# Split, reduce and convert PDF to JPEG using PHP ImageMagick

During these days I’m working on a PHP project based on manipulation of PDF files uploaded by users. Documents must be truncated, splitted to images, quality reducted, etc.

Because of the project’s nature, libraries’ inclusion is problematic, so I was forced to study a solution for the issue that involves only PHP official extensions. The result is some code based on ImageMagick extension.

In this post I’ll note down what I found because I hope that it can be useful for someone in my own situation.

## Example 1: convert to low quality PDF

Let’s start with very simple example. Following lines will produce a new PDF file containing a low quality image for each page from original PDF:

 1 2 3 4  readImage('original.pdf'); $image->writeImages('poor_quality.pdf', true);  Not really so much to explain… ## Example 2: convert to low quality JPEGs This code will produce a list of low quality JPGs, one for each page from original PDF:  1 2 3 4  readImage('original.pdf');$image->writeImages('poor_quality.jpg', false); 

Only difference from previous example is in line 4: * file extension is .jpg * last parameter is false, otherwise imagemagick will output a single jpg with all pages overlapped

Output of this code will be a list of files like:

$ls poor_quality-0.jpg poor_quality-1.jpg poor_quality-2.jpg ## Example 3: convert to high quality JPEGs This code will produce a list of JPGs, like previous example, but with high quality images:  1 2 3 4 5 6  setResolution(300, 300);$image->readImage('original.pdf'); $image->setImageCompressionQuality(100);$image->writeImages('high_quality.jpg', false); 

Most important line, compared to previous example, is line 3 in which we set resolution parameters. It’s important to execute this instruction before read PDF file.

## Example 4: convert subset of pages to JPEGs (only for small PDFs)

Following code can be used to convert a subset of page to JPGs. Unfortunately it is very inefficient and can be very slow for PDF with a lot of pages. For this reason, I point it out for educational reasons only:

  1 2 3 4 5 6 7 8 9 10 11 12 13  setResolution(300,300); $image->readImage('original.pdf');$image->setImageCompressionQuality(100); $image->setBackgroundColor( new ImagickPixel( 'white' ) ); // Remove pages > 10 while ($image->setLastIterator() && $image->getIteratorIndex() > 10 ) {$image->removeImage(); } $image->writeImages( 'high_quality.jpg',false);  There also are methods like setIteratorIndex or setFirstIterator that can be useful for building various types of cycles but, as reported before, in my experience this method can be very slow. More details can be found at ImageMagick. ## Example 5: convert subset of pages to JPEGs Last example is a bit more complex because it reads and writes pages one by one. Anyway it’s also much more efficient than the previous one. It produces a sublist of high quality JPEGs also from very long PDFs in an acceptable time:   1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17  setResolution(300,300);$first_page = 0; $last_page = 10; try { for ($i = $first_page;$i < $last_page;$i ++ ) { $image->readImage('original.pdf' . '[' .$i . ']' ); $image->setImageCompressionQuality( 100 );$image->setBackgroundColor( new ImagickPixel( 'white' ) ); $image->writeImage( 'high_quality-' .$i . '.jpg' ); } } catch (ImagickException $e){ if ($e->getCode() == 1 ) { // It's OK. This exception means that there are no more pages to convert. } } 
Using [index] after PDF path, Imagemagick read and converts only the page corresponding to the index. In this way, only needed pages will be processed.

## More details

I know well that this is not a complete guide, but I hope that it can be useful for someone in my same situation. Anyway, you can merge previous example to obtain different results and find much more details at ImageMagick.

## Final thoughts

PHP ImageMagick extension is a very powerful tool but, imho, it’s not very well documented (like most PHP features). Probably, a more complete guide could allow to develop a lot of code without need to include external libraries that adds complexity, maintenance issues, etc.