Thursday, October 05, 2006

okular features #2

Word processor-like text selection



Okular provides real text selection, that is you can select from half one line to the half of a line below it and get the lines in between completely selected. This fixes the problem kpdf had with the old rectangle based selection.

9 comments:

Anonymous said...

But can we still use the old block based copy selection as well?!

Anonymous said...

yes, the old rectangular selection is preserved and working as before.

Anonymous said...

Is it a normal text selection or does it work like in Adobe's Pdf Reader?
I.e., does the mouse pointer automatically become a selection marker when you stay over a bit of text for a certain amount of time, and can you also make a selection and move the start and end markers later?
The first one should be useful for its automation, the second one is imho always useful because one often changes the final selection afterwards.

Anonymous said...

Another idea: Integrate an OCR-engine so that scanned pages get converted to text when selecting. This would be a killer feature.

Janne said...

Can I make a suggestion :)? Notice the red box in the sidebar that displays the area being displayed. In the screenshot we have two boxes, one bigger and one smaller, divided between two pages. What if cases like this (where the viewed area is spread across two pages), the top "box" would lose it's bottom line, ant the bottom-box would lose it's topmost line. That way we wouldn't have two separate boxes, but one box, that just happens to be split across two pages.

I think it would look a lot nicer :).

Anonymous said...

Why is the anti-aliasing quality of the PDF worse than the quality of the menu font? Does Okular not support hinting or subpixel rendering?

Albert Astals Cid said...

@janne:

That's not possible as we support more than 1 column of pages in the view so there is no way to show a rectable that marks all the visible parts of the pages as you can be viewing some parts of 4 pages for example.

Anonymous said...

@anonymous (2nd): the actual text selection is another mode (just like "zoom mode", for example). Maybe we could implement the starting of the text selection after a pair of seconds of non-moving mouse, but we're not looked into it yet. About the possibility to move start and end anchor, we should first talk with our usability master.

@anonymous (3rd): that it's already in our TODO ;)

Janne said...

"That's not possible as we support more than 1 column of pages in the view so there is no way to show a rectable that marks all the visible parts of the pages as you can be viewing some parts of 4 pages for example."

I don't think that would be an issue, really. Think of it like this: the red rectangle would be red only in the white areas (in other words, the actual pages) in the sidebar. In the area between the pages, it would be transparent. This way you would have just one box (with some parts being transparent), instead of having several small boxes scattered around the sidebar. The end-result would look neater.

So if you had two colums (for example), and you would have four pages visible, you would see the four corners of the box in the sidebar, instead of having four small boxes in there.

I think this is doable :).