Re: Minutes of the GTK+ Team Meeting - 2010-12-14

From: Kristian Rietveld <kris gtk org>

Ok, thank you very much.

> On Wed, Dec 15, 2010 at 6:34 PM, Piñeiro <apinheiro igalia com> wrote:
>> I looked on gailtreeview code, and it is used gtk_tree_view_set_cursor
>> with "start_editing=TRUE" in one case. In the method cell_edit.
>> The ATK framework allows to define some actions in order to control
>> the accessibility objects. In this case it is possible to add actions
>> for the cells of the tree view. If it is a text cell renderer
>> editable, the action "edit" is added, described as "creates a widget in which
>> the contents of the cell can be edited".
>> Anyway this description is somewhat misleading. cell_edit basically
>> finds the cell, and use gtk_tree_view_set_cursor to give the keyboard
>> focus. And I guess that sets start_editing as TRUE in order to start
>> to edit this cell.
> That makes sense, so that function sets the cursor on the given cell
> and starts the edit operation, like you would expect from the name
> cell_edit().
>> So, after this change on start_editing, will this method work
>> properly? Could someone still using this action to start to edit on a
>> cell, and then just start to write on this cell?
> Yes, nothing has changed for editable cells, such as the text renderer
> or combo box renderer.  Only activatable cells, which are usually
> modified with a mouse click or a press on a single keyboard key (and
> do not allow input of free-form text), are affected.  Then again, it
> is highly unlikely that anybody was using gtk_tree_view_set_cursor()
> with start_editing=TRUE to toggle the state of for example a toggle
> cell renderer.
>> If this is the case, as Murray Cummings asked, what it is the purpose
>> of start_editing? It is now a superfluous parameter?
> The purpose of the start_editing argument is to indicate whether
> GtkTreeView should attempt to start editing the cell that is being
> focused.  Previously, it would also attempt to "activate" an
> activatable cell renderer, which is now no longer the case. The
> parameter is not superfluous.
> regards,
> -kris.

API (apinheiro igalia com)

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]