Pulling in image from a resource? about pdf-api6

[dwarring]

This PDF has image resources, rather than inline images. If every page has an image resource named FFX1, then to get at the raw image data:

use PDF::API6;
use PDF::XObject::Image;

my PDF::API6 $pdf .= open: "test.pdf";

my $n := $pdf.page-count;
my PDF::XObject::Image @images = (1 .. $n).map: -> $page-num {
    $pdf.page($page-num)<Resources><XObject><FXX1>;
}

with @images.head -> $img {
    my blob8 $raw-compressed = $img.encoded.encode('latin-1');
    my blob8 $raw-uncompressed = $img.decoded;
    # partially implemented. probably returns Nil
    note (try $img.to-png1).WHAT.raku;
}

However, a PDF may contain several image formats, and the to-png is only partially implemented for some image formats. It will most likely return Nil rather than an object of PDF::Content::Image::PNG. It could be that image magick, or other image manipulation libraries can load the raw compressed or uncompressed data.

Reading of images from PDF files could, in general, do with a lot more work. I'd been thing of trying to integrate MagicWand, or something similar, but would need to investigate further.

[vrurg]

Thank you! I expected the format to be anything but hoped that perhaps PDF::To::Cairo can manage it. But with the current state of affairs it looks like using something like pdftoppm would be preferable.

BTW, unrelated. A while ago I desperately wanted to intercept PDF rendering to pull out data from it. No idea how I managed to overlook PDF::API6 being capable of this, but at the moment I'm biting own elbows seeing the possibilities I would have!

[dwarring]

pdftoppm Sound reasonable.

Yeah PDF::To::Cairo is just an advanced application of rendering also available in PDF::API6. PDF::Tags::Reader is another example that renders to extract structure and text from a tagged PDF file.