Changing the version in the file header would make
a preset file not readable by older versions of the
tool. If we just append new data always at the end
of the save procedure we should have no problems
reading them with different versions, as they all
just read to a certain point and ignore the rest
of the file.
This patch implements the logic to save the presets
first to a file called file_name.tmp and appends any
extra data that would come after in a possibly present
file_name file.
Any data written by newer versions would remain in
the preset file and from now on no old version should
remove data written by newer versions.
Also since we write to a temporary file and rename
afterwards an extra feature is gained of not corrupting
the preset should we crash.
Signed-off-by: Jasmin Fazlic <superfassl@gmail.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Some variable types are referring to the field in ioctl struct, which
are actually __u32 or __u64 instead of uint32_t or uint64_t.
This inconsistency may result in the compile error.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This signifies that this code is now compatible with both Python 2 and
Python 3, as per PEP-0394[1].
[1] https://www.python.org/dev/peps/pep-0394/
Signed-off-by: Emmanuel Gil Peyrot <linkmauve@linkmauve.fr>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This feature has been added in Python 2.5 and automatically closes an
open file once the context exits.
Signed-off-by: Emmanuel Gil Peyrot <linkmauve@linkmauve.fr>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This replaces VBox and HBox with Grid (using Gtk.Orientation), HScale
with Scale, creates labels with mnemonics, set hexpand and vexpand
properly, use the correct enum container classes, use the correct getter
for size request, and finally update to the correct GLib watch function.
Signed-off-by: Emmanuel Gil Peyrot <linkmauve@linkmauve.fr>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This doesn’t work yet, we require Gtk 3.0 rather than 2.0 and the API
changed quite a lot, so this is but a preparatory patch.
This is done so that we can get rid of GTK+ 2 which has been EOL for
many years already, and to add Python 3 support because Python 2 will
very soon be EOL as well.
Signed-off-by: Emmanuel Gil Peyrot <linkmauve@linkmauve.fr>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This is the recommended coding style for all Python programs, as
specified in PEP-0008[1].
[1] https://www.python.org/dev/peps/pep-0008/
Signed-off-by: Emmanuel Gil Peyrot <linkmauve@linkmauve.fr>
Reviewed-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
__sighandler_t is a glibc internal type which doesn't exist in musl. By using
AC_USE_SYSTEM_EXTENSIONS to get _GNU_SOURCE defined, both glibc and musl
then expose sighandler_t.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The gitcompile script is used from the top-level Makefile, so it needs
to be included in the tarball.
Signed-off-by: Tanu Kaskinen <tanuk@iki.fi>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
In current implementation for a series of echoaudio drivers, some control
element set delivers information to use element values for
multi-dimensional array. Essentially, ALSA control interface of asound.h
has no common way to parse the information. As a result, the drivers
forces 'echomixer' application to parse the information by several ways.
This issue was reported by a commit 51db452df07b ('Revert "ALSA: echoaudio:
purge contradictions between dimension matrix members and total number of
members"') to Linux kernel. As a result of discussion at Linux
miniconference 2017, usage of 'dimen' member of 'struct snd_ctl_elem_info'
is going to be deprecated for future removal.
According to the above decision, this commit obsoletes usage of 'dimen'
information in 'echomixer' application. In a series of echoaudio drivers,
a control element set named as 'Channels info' includes duplicated
information of the dimension information on value array. Additionally,
some of dimension information have fixed values. This commit utilizes them
to obsolete usage of dimension information.
This patch is tested with Echo Gina 3G.
Signed-off-by: Takashi Sakamoto <o-takashi@sakamocchi.jp>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
output_tram_line() has a superfluous inline prefix and this interferes
the build with clang. Let's drop it.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
gcc6 narrowing error compiling for PowerPC archi
===
[ 51s] Cus428Midi.cc: At global scope:
[ 51s] Cus428Midi.cc:83:1: error: narrowing conversion of '-1' from 'int' to 'char' inside { } [-Wnarrowing]
[ 51s] };
===
proposed patch as suggested in
https://gcc.gnu.org/gcc-6/porting_to.html
Signed-off-by: Michel Normand <normand@linux.vnet.ibm.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
This allows for simpler setting of dock headphone, dock mic,
dock line out and dock line in.
Also improve detection a bit so that an existing dock headphone
shows up as such and not as regular headphone.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
I previously had a small python script doing the same thing,
but it depended on hda-analyzer, which always breaks when something
new is added to the codec proc file.
I got tired and rewrote it as a small C program instead, which I
hope will be a useful addition to alsa-tools.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The kernel driver has some hints you can send to it that changes
parser behaviour. This patch exposes that functionality to the user.
(This patch also includes minor fixes for documentation, GTK warnings
and whitespace.)
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
On a 3.15 based kernel, both "hdaudioC*D*" and "hwC*D*" shows up under
/sys/class/sound, and pointing to the same object. Make sure we only
match on the hwC*D* name to avoid duplicate entries.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
AIO allows to connect Analogue Extension Boards (4in/4out).
Provide those channels in hdspmixer, too.
The channel maps were copied from the driver, hence the changed casing
in the comments.
Signed-off-by: Adrian Knoth <adi@drcomp.erfurt.thur.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The standard configuration directory for PulseAudio has changed,
I think since 3.0. Adjust the directory lookup mechanism to match.
Signed-off-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
When compiled with hardening-flags enabled, ld10k1 fails to build due to
a pair of unsafe printf usages that trigger -Werror=format-security.
Signed-off-by: Jordi Mallach <jordi@debian.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
The “function” builtin is a bashism that can be simply omitted with
no behaviour change, if parentheses are supplied.
Signed-off-by: Jordi Mallach <jordi@debian.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
