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

Re: r23706 - in trunk/packages/kbd-chooser: . debian



Joey Hess wrote:
> keyboard type at: present: false
> keyboard type usb: present: false
 
Hmm, ok, why does it think usb is not present? This is obviously the
reason behind displaying a random hunk of memory, since buf is not
initialised. I don't understand the hack of setting present to false for
a non-usb keyboard that you added in the patch, but notice that indeed
if the arch is not i386 or powerpc, the new code never sets it to true.

I'll bet that Frans can fix in now in 5 minutes, so I'll leave it to
him. :-)

(Just a guess:

Index: usb-kbd.c
===================================================================
--- usb-kbd.c	(revision 23706)
+++ usb-kbd.c	(working copy)
@@ -68,7 +68,7 @@
 				p->present = TRUE;
 			} else {
 				di_debug ("non-Apple USB keyboard detected");
-				p->present = FALSE;
+				p->present = TRUE;
 #if defined(__i386__) && defined (AT_KBD)
 				di_debug ("Forcing keymap list to AT (i386)");
 				p->name = "at";           // Force installer to show AT keymaps
@@ -80,6 +80,8 @@
 					p->name = "at";   // Force installer to show AT keymaps
 					p->present = TRUE;
 				}
+				else
+					 p->present = FALSE;
 #endif
 			}
 			if (strcmp(p->name,"usb") == 0)


)
-- 
see shy jo

Attachment: signature.asc
Description: Digital signature


Reply to: