So I just noticed this little perversion.
I'm looking at a PDF, which is an exported version of a spreadsheet. It's some boring requirements thing, lots of text, no actual formulas. So it doesn't matter that I'm reading it on PDF form, right?
I search for a multi-word term. Nothing. What? No, that must be used in here, a dozen times.
I search for one word, and find an occurrence. Yeah, no shit, it's there. Okay, select and copy the text, and search on that. Finds all of them.
WTF?!
Start poking at the search string. The letters are normal letters. The spaces are... breaking it?! What?
In fact, Excel, in its infinite wisdom, has chosen to use non breaking spaces, " ", \u00a0, in the export. Making text search nearly useless.
Thanks Microsoft.
And no, it's not in the source document, those are normal spaces (\u0020).
Just FYI, in case you encounter this, and in case you needed another occasion to vent about M$...

So on a related note, I wonder if there is a way to add a plug-in to Windows itself, to add more functionality to OS widgets. Namely, for the present case -- more features for text inputs and edit boxes, like viewing and editing direct Unicode code points, and rich-text formatting elements. Maybe this already exists? Maybe as an Office or $PDF-viewer plugin rather than the OS? That'd be a bit less useful, but might help illuminate cases like the present one.
Tim