Re: [Denemo-devel] Selections, Copy, Cut and Paste - Bugs and Comments.

[Richard Shann]

On Fri, 2009-10-09 at 11:01 +0200, Nils wrote:
> On little thing: 
> 
> After pasting the cursor stays at at the position where you gave the 
> paste-command. GTK standard behaviour is that the cursor jumps to the end of 
> the pasted part. 
> 
Thinking about it, this can probably be scripted.
The script inserts a marker (d-DirectivePutStandalone-display
"RehearsalMark" "temp") then (d-CursorLeft) (d-Paste) (d-NextBookmark)
and (d-Delete) to delete the bookmark
There would be fiddly details to do with the bookmark system, it needs
to be a specially named bookmark otherwise it might find bookmarks in
the pasted text. Of course, such desirable behavior should be built in.

Richard




> Nils
> 
> 
> 
> 
> On Wed, 07 Oct 2009 21:37:27 +0100
> Richard Shann <address@hidden> wrote:
> 
> > On Wed, 2009-10-07 at 21:17 +0200, Nils wrote:
> > > More on this topic:
> > > 
> > >  > > 4) "Delete" on a Selection does not work. Cut does, but it 
> > > unfortunatly copies, too :) (And there is still Backspace. Don't know 
> > > about it)
> > > > There is no RemoveSelection function. Could be scripted
> > > 
> > > See my new mail for "Apply to Selection" for further plans.
> > > 
> > > > > 2) Pasting more than one measure (Multi- or single staff selection 
> > > > > does not matter)  into an empty measure  does not fill the empty 
> > > > > measure but leaves it empty and creates new ones behind it. Shouldn't 
> > > > > it fill the empty measure first?
> > > > > 
> > > > ? well it does now; note if you are pasting into an empty measure, then
> > > > ensure you have enough empty measures for what you are pasting (unless
> > > > you want overfull measures)
> > > 
> > > Don't you agree that it would be better if the new measures would be 
> > > created automatically? The most common task, I can imagine, is that 
> > > someone has an empty measure (mabye as a placeholder) but wants to paste 
> > > a whole part (>1 measure) in it. 
> > I think it would be better if it always created enough measures for the
> > pasted music, which it does if you are not pasting into an empty
> > measure. The only reason it doesn't do that is that I was struggling to
> > get it to do anything reasonable. There was one major structural error
> > and numerous other gotchas in there. Enough is enough.
> > 
> > I have not used paste much, but when I have it has been to paste some
> > notes into an empty measure - in that case I would not want the empty
> > measure to survive. Occasionally I have pasted two or three measures in,
> > by opening up space for them. With the new implementation I could just
> > paste without having any empty measure - I haven't tried it yet (except
> > testing this work).
> > 
> > 
> > > a)Like said in IRC: Dragging the mouse to select should scroll too, so 
> > > that you can select more than the current visible measures. 
> > yes, I would like that.
> > > 
> > > b)Is it possible to make cursor movement deselect? 
> > you would need to have the dragging causing scrolling first, else how
> > would you extend the selection?
> > > I know Richard does not like it so we could make it optional. But I know 
> > > many people expect programs to behave this way.
> > > 
> > > c) Copying (complete) empty measures works, which is good.
> > I don't think you can copy empty measures generally, I think what you
> > have seen is perhaps that it has inserted enough empty measures on the
> > assumption that they are all non-empty, and that leaves some empties at
> > the end.
> > >  But if you have an (non-notation-)object in one of those empty m. it 
> > > gets buggy. 
> > > Lets say you have 4 measures, full, empty, full, empty.
> > > And now you put a bookmark/rehearsel mark in one of those empty measures 
> > > and copy/paste it: The rehearsel mark will jump to the last non-empty 
> > > measure and all empty measures will gather at the end of the pasted area.
> > > 
> > > My personal point of view is to make empty measures more robust, but its 
> > > also possible to just forbid them. Whatever the solution is, this is a 
> > > showstopper bug.
> > I don't think it is a showstopper, you never lose any data; it would be
> > nice to be able to ensure that there is a whole measure rest object
> > there which gets eaten away when notes are added or something, but that
> > would be an undertaking of some magnitude. Meanwhile it would be
> > possible to fill all empty measures with whole measure rests
> > (d-WholeMeasureRest or whatever it is called) before cutting and pasting
> > and, indeed restore them as empty measues afterwards.
> > > 
> > > d) As a sidenote: It would be really great if successive empty measures 
> > > would be gathered as multi-measure rests as lilypond output. This way we 
> > > can keep the timeline and vertical arranging in Denemo intact but also 
> > > save space in printout. Maybe its possible to define a threshold in 
> > > Denemo preferences, maybe 3. And if there are more empty measures than 
> > > (d-GetTreshold) in a row they are converted to multi-measure-rests.
> > > 
> > Yes, I will want this - I have just got as far as a usable whole measure
> > rest command, I forget the syntax for multi-measure rests, so I am not
> > sure how easy it might be.
> > 
> > Richard
> > 
> > > Nils
> > > 
> > > 
> > 
> > 
> > 
> > _______________________________________________
> > Denemo-devel mailing list
> > address@hidden
> > http://lists.gnu.org/mailman/listinfo/denemo-devel
> > 
> 
>