Hello, I've tweaked poppler a little bit, so it won't load all the pages when the document is created, but when they are actually needed (btw if I submit this kind of patch will it be accepted, there are no major changes?). That's because I need to load large files from the web, so I load the files in chunks when data is actually needed. The problem is with links. The function that finds a page from its object number just compares all pages, assuming they are already loaded. Is there a way to know the page number the link leads to without the target page actually being loaded? Thank you, Tosho