I have two issues I hope you can help me with. I am writing my dissertation and these two things are getting in my way. They only happen on very large PDF documents I am using (other people’s dissertations). Here are the issues:
White boxes take up the entire screen when I scroll 1/3 of the way through a document. I can turn to single page view, but then I can’t search and move quickly through the document
When I search for a keyword, it will act like the word does not occur in the document even when I am staring at the word. After a few minutes it will be able to find it. I think this may simply be lag in the recognition system, but I want to see if there is a way to make it quicker or eliminate this problem
This is an OS X 10.9 bug in the PDF engine that we use (and thus unfortunately we cannot fix it, so please report to Apple if you have the time, and make sure to mention it is widespread through multiple programs using PDF Kit). If you experience this problem with a PDF, the workaround is to switch display mode to single page, rather than continuous, with the View/Media/PDF Display/Page Breaks menu command.
Could you clarify what you mean by “keyword” here? Strictly speaking, Keywords (as in the feature that lets you assign keywords as meta-data to documents) will not be in the main text editor at all. If you are setting your search criteria to “Keywords” and then looking for a result in the main editor, you won’t find it. You need to set your search criteria to “Text” or “All”.
That said, your description with lag doesn’t match the above, so if you just mean “keyword” in the general sense of the word, then a “few minutes” of time to search is totally abnormal. That should be pretty much instantaneous even in very large projects with hundreds of thousands of words.
Perhaps there is some content in the project that is slowing down the search engine though. Can you confirm that this problem exists if you create a new blank project as a test (just put it somewhere temporary like the desktop), and drag and drop one single file from your main working project into this test project, then search for the same word—faster?
If so, it could be those PDFs you mentioned. They will be indexed, and if they are adding tons of text to the search index perhaps they are slowing things down. That would be unusual however. We have people that store gigabytes worth of PDFs in their projects and search speed is not a problem. Not all PDFs are created equal however. If you discover it is a problem with that PDF, would it be something you can send to use for evaluation?
Regarding the white box issue, it seems to be that Apple’s PDF frameworks (which Scrivener uses) have these drawing issues when running in 32-bit, a bug introduced in Mavericks. The problem doesn’t seem to arise when the PDF frameworks are running in 64-bit. The problem is that Scrivener is currently a 32-bit program. The problem should disappear when Scrivener becomes 64-bit (which it will be doing later this year, I hope).