How to find x and y coordinates of a text in PDF

[ccpplinux]

Hi,
Is it possible to get the value of X and Y coordinates of a text in a PDF file using this library? If yes then please send me a sample code.

Best Regards ...
Pankaj Kumar

[k00ni]

Yes. Here is an example.

[ccpplinux]

Thanks for reply. I have installed it on my server and then tested it. I am using the following code to extract X and Y coordinates of all words in a PDF file:

<?php
// Include Composer autoloader if not already done.
include 'vendor/autoload.php';
 
// Parse pdf file and build necessary objects.
$parser = new \Smalot\PdfParser\Parser();
$pdf    = $parser->parseFile('result_sheet_format_llb_1.pdf');
 
$pages = $pdf->getPages();
$page = $pages[0];

$dataTm = $page->getDataTm();

echo("<pre>");
var_dump($dataTm);
echo("</pre>");
?>

It is working and you can see its output at the URL https://glug4muz.org/php/pdfparser/parse.php

But there are two small problems that I would like to get rid of. At this moment I am using the package https://github.com/measuresforjustice/textricator to extract X and Y coordinates of each word of a PDF file. I am using this tool to extract all words and corresponding X/Y coordinates in the form of a CSV file. Then I am using importing that CSV file into a MySQL database. Then I am reading the values of X/Y coordinates using PHP script.

In case of this PHP script that is pdfparser, the X coordinate of each word is same as that received using the tool textricator but there is difference in the value of Y coordinate. For example the value of Y coordinate of the word $roll_no as per textricator is 115.216255 but as per pdfparser is 721.786 but X coordinate of the same word using both tools is 48.503. Can you please tell me why it is so?

Further pdfparser is not returning the X/Y coordinate of every distinct word. Sometimes it is combining two or more words. For example as you can see that it is returning X/Y coordinates of two words $held_month $held_year as combined at https://glug4muz.org/php/pdfparser/parse.php. Why it is not returning separate X/Y coordinates of $held_month and $held_year?

Can you please explain how these two issues can be resolved? Then I will integrate this tool in my project.

Best Regards ...
Pankaj Kumar

[ccpplinux]

parse.php.txt

[ccpplinux]

result_sheet_format_llb_1.pdf

[ericksho]

I have the same problem, I believe that using tables (I convert html to PDF) messes with y coordinate.
An example: Im searching for the tag {f{Docente}f} in this document in white font at the bottom but x, y is x: 251.367 - y: 109.779

[hohohoc]

This is what I found out, you need to do the conversion from Point to mm before you use SetXY.
Plus, to get the real y-coordinate, you need to use the page height minus with the y-coordinate.

Point to mm conversion refer to this link below:
https://stackoverflow.com/questions/34545339/the-size-of-pdf-documents-how-do-i-convert-from-millimeters-to-pixels-using-spi

$parser = new \Smalot\PdfParser\Parser();

$pdf = $parser->parseFile('document.pdf');
// .. or ...
$pdf = $parser->parseContent(file_get_contents('document.pdf'))

// get the page you want; put looping if need to read multiple pages. 
$pages = $pdf->getPages(); 

// get coordinate info of each string in the page
$dataTm = $pages->getDataTm();

// Find keyword
$keyword = "testing"; 

// To get page height 
$details = $pages->getDetails();
$page_height = $details['MediaBox'][3];

$x=0;
$y=0;

// Matching the string with keyword
foreach ($dataTm as $element){
        $pos = strpos($element[1], $keyword);
        if($pos !== false){
               
		// Convert point to mm;
                $x = ($element[0][4])*0.352777778;
                $y = ($page_height - $element[0][5])*0.352777778;
        }
}

print_r($x);
print_r($y);

[crThiago]

This is what I found out, you need to do the conversion from Point to mm before you use SetXY. Plus, to get the real y-coordinate, you need to use the page height minus with the y-coordinate.

Point to mm conversion refer to this link below: https://stackoverflow.com/questions/34545339/the-size-of-pdf-documents-how-do-i-convert-from-millimeters-to-pixels-using-spi

$parser = new \Smalot\PdfParser\Parser();

$pdf = $parser->parseFile('document.pdf');
// .. or ...
$pdf = $parser->parseContent(file_get_contents('document.pdf'))

// get the page you want; put looping if need to read multiple pages. 
$pages = $pdf->getPages(); 

// get coordinate info of each string in the page
$dataTm = $pages->getDataTm();

// Find keyword
$keyword = "testing"; 

// To get page height 
$details = $pages->getDetails();
$page_height = $details['MediaBox'][3];

$x=0;
$y=0;

// Matching the string with keyword
foreach ($dataTm as $element){
        $pos = strpos($element[1], $keyword);
        if($pos !== false){
               
		// Convert point to mm;
                $x = ($element[0][4])*0.352777778;
                $y = ($page_height - $element[0][5])*0.352777778;
        }
}

print_r($x);
print_r($y);

This solution work to me