Fprintd is dependent on the deprecated dbus-glib, also this doesn't provide
various features we can take advantage of, like the ones for async
authentication mechanism.
So, remove all the dbus-glib dependencies and simplify the code, but without
any further refactor, and keeping everything as it used to work, while this
will give room for further improvements in subsequent commits.
Internally, we just use dbus-codegen to generate the skeletons, and we
use the generated FprintdDBusManager with composition, while we
implement the device skeleton interface in FprintDevice, so that we
don't have to use it as a proxy, and keep being closer to what it used
to be with dbus-glib.
Fixes: #61
In case of early return we may not free them consistently, while this is not
a big problem in a main function, is better to have a cleaner management,
and we did get valgrind reports.
When coverage is enabled fprintd test won't generate any .gcda file and so
apparently no data, this happens because gcov doesn't handle properly the
process termination when SIGTERM is used, and so when in fprintd.py we
terminate the process no coverage data is reported.
To avoid this, quit the main loop cleanly on SIGTERM, so that we will exit
from the main function cleanly, making libc to perform a gcov flush when we
exit the program.
There's no need to declare it as extern in the header as it is already
declared in the source files where it's used.
Fixes:
../src/device.c:51:25: error: redundant redeclaration of ‘fprintd_dbus_conn’ [-Werror=redundant-decls]
51 | extern DBusGConnection *fprintd_dbus_conn;
| ^~~~~~~~~~~~~~~~~
In file included from ../src/device.c:34:
../src/fprintd.h:29:25: note: previous declaration of ‘fprintd_dbus_conn’ was here
29 | extern DBusGConnection *fprintd_dbus_conn;
| ^~~~~~~~~~~~~~~~~
Fix linking error as the "store" global variable gets redeclared in
each C file that includes the header. Move the actual declaration to
main.c.
Fixes:
/usr/bin/ld: ./.libs/libfprintd.a(device.o):/builds/libfprint/fprintd/src/storage.h:51: multiple definition of `store'; main.o:/builds/libfprint/fprintd/src/storage.h:51: first defined here
When creating the FprintManager object the devices will be enumerated.
This operation calls the mainloop recursively. We do not want to receive
any client requests before the initial enumeration has happened. Because
of this, move the registration of the common name to happen after the
enumeration has finished.
For sensors with internal storage we may want to garbage collect prints.
Adding this API means we can list all local prints, allowing us to find
out whether there are prints on the device's storage with no
corresponding print on the host.
the dbus activation machinery depends on daemons taking a name on
the bus to complete activation without timeouts.
fprintd fails prematurely if there is USB bus (as found in some
virtual machine setups).
This commit moves the bus-name-acquisition code to happen before
the fail-from-no-usb-bus code to keep callers from timing out
when activating fprintd.
https://bugs.freedesktop.org/show_bug.cgi?id=57025
Make all of FPrintManager's struct members private,
warn of errors using a _get_error() function, as object
creation can never fail.
The only error possible shouldn't really be an error though,
as it only means that no devices will be enumerated, but it
could tell us about newly plugged devices instead.
This will be needed to get better error messages to
users of the D-Bus API. We use glib's builtin gettext
support. No strings marked as translatable yet, we'll
need to mark those user-visible strings carefully and
review them.
With this patch, and the D-Bus activation patches posted earlier the
fprintd daemon is started automatically when needed.
We should have a way to exit it nicely when there are no users
anymore as well...
