Hello

Let's consider the problematic cases and try to make some suggestions. As I 
recall, there are several possibilities of unintentional actions:

1. In the darkroom mode, ratings etc are applied to an image under cursor, not 
to the selected or processed one. That can occasionally lead to unpredicted and 
even unnoticed results if mouse cursor have been moved to the filmstrip. 
A possible solution: apply changes to an image from the filmstrip if it is 
selected. Should it be applied to the whole number of multiple selected images 
or to the one and only under cursor - it is a question of settings. But the 
image should be already selected.

2. Focus steal from text input controls without explicit mouse click. As I 
understood from the discussion, it is mainly made for spinbuttons controls. So 
let's consider the use-case of spinbutton control. Does it need keyboard 
actions? Yes, maybe. In which case? If keyboard was used for navigation I 
guess. That is traditional tabs switch between controls within a window. 
Opposite, in the case a _mouse_ was used to select (simply hovering over the 
control), keyboard cursors are hardly used to change its value - it is much 
easier to use a scroll button from the same mouse. So if the control has 
received focus with click or keyboard, it's OK to it to hook keyboard actions. 
In the other case, it can only hook mouse actions. And focus for keyboard input 
remains for the last clicked control or at least the keyboard actions are 
translated to it.

Any thoughts? 


With respect,
Alexander Rabtchevich


> Von: "David Vincent-Jones" <david...@gmail.com>
> There are standards that most have used across numerous systems and over 
> many years.
> 
> Double click to select a word element or number and then simply type-in 
> a replacement. This input method is respected across all programs on my 
> system .. except for darktable on which it can, unfortunately, lead to 
> lost material.
> 
> As you suggested I have now trained myself to to handle dt input in a 
> different manner; it would however be preferred if a more common method 
> was available.
> 
> David
___________________________________________________________________________
darktable developer mailing list
to unsubscribe send a mail to darktable-dev+unsubscr...@lists.darktable.org

Reply via email to