Re: ATK crash

[Samuel Thibault <samuel thibault ens-lyon org>]

Sebastian Humenda, on Tue 31 Jan 2017 17:38:13 +0100, wrote:
> #0  0x00007f7524543ec2 in __memmove_avx_unaligned_erms () at 
> ../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S:396
> #1  0x00007f7524d54a2d in memcpy (__len=18446744073709551615, __src=0x7f75080f01a3, __dest=<optimized out>) 
> at /usr/include/x86_64-linux-gnu/bits/string3.h:53
> #2  0x00007f7524d54a2d in g_utf8_substring (str=str@entry=0x7f75080f0160 "Recent developments\nin the 
> Hurd\nSamuel Thibault\n2013 August 24th\n1", start_pos=start_pos@entry=67, end_pos=end_pos@entry=66) at 
> ././glib/gutf8.c:288
> #3  0x00007f7527243774 in ev_page_accessible_get_substring (text=<optimized out>, start_offset=67, 
> end_offset=<optimized out>) at ev-page-accessible.c:494
>         startOffset = 67
>         endOffset = 68
>         __func__ = "impl_GetText"

Oh, I'm sorry, I had told you on irc that it looked like an atk issue,
but ev_page_accessible_get_substring is actually part of evince itself,
which is given offsets 67 and 68, and translates them into 67 and 66,
i.e. it caps end_offset to 66, but doesn't cap start_offset to 66. So it
should be reported to evince.

> #5  0x00007f75225a24b8 in handle_other (pathstr=0x5608a4767c98 "/org/a11y/atspi/accessible/200", 
> member=<optimized out>, iface=0x5608a4767cd0 "org.a11y.atspi.Text", path=0x5608a3f8dac0, 
> message=0x5608a3f90fa0, bus=0x5608a40a8ec0) at droute.c:553
>         func = 0x7f75225ae1c0 <impl_GetText>
> #7  0x00007f751e18c6a3 in  () at /lib/x86_64-linux-gnu/libdbus-1.so.3
> #8  0x00007f751e17dd24 in dbus_connection_dispatch () at /lib/x86_64-linux-gnu/libdbus-1.so.3

Samuel