I haven’t been able to come up with a good, quick search routine. The usual scenario is:
I do a database-wide search for some words or a phrase and I get a fairly long list of promising-looking documents, many of which are long PDFs. I open a PDF and can’t immediately find the expression I’m looking for. Often, if I’m not thinking too clearly, I’ll open up the words sidepane in the PDF and double-click on one of my target words, expecting to be taken to the line in the document where that word is. However, as you know, that’s not how this works: instead a new search window opens up, database-wide by default.
I can understand that this behaviour makes perfect sense for exploring the database as a whole: we find a document that interests us and use that as a launchpad to find lots of other related and interesting documents. But what’s the best way to find the place within a document where a particular word or phrase occurs?
From the list of search results, select one. For text or PDF documents the text panel view will scroll to the first occurrence of a search term, which is highlighted. For HTML or WebArchive document types the search terms are highlighted but the document is not automatically scrolled to the first occurrence.
For Phrase or for single-term searches pressing Command-G will move to the next highlighted occurrence of the search term(s).
Ah, I see. I have hitherto tended to use “any word”, which I realise now renders Command-G meaningless, so I guess rethinking my searches a phrases would be sensible. And the lack of automatic scrolling with HTML and WebArchives fooled me into thinking that DT couldn’t locate the terms.
One thing I have noticed when searching PDFs is that the search term is highlighted in grey rather than in my usual highlight colour, and sometimes it is therefore rather difficult to pick out the search term. I don’t suppose there is any way of changing this behaviour?