;
$toc_html = '<ol id="toc">' . "\n";
$old_depth = 0;
$hadfirst = false;
preg_match_all('/<h[0-9].*?>.*?<\/h[0-9]>/i', $content, $matches);
if (!empty($matches[0]) && count($matches[0]) > 0)
for ($i=0; $i < count($matches[0]); $i++) {
$headertag = $matches[0][$i];
preg_match('/<h[0-9][^>]*?id="(.*?)".*?>(.*?)<\/(h[0-9])>/i', $headertag, $smatches);
if (!empty($smatches[1]) && count($smatches[1]) > 0) {
$headerid = trim($smatches[1]);
$header_depth = intval(trim(str_ireplace('h', '', $smatches[3]))) - 1;
// don't take heigher than h3 in TOC
if ($header_depth > 2)
continue;
if ($header_depth < $old_depth) {
$diff = $old_depth - $header_depth; //if going multiple levels up
$toc_html .= '</li>'.str_repeat('</ol></li>', $diff);
} elseif ($header_depth > $old_depth) {
$toc_html .= '<ol>';
} else {
$toc_html .= ($hadfirst) ? '</li>' : null;
}
$toc_html .= '<li>' . htmlentities(trim(strip_tags($smatches[2]))) . '';
$old_depth = $header_depth;
$hadfirst = true;
}
}
$toc_html .= str_repeat('</li></ol>', ($old_depth + 1));

Answer 3

You may have solved this already? I've not used dompdf, but i did a similar thing in Zend_Pdf: I made a blank page for the table of contents and then went on creating all the other later pages, keeping an array of page_number => title. At the end I went back and updated the contents page using the reference saved earlier...

Answer 4

As an extension to vicky shrestha's answer here is what I have for a table of contents that expands more than a single page.
36 is just an arbitrary number of items that fit in the design.
foreach ($GLOBALS['entity_page'] as $key => $val) {
$GLOBALS["entity_y"] = 88;
$GLOBALS["entity_val"] = 0;
$GLOBALS["entity_per_page"] = 36;
if($val) {
$canvas->page_script('
if(isset($GLOBALS["entity_page"][$GLOBALS["entity_val"]])) {
if($PAGE_NUM == $GLOBALS["entity_page_number"]){
$x = 505;
$y = $GLOBALS["entity_y"];
$font = $fontMetrics->get_font("Open Sans", "Helvetica Neue", "Helvetica, Arial, sans-serif");
$pdf->text($x, $y, $GLOBALS["entity_page"][$GLOBALS["entity_val"]]."", $font, 7, array(0, 0, 0, 1));
$GLOBALS["entity_y"] = $GLOBALS["entity_y"] + 19;
$GLOBALS["entity_val"] = $GLOBALS["entity_val"] + 1;
if (($GLOBALS["entity_val"] + 1) % $GLOBALS["entity_per_page"] == 0 ) {
$GLOBALS["entity_page_number"] = $GLOBALS["entity_page_number"] + 1;
$GLOBALS["entity_y"] = 31;
}
}
}');
}
}
the isset is important as for some reason the foreach will loop an additional time and you will get an out of bounds exception thrown at the end.

Answer 5

My idea to create TOC in pdf is following:
generate pdf where title of page would contain random token instead of name
parse generated pdf (using Smalot\PdfParser) and find page number which contains the token
generate the pdf where replace tokens with real name and found page number.
Example:
$html = "<html>
<div class='toc'>
<a>Article1 ..... {article_1_num}</a>
</div>
...
<div class='article'>
<div>{article_1_title}</div>
<div>Article content</div>
</div>
</html>";
//prepare variables to replace in template with random token
$vars = ["{article_1_title}"=>'676TGZGHVGFTRR655R66TTFTF', "{article_1_num}"=>0];
//genetate pdf
$options = new Options();
$options->set('defaultFont', 'Courier');
$options->set('isRemoteEnabled', TRUE);
$options->set('debugKeepTemp', TRUE);
$options->set('isPhpEnabled', TRUE);
$options->set('isHtml5ParserEnabled', true);
$dompdf = new Dompdf($options);
//load html with variables replaced
$dompdf->loadHtml(strtr($html, $vars));
$dompdf->setPaper('A4');
#$dompdf->render();
//create tamporary file
$temp = tempnam(sys_get_temp_dir(), 'prefix');
$output = $dompdf->output();
//save to temporary file
file_put_contents($temp, $output);
// parse pdf
$parser = new \Smalot\PdfParser\Parser();
$pdf = $parser->parseFile($temp);
$pages = $pdf->getPages();
$pageNum = 0;
//loop the pages and find the one with the token
foreach ($pages as $k=>$page) {
if(strpos($page,'676TGZGHVGFTRR655R66TTFTF') !== false){
$pageNum = $k+1;
break;
}
}
// remove temp file
unlink($temp);
//prepare variables with real values to replace in template
$vars = ["{article_1_title}"=>'Article no. 1', "{article_1_num}"=>$pageNum];
// generate pdf and stream it to user
$options = new Options();
$options->set('defaultFont', 'Courier');
$options->set('isRemoteEnabled', TRUE);
$options->set('debugKeepTemp', TRUE);
$options->set('isPhpEnabled', TRUE);
$options->set('isHtml5ParserEnabled', true);
$dompdf = new Dompdf($options);
//load html with variables replaced
$dompdf->loadHtml(strtr($html, $vars));
$dompdf->setPaper('A4');
#$dompdf->render();
$dompdf->stream("pdf.pdf", array('Attachment' => 0));