Ok. Gotcha. What I'm doing is to create the image on the fly (as buffered
image), then directly drawing the image using a JPEGFactory response. So maybe
that's where it's getting hosed up. There are options to drawImage from content
read from a file but we're trying to avoid that if possible to eliminate disk
I/O as a bottleneck when dealing with multiple thousands of images.
[cid:ba407754-72d1-481d-8d9c-a271163b0497]
Maybe one of these 2 steps (likely the first one) is where this is getting
introduced. I haven't found a way to directly draw a bufferedImage of a PNG
into a PDF without writing to file first. I'm new to the PDFBox side of things
so don't have a firm grasp on all of the possibilities yet.
________________________________
From: Tilman Hausherr <thaush...@t-online.de>
Sent: Wednesday, August 2, 2023 11:27 AM
To: users@pdfbox.apache.org <users@pdfbox.apache.org>
Subject: Re: Border / Box around images and form elements with backgrounds
EXTERNAL: Do not click links or open attachments if you do not recognize the
sender.
On 02.08.2023 18:11, JJ Blodgett wrote:
Not sure what you mean about ARGB being jpeg. The examples I provided should
have been PNG.
Here's what I mean - the image in the PDF is a JPEG encoded image with a b/w
JPEG ("DCTDecode" is JPEG) encoded mask. Using JPEG is a weird idea for b/w
images, CCITT 4 is best. I'm wondering if the non-matching is because of the
weird compression.
[cid:part1.0ze00fN2.5NiM0jXY@t-online.de]
Tilman
