PDF has supported "fast web viewing" for over 10 years now via a feature of the language called Linearization. This feature, in conjunction with a feature of HTTP called "byte serving" is how Acrobat/Reader are able to display documents in browsers quickly.
I recommend you read up on these two technologies and simply use them - rathe than reinventing the wheel. Leonard On Oct 5, 2008, at 1:20 PM, Тодор Африкански wrote: > Hello all, > > I have this task I need some help with. I need to load large pdf > files (over 100mb) from the web, so I can not afford to wait for the > whole file to load before displaying something on the screen. I plan > to write a custom stream class that will be able to download the > files in chunks. So when poppler needs to read at the end of the > file I will download this chunk. Then, when some page needs to be > displayed, I will read the part of the file containing the page. But > I need to make poppler load the pages only when they are needed. For > example, when I display the first page, I don't want to download the > rest 1400 pages, because nobody reads 1000 pages in one sitting. So > I need to make some changes to the poppler source. Can you give me > some advice what I need to modify to make poppler read the page data > only when it is actually needed (eg displayPage), not all pages at > once. > > Thank you, > Tosho > _______________________________________________ > poppler mailing list > [email protected] > http://lists.freedesktop.org/mailman/listinfo/poppler _______________________________________________ poppler mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/poppler
