Re: GMountOperation issues

[David Zeuthen <david fubar dk>]

On Mon, 2009-02-16 at 10:48 -0500, Matthias Clasen wrote:
> >>  Introduce ::cancelled signal
> >>   on GMountOperation and add suitable docs.
> >
> > Sounds reasonable. i'd consider adding that signal if Alex agrees that
> > it is a good idea and a patch shows up before Monday night...
> 
> Alex was in favour and proposed to name it "aborted" instead. 

Sounds good to me.

> Here is
> a patch. Does that look like what you need ?

Thanks for doing this

+   * Implementations of GMountOperation should handle this signal
+   * by dismissing open password dialogs, taking possible further
+   * action to inform the user about what happened, and return
+   * #G_IO_ERROR_FAILED_HANDLED.

GMountOperation derived classes don't really return anything (the
backends do however). Suggest to replace it with

+   * Implementations of GMountOperation should handle this signal
+   * by dismissing open password dialogs.

and then a patch for the g_volume_mount() docs saying something along
the lines of 

 "#GVolume implementations must emit the #GMountOperation::cancelled
  signal before completing a mount operation that is aborted while
  awaiting input from the user through a #GMountOperation instance."

     David