Comment 34 for bug 1969315

as followup c.q. correction to #21:
After a good while without any problems, the bug is back again.
So I think it never went away but just wasn't triggered for a long period of time.

In my logs, this is repeated over and over again:

  invalid unclassed pointer in cast to 'GtkTreeModel'

  gtk_tree_model_iter_nth_child: assertion 'GTK_IS_TREE_MODEL (tree_model)' failed
  gtk_widget_set_visible: assertion 'GTK_IS_WIDGET (widget)' failed

  invalid unclassed pointer in cast to 'GtkToggleButton'

  gtk_toggle_button_get_active: assertion 'GTK_IS_TOGGLE_BUTTON (toggle_button)' failed
  gtk_widget_set_sensitive: assertion 'GTK_IS_WIDGET (widget)' failed

  invalid unclassed pointer in cast to 'GtkLabel'

  gtk_label_get_text: assertion 'GTK_IS_LABEL (label)' failed
  gtk_label_set_text: assertion 'GTK_IS_LABEL (label)' failed

Regarding this, I am pretty convinced that the first line is the reason for the rest of the failures.

  invalid unclassed pointer in cast to 'GtkTreeModel'

This hints to a declaration of a void pointer which is handed over *uninitialized* to a GTK-object-something and therefore sometimes invalid and sometimes not.
This kind of thing can go randomly wrong (actually depending on what's in memory at that moment)
which would explain why the bug would sometimes pop up and sometimes not.

I can reasonably read and write "C" but I really can't wrap my head around any GTK code.
I swear if I could understand GTK code, I would personally wholeheartily get to the bottom of this.