On Mon, 17 Jul 2006 14:51:52 -0400
"Alexandre Leclerc" <[EMAIL PROTECTED]> wrote:
> 2006/7/17, Mattias Gaertner <[EMAIL PROTECTED]>:
> > You are a mouse user, aren't you?
>
> No, I'm using the keyboard much mode of the time. But When I do form
> design, I use the mouse. When I have too many pages opened, so that
> Alt+[1..0] are not enought, I do use the mouse to get the tab I
> desire. -- In fact I hate wasting my time with the mouse, but when my
> had is already on it...
>
> > You chose the page with the mouse. Others rarely use the mouse and
switch
> > between the pages with (shift-)ctrl+Tab. It would be quite annoying, if
for
> > moving from page 1 to 5 I get 4 history entries. Think about editing,
moving
> > to another page, editing, moving to another page, editing and my history
is
> > gone, filled with page moves. Or think about controls.pp with its
include
> > files. Just browsing through the code, by using Ctrl-Shift-Down results
in
> > jumping very often between two pages.
> >
> > If you want that feature, then add an editor option, so that people can
> > enable it, if they want.
>
> I see what you mean. I did not test to that point. This is
> unfortunate. Adding a browsing point should be more brilliant in this
> case: an history simplifier would remove duplicated group entries...
> for now only groups of two would be required.
>
> [p3][p1][p2][p1][p2][p3][p2] would become [p3][p1][p2][p3]
>
> So after writing a new entry, we could run the simplification
> algorythm. But again the problem is not that simple, because the caret
> position would be different.
>
> What about this approach? When to do about cursor positionning; should
> we ignore: that would solve the problem; because the goal is mainly
> page browsing... only the last position could be kept?
>
> [p3][p1.1][p2.1][p1.2][p2.2][p3][p2] where the .1/.2 represent caret
> position would give after simplification: [p3][p1.2][p2.2][p3][p2]
Removing 'doubles' has a drawback. When trying to understand the working of
some code, I explore the code, as it would be executed. I use the history as
stack. So, when the code calls A, B, C, B, D. The B is the same position,
but it is a different context. That the history works like a stack helps a
lot here.
The second purpose of the jump history is to undo cursor moves, that would
be a lot of work to undo. A find declaration jumps to another page, line and
column. You can not undo that with a few keys. But a page jump from page 2
to 3 can be easily undone with one key. So, here I don't see the gain.
What about his:
Add a jump point everytime the user switches to a page far away. For example
switching from 2 to 3 does not add a jump, but a jump from 1 to 3 does.
Mattias
_________________________________________________________________
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject
archives at http://www.lazarus.freepascal.org/mailarchives
