First of all thank you for a decent PDF viewer! I have just started using Okular mostly because it has a 'Trim To Selection' feature. Often my use case for this... would be when I open an A4 PDF which has a 4x6" shipping label in the corner and a bunch of instructions elsewhere on the page. In these circumstances I'm expecting to be able to 'Trim To Selection' to the shipping label... and then go to print and choose my Dymo LabelWriter 4XL (which prints to 4x6" shipping labels) and bingo! Except, no ... it prints the entire page squished into that label... doh! This would seem to be counter intuitive really... most users would expect that it would print what you've trimmed to. Heck if I put into Google... okular "trim to selection" print Google AI / Gemini answers... "To print only the selected area in Okular using the "trim to selection" feature, first select the desired area using the "Trim View" -> "Trim to Selection" tool. Then, navigate to "File" -> "Print", and Okular will automatically use the trimmed selection for printing. " Which seems to not be true? Typical of Gemini really. But even so, if both a human and an AI think it should be obvious what should happen... I think giving this experience the label of being counterproductive is warranted. Any chance we can change this behavior?
It is on the View menu, so affects what you view. There was never any intention of it affecting printing.
Consider it a feature request? Maybe present an option to print only what you've trimmed to, in some other fashion or menu? Essentially it is a similar function (from Adobe Acrobat Reader) that I was hoping for... https://helpx.adobe.com/uk/acrobat/kb/print-portion-page-acrobat-reader.html Perhaps if someone goes to the 'Tools' menu and 'Area Selection' then one of the options that appears on the context menu (under the 'Image' section) could be print? What **shouldn't** happen though... is that it just takes the area, rasterizes it, and then sends that to print. It would be preferred if it could keep that area as a vector image (e.g. postscript) when sending to the printer. Mostly because in my case I'd be printing things like QR/barcodes which might not be as clear if they are converted to a raster image first (as 'Copy to Clipboard' seems to do).
Implementing this without rasterization will likely need something like https://invent.kde.org/graphics/okular/-/merge_requests/411. Based on that I expect it to be reasonably straightforward.