Package: pinentry / 1.1.0-3

Metadata

Package Version Patches format
pinentry 1.1.0-3 3.0 (quilt)

Patch series

view the series file
Patch File delta Description
0001 avoid self describing as a beta.patch | (download)

autogen.sh | 4 2 + 2 - 0 !
configure.ac | 4 1 + 3 - 0 !
2 files changed, 3 insertions(+), 5 deletions(-)

 avoid self-describing as a beta

Using autoreconf against the source as distributed in tarball form
invariably results in a package that thinks it's a "beta" package,
which produces the "THIS IS A DEVELOPMENT VERSION" warning string.

since we use dh_autoreconf, i need this patch to avoid producing
builds that announce themselves as DEVELOPMENT VERSIONs.

See discussion at:

 http://lists.gnupg.org/pipermail/gnupg-devel/2014-November/029065.html

0007 gtk2 When X11 input grabbing fails try again over 0..patch | (download)

gtk+-2/pinentry-gtk-2.c | 10 6 + 4 - 0 !
1 file changed, 6 insertions(+), 4 deletions(-)

 gtk2: when x11 input grabbing fails, try again over ~0.4 seconds.

* gtk+-2/pinentry-gtk-2.c (grab_keyboard): Delay 1 ms before retrying
keyboard grab, limit to about 0.2 seconds.
(grab_pointer): Delay 1 ms before retrying pointer_grab, limit to
about 0.2 seconds.

--

These changes were proposed by Vincent Bernat in
https://bugs.debian.org/851707#124

Debian-bug-id: 851707
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>

0003 Fix format string errors in fltk pinentry.patch | (download)

fltk/main.cxx | 6 3 + 3 - 0 !
1 file changed, 3 insertions(+), 3 deletions(-)

 fix format string errors in fltk pinentry

if the mesage in question had formatting escape chars, this might be
able to cause a crash (or worse) in the pinentry.

0004 gnome3 Use the default dbus timeout when checking fo.patch | (download)

gnome3/pinentry-gnome3.c | 2 1 + 1 - 0 !
1 file changed, 1 insertion(+), 1 deletion(-)

 gnome3: use the default dbus timeout when checking for screenlock

* gnome3/pinentry-gnome3.c (pe_gnome_screen_locked): use default dbus
timeout instead of 0ms.

--

At the moment, every single check to see if the screen is locked fails
immediately with a timeout, which is then silently ignored.

This is happening because the definition of the timeout argument is a
millisecond count, with -1 as the (unspecified) default, and G_MAXINT for no
timeout.

This means 0 times out immediately, and that leads to a complete inability for
pinentry-gnome3 to fallback to curses input when the screen is locked.

(Which then prevents a remote session from ever asking for a PIN, even when
the local session is locked.)

Debian-bug-id: 927105
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
(cherry picked from commit 65d2c6d5911ab04b19f953153d01116198cc4cb8)

0005 gnome3 correctly detect when no GNOME screenlock exi.patch | (download)

gnome3/pinentry-gnome3.c | 4 2 + 2 - 0 !
1 file changed, 2 insertions(+), 2 deletions(-)

 gnome3: correctly detect when no gnome screenlock exists

* gnome3/pinentry-gnome3.c (pe_gnome_screen_locked): skip
G_IO_ERROR_IS_DIRECTORY responses, not G_IO_ERROR_TIMED_OUT responses
from dbus.

--

After the previous commit, testing demonstrates that the normal
response when no screenlock is running is G_IO_ERROR_IS_DIRECTORY, not
G_IO_ERROR_TIMED_OUT.

Debian-bug-id: 927105
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
(cherry picked from commit f74c4673e6b60508012c14f7574dbc3215345210)

0006 qt Fix use of dangling pointer in QApplication.patch | (download)

qt/main.cpp | 12 10 + 2 - 0 !
1 file changed, 10 insertions(+), 2 deletions(-)

 qt: fix use of dangling pointer in qapplication

* qt/main.cpp (main): Use a new variable for argc that stays
valid.
--
The QApplication constructor takes argc as a reference and the referenced
integer has to stay alive for at least as long as the QApplication.

GnuPG-Bug-Id: T4658