2010-01-30 23:24:23 +00:00
|
|
|
/*
|
|
|
|
* This file is part of MPlayer.
|
|
|
|
*
|
|
|
|
* MPlayer is free software; you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License as published by
|
|
|
|
* the Free Software Foundation; either version 2 of the License, or
|
|
|
|
* (at your option) any later version.
|
|
|
|
*
|
|
|
|
* MPlayer is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License along
|
|
|
|
* with MPlayer; if not, write to the Free Software Foundation, Inc.,
|
|
|
|
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
|
|
|
|
*/
|
2006-04-24 19:20:04 +00:00
|
|
|
|
|
|
|
/// \file
|
|
|
|
/// \ingroup Config
|
|
|
|
|
2002-11-12 01:56:42 +00:00
|
|
|
#include "config.h"
|
|
|
|
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <errno.h>
|
|
|
|
#include <string.h>
|
|
|
|
#include <assert.h>
|
2011-07-27 17:59:44 +00:00
|
|
|
#include <stdbool.h>
|
|
|
|
|
|
|
|
#include "talloc.h"
|
2002-11-12 01:56:42 +00:00
|
|
|
|
|
|
|
#include "m_config.h"
|
2012-11-09 00:06:43 +00:00
|
|
|
#include "core/m_option.h"
|
|
|
|
#include "core/mp_msg.h"
|
2002-11-12 01:56:42 +00:00
|
|
|
|
2013-07-31 18:43:31 +00:00
|
|
|
// Profiles allow to predefine some sets of options that can then
|
|
|
|
// be applied later on with the internal -profile option.
|
2006-01-24 11:16:13 +00:00
|
|
|
#define MAX_PROFILE_DEPTH 20
|
|
|
|
|
2013-07-31 18:43:31 +00:00
|
|
|
struct m_profile {
|
|
|
|
struct m_profile *next;
|
|
|
|
char *name;
|
|
|
|
char *desc;
|
|
|
|
int num_opts;
|
|
|
|
// Option/value pair array.
|
|
|
|
char **opts;
|
|
|
|
};
|
|
|
|
|
2013-07-31 18:44:16 +00:00
|
|
|
// In the file local case, this contains the old global value.
|
|
|
|
struct m_opt_backup {
|
|
|
|
struct m_opt_backup *next;
|
|
|
|
struct m_config_option *co;
|
|
|
|
void *backup;
|
|
|
|
};
|
|
|
|
|
2012-05-07 23:10:39 +00:00
|
|
|
static int parse_include(struct m_config *config, struct bstr param, bool set)
|
|
|
|
{
|
|
|
|
if (param.len == 0)
|
|
|
|
return M_OPT_MISSING_PARAM;
|
|
|
|
if (!set)
|
|
|
|
return 1;
|
|
|
|
char *filename = bstrdup0(NULL, param);
|
|
|
|
config->includefunc(config, filename);
|
|
|
|
talloc_free(filename);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int parse_profile(struct m_config *config, const struct m_option *opt,
|
|
|
|
struct bstr name, struct bstr param, bool set)
|
2010-09-27 12:10:59 +00:00
|
|
|
{
|
2011-07-28 08:07:47 +00:00
|
|
|
if (!bstrcmp0(param, "help")) {
|
2011-07-05 19:31:44 +00:00
|
|
|
struct m_profile *p;
|
2010-09-27 12:10:59 +00:00
|
|
|
if (!config->profiles) {
|
2011-07-05 19:31:44 +00:00
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_INFO,
|
|
|
|
"No profiles have been defined.\n");
|
|
|
|
return M_OPT_EXIT - 1;
|
2010-09-27 12:10:59 +00:00
|
|
|
}
|
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_INFO, "Available profiles:\n");
|
|
|
|
for (p = config->profiles; p; p = p->next)
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, "\t%s\t%s\n", p->name,
|
|
|
|
p->desc ? p->desc : "");
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, "\n");
|
2011-07-05 19:31:44 +00:00
|
|
|
return M_OPT_EXIT - 1;
|
2010-09-27 12:10:59 +00:00
|
|
|
}
|
2006-01-24 11:16:13 +00:00
|
|
|
|
2012-05-07 23:10:39 +00:00
|
|
|
char **list = NULL;
|
2012-08-05 21:34:28 +00:00
|
|
|
int r = m_option_type_string_list.parse(opt, name, param, &list);
|
2010-09-27 12:10:59 +00:00
|
|
|
if (r < 0)
|
|
|
|
return r;
|
|
|
|
if (!list || !list[0])
|
|
|
|
return M_OPT_INVALID;
|
2012-05-07 23:10:39 +00:00
|
|
|
for (int i = 0; list[i]; i++) {
|
2013-07-27 19:26:00 +00:00
|
|
|
struct m_profile *p = m_config_get_profile0(config, list[i]);
|
2012-05-07 23:10:39 +00:00
|
|
|
if (!p) {
|
2010-09-27 12:10:59 +00:00
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_WARN, "Unknown profile '%s'.\n",
|
|
|
|
list[i]);
|
|
|
|
r = M_OPT_INVALID;
|
2012-05-07 23:10:39 +00:00
|
|
|
} else if (set)
|
|
|
|
m_config_set_profile(config, p);
|
2010-09-27 12:10:59 +00:00
|
|
|
}
|
|
|
|
m_option_free(opt, &list);
|
2012-05-07 23:10:39 +00:00
|
|
|
return r;
|
2010-09-27 12:10:59 +00:00
|
|
|
}
|
2006-01-24 11:16:13 +00:00
|
|
|
|
2013-07-27 19:26:00 +00:00
|
|
|
static int show_profile(struct m_config *config, bstr param)
|
2010-09-27 12:10:59 +00:00
|
|
|
{
|
2011-07-05 19:31:44 +00:00
|
|
|
struct m_profile *p;
|
2010-09-27 12:10:59 +00:00
|
|
|
int i, j;
|
2013-07-27 19:26:00 +00:00
|
|
|
if (!param.len)
|
2010-09-27 12:10:59 +00:00
|
|
|
return M_OPT_MISSING_PARAM;
|
|
|
|
if (!(p = m_config_get_profile(config, param))) {
|
2013-07-27 19:26:00 +00:00
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_ERR, "Unknown profile '%.*s'.\n",
|
|
|
|
BSTR_P(param));
|
2010-09-27 12:10:59 +00:00
|
|
|
return M_OPT_EXIT - 1;
|
|
|
|
}
|
|
|
|
if (!config->profile_depth)
|
2013-07-27 19:26:00 +00:00
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_INFO, "Profile %s: %s\n", p->name,
|
2010-09-27 12:10:59 +00:00
|
|
|
p->desc ? p->desc : "");
|
|
|
|
config->profile_depth++;
|
|
|
|
for (i = 0; i < p->num_opts; i++) {
|
|
|
|
char spc[config->profile_depth + 1];
|
|
|
|
for (j = 0; j < config->profile_depth; j++)
|
|
|
|
spc[j] = ' ';
|
|
|
|
spc[config->profile_depth] = '\0';
|
|
|
|
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, "%s%s=%s\n", spc,
|
|
|
|
p->opts[2 * i], p->opts[2 * i + 1]);
|
|
|
|
|
|
|
|
if (config->profile_depth < MAX_PROFILE_DEPTH
|
|
|
|
&& !strcmp(p->opts[2*i], "profile")) {
|
|
|
|
char *e, *list = p->opts[2 * i + 1];
|
|
|
|
while ((e = strchr(list, ','))) {
|
2011-07-05 19:31:44 +00:00
|
|
|
int l = e - list;
|
2010-09-27 12:10:59 +00:00
|
|
|
char tmp[l+1];
|
|
|
|
if (!l)
|
|
|
|
continue;
|
|
|
|
memcpy(tmp, list, l);
|
|
|
|
tmp[l] = '\0';
|
2013-07-27 19:26:00 +00:00
|
|
|
show_profile(config, bstr0(tmp));
|
2011-07-05 19:31:44 +00:00
|
|
|
list = e + 1;
|
2010-09-27 12:10:59 +00:00
|
|
|
}
|
|
|
|
if (list[0] != '\0')
|
2013-07-27 19:26:00 +00:00
|
|
|
show_profile(config, bstr0(list));
|
2010-09-27 12:10:59 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
config->profile_depth--;
|
|
|
|
if (!config->profile_depth)
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, "\n");
|
|
|
|
return M_OPT_EXIT - 1;
|
|
|
|
}
|
2006-01-24 11:16:13 +00:00
|
|
|
|
2013-07-27 19:26:00 +00:00
|
|
|
static int list_options(struct m_config *config)
|
2010-09-27 12:10:59 +00:00
|
|
|
{
|
|
|
|
m_config_print_option_list(config);
|
|
|
|
return M_OPT_EXIT;
|
|
|
|
}
|
2006-01-24 11:34:24 +00:00
|
|
|
|
2012-08-06 15:45:17 +00:00
|
|
|
// The memcpys are supposed to work around the struct aliasing violation,
|
|
|
|
// that would result if we just dereferenced a void** (where the void** is
|
|
|
|
// actually casted from struct some_type* ).
|
|
|
|
static void *substruct_read_ptr(void *ptr)
|
2008-03-31 03:19:29 +00:00
|
|
|
{
|
2012-08-06 15:45:17 +00:00
|
|
|
void *res;
|
|
|
|
memcpy(&res, ptr, sizeof(void*));
|
|
|
|
return res;
|
2008-03-31 03:19:29 +00:00
|
|
|
}
|
2012-08-06 15:45:17 +00:00
|
|
|
static void substruct_write_ptr(void *ptr, void *val)
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
{
|
2012-08-06 15:45:17 +00:00
|
|
|
memcpy(ptr, &val, sizeof(void*));
|
2008-03-31 03:19:29 +00:00
|
|
|
}
|
|
|
|
|
2013-07-27 19:30:00 +00:00
|
|
|
static struct m_config_option *m_config_add_option(struct m_config *config,
|
|
|
|
const char *prefix,
|
|
|
|
struct m_config_option *parent,
|
|
|
|
const struct m_option *arg);
|
2010-09-27 12:10:59 +00:00
|
|
|
|
2013-07-27 19:26:00 +00:00
|
|
|
static void add_options(struct m_config *config,
|
|
|
|
struct m_config_option *parent,
|
|
|
|
const struct m_option *defs);
|
|
|
|
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
static int config_destroy(void *p)
|
|
|
|
{
|
|
|
|
struct m_config *config = p;
|
2013-07-31 18:44:16 +00:00
|
|
|
if (config->file_local_mode)
|
|
|
|
m_config_leave_file_local(config);
|
2013-07-31 18:44:34 +00:00
|
|
|
for (struct m_config_option *copt = config->opts; copt; copt = copt->next)
|
|
|
|
m_option_free(copt->opt, copt->data);
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-07-27 19:26:00 +00:00
|
|
|
struct m_config *m_config_new(void *talloc_parent, size_t size,
|
|
|
|
const void *defaults,
|
2013-07-31 18:45:06 +00:00
|
|
|
const struct m_option *options,
|
|
|
|
const char *suboptinit)
|
2012-08-04 09:33:24 +00:00
|
|
|
{
|
2013-07-27 19:26:00 +00:00
|
|
|
struct m_config *config = talloc(talloc_parent, struct m_config);
|
2012-08-04 09:33:24 +00:00
|
|
|
talloc_set_destructor(config, config_destroy);
|
2013-07-27 19:26:00 +00:00
|
|
|
*config = (struct m_config) {
|
|
|
|
.optstruct_size = size,
|
|
|
|
.optstruct_defaults = defaults,
|
|
|
|
.options = options,
|
2013-07-31 18:45:06 +00:00
|
|
|
.suboptinit = suboptinit,
|
2013-07-27 19:26:00 +00:00
|
|
|
};
|
|
|
|
if (size) { // size==0 means a dummy object is created
|
|
|
|
config->optstruct = talloc_zero_size(config, size);
|
|
|
|
if (defaults)
|
|
|
|
memcpy(config->optstruct, defaults, size);
|
|
|
|
if (options)
|
|
|
|
add_options(config, NULL, options);
|
|
|
|
}
|
2013-07-31 18:45:06 +00:00
|
|
|
if (suboptinit) {
|
|
|
|
bstr s = bstr0(suboptinit);
|
|
|
|
int r = m_obj_parse_sub_config(bstr0("internal"), bstr0("-"), &s,
|
|
|
|
config, 0, NULL);
|
|
|
|
if (r < 0 || s.len > 0)
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_ERR, "Internal error: preset broken\n");
|
|
|
|
}
|
2012-08-04 09:33:24 +00:00
|
|
|
return config;
|
|
|
|
}
|
|
|
|
|
options: use m_config for options instead of m_struct
For some reason, both m_config and m_struct are somewhat similar, except
that m_config is much more powerful. m_config is used for VOs and some
other things, so to unify them. We plan to kick out m_struct and use
m_config for everything. (Unfortunately, m_config is also a bit more
bloated, so this commit isn't all that great, but it will allow to
reduce the option parser mess somewhat.)
This commit also switches all video filters to use the option macros.
One reason is that m_struct and m_config, even though they both use
m_option, store the offsets of the option fields differently (sigh...),
meaning the options defined for either are incompatible. It's easier to
switch everything in one go.
This commit will allow using the -vf option parser for other things,
like VOs and AOs.
2013-07-21 17:33:08 +00:00
|
|
|
struct m_config *m_config_from_obj_desc(void *talloc_parent,
|
|
|
|
struct m_obj_desc *desc)
|
|
|
|
{
|
2013-07-27 19:26:00 +00:00
|
|
|
return m_config_new(talloc_parent, desc->priv_size, desc->priv_defaults,
|
2013-07-31 18:45:06 +00:00
|
|
|
desc->options, desc->init_options);
|
options: use m_config for options instead of m_struct
For some reason, both m_config and m_struct are somewhat similar, except
that m_config is much more powerful. m_config is used for VOs and some
other things, so to unify them. We plan to kick out m_struct and use
m_config for everything. (Unfortunately, m_config is also a bit more
bloated, so this commit isn't all that great, but it will allow to
reduce the option parser mess somewhat.)
This commit also switches all video filters to use the option macros.
One reason is that m_struct and m_config, even though they both use
m_option, store the offsets of the option fields differently (sigh...),
meaning the options defined for either are incompatible. It's easier to
switch everything in one go.
This commit will allow using the -vf option parser for other things,
like VOs and AOs.
2013-07-21 17:33:08 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
int m_config_set_obj_params(struct m_config *conf, char **args)
|
|
|
|
{
|
|
|
|
for (int n = 0; args && args[n * 2 + 0]; n++) {
|
|
|
|
int r = m_config_set_option(conf, bstr0(args[n * 2 + 0]),
|
|
|
|
bstr0(args[n * 2 + 1]));
|
|
|
|
if (r < 0)
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int m_config_initialize_obj(struct m_config *config, struct m_obj_desc *desc,
|
|
|
|
void **ppriv, char ***pargs)
|
|
|
|
{
|
|
|
|
if (desc->priv_size) {
|
|
|
|
int r = m_config_set_obj_params(config, *pargs);
|
|
|
|
if (r < 0)
|
|
|
|
return r;
|
|
|
|
*ppriv = config->optstruct;
|
|
|
|
*pargs = NULL;
|
|
|
|
} else if (*pargs && !strcmp((*pargs)[0], "_oldargs_")) {
|
|
|
|
// Handle things which still use the old subopt parser
|
|
|
|
*pargs = (char **)((*pargs)[1]);
|
|
|
|
} else {
|
|
|
|
*pargs = NULL;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
static void ensure_backup(struct m_config *config, struct m_config_option *co)
|
2012-05-17 00:31:11 +00:00
|
|
|
{
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
if (!config->file_local_mode)
|
|
|
|
return;
|
|
|
|
if (co->opt->type->flags & M_OPT_TYPE_HAS_CHILD)
|
|
|
|
return;
|
2012-08-04 09:39:23 +00:00
|
|
|
if (co->opt->flags & M_OPT_GLOBAL)
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
return;
|
2013-07-31 18:44:16 +00:00
|
|
|
if (!co->data)
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
return;
|
2013-07-31 18:44:16 +00:00
|
|
|
for (struct m_opt_backup *cur = config->backup_opts; cur; cur = cur->next) {
|
|
|
|
if (cur->co->data == co->data) // comparing data ptr catches aliases
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
struct m_opt_backup *bc = talloc_ptrtype(NULL, bc);
|
|
|
|
*bc = (struct m_opt_backup) {
|
|
|
|
.co = co,
|
|
|
|
.backup = talloc_zero_size(bc, co->opt->type->size),
|
|
|
|
};
|
|
|
|
m_option_copy(co->opt, bc->backup, co->data);
|
|
|
|
bc->next = config->backup_opts;
|
|
|
|
config->backup_opts = bc;
|
2012-05-17 00:31:11 +00:00
|
|
|
}
|
|
|
|
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
void m_config_enter_file_local(struct m_config *config)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
assert(!config->file_local_mode);
|
|
|
|
config->file_local_mode = true;
|
2012-09-18 12:29:22 +00:00
|
|
|
for (struct m_config_option *co = config->opts; co; co = co->next) {
|
|
|
|
if (co->opt->flags & M_OPT_LOCAL)
|
|
|
|
ensure_backup(config, co);
|
|
|
|
}
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
|
|
|
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
void m_config_leave_file_local(struct m_config *config)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
assert(config->file_local_mode);
|
|
|
|
config->file_local_mode = false;
|
2013-07-31 18:44:16 +00:00
|
|
|
while (config->backup_opts) {
|
|
|
|
struct m_opt_backup *bc = config->backup_opts;
|
|
|
|
config->backup_opts = bc->next;
|
|
|
|
|
|
|
|
m_option_copy(bc->co->opt, bc->co->data, bc->backup);
|
|
|
|
m_option_free(bc->co->opt, bc->backup);
|
|
|
|
talloc_free(bc);
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2013-04-10 19:06:00 +00:00
|
|
|
void m_config_mark_file_local(struct m_config *config, const char *opt)
|
|
|
|
{
|
|
|
|
struct m_config_option *co = m_config_get_co(config, bstr0(opt));
|
|
|
|
if (co) {
|
|
|
|
ensure_backup(config, co);
|
|
|
|
} else {
|
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_ERR, "Option %s not found.\n", opt);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void m_config_mark_all_file_local(struct m_config *config)
|
|
|
|
{
|
|
|
|
for (struct m_config_option *co = config->opts; co; co = co->next)
|
|
|
|
ensure_backup(config, co);
|
|
|
|
}
|
|
|
|
|
options: change handling of "no-" options yet again
Commit 4a40eed "options: change handling of "no-" options" generally
improved the handling of automatically added negation options
(recognizing "--no-opt", even though only "--opt" is declared in the
option list).
Unfortunately, one corner case was missed, which broke the option
"--input=no-default-bindings" (other suboptions, e.g. VO suboptions,
were not affected, and this is the only option where this mattered).
Instead of increasing the complexity further, use a completely different
approach: add the "--no-" options at runtime, and make them behave like
real options. This approach could be considered slightly less elegant,
because the code now has to worry about some option implementation
details rather than leaving it to the parser, but all in all the new
code is simpler and there are less weird corner cases to worry about.
2013-02-16 18:57:57 +00:00
|
|
|
// Given an option --opt, add --no-opt (if applicable).
|
|
|
|
static void add_negation_option(struct m_config *config,
|
|
|
|
struct m_config_option *parent,
|
|
|
|
const struct m_option *opt)
|
|
|
|
{
|
|
|
|
int value;
|
|
|
|
if (opt->type == CONF_TYPE_FLAG) {
|
|
|
|
value = opt->min;
|
|
|
|
} else if (opt->type == CONF_TYPE_CHOICE) {
|
|
|
|
// Find out whether there's a "no" choice.
|
|
|
|
// m_option_parse() should be used for this, but it prints
|
|
|
|
// unsilenceable error messages.
|
|
|
|
struct m_opt_choice_alternatives *alt = opt->priv;
|
|
|
|
for ( ; alt->name; alt++) {
|
|
|
|
if (strcmp(alt->name, "no") == 0)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
if (!alt->name)
|
|
|
|
return;
|
|
|
|
value = alt->value;
|
|
|
|
} else {
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
struct m_option *no_opt = talloc_ptrtype(config, no_opt);
|
|
|
|
*no_opt = (struct m_option) {
|
|
|
|
.name = talloc_asprintf(no_opt, "no-%s", opt->name),
|
|
|
|
.type = CONF_TYPE_STORE,
|
|
|
|
.flags = opt->flags & (M_OPT_NOCFG | M_OPT_GLOBAL | M_OPT_LOCAL |
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
M_OPT_PRE_PARSE),
|
options: change handling of "no-" options yet again
Commit 4a40eed "options: change handling of "no-" options" generally
improved the handling of automatically added negation options
(recognizing "--no-opt", even though only "--opt" is declared in the
option list).
Unfortunately, one corner case was missed, which broke the option
"--input=no-default-bindings" (other suboptions, e.g. VO suboptions,
were not affected, and this is the only option where this mattered).
Instead of increasing the complexity further, use a completely different
approach: add the "--no-" options at runtime, and make them behave like
real options. This approach could be considered slightly less elegant,
because the code now has to worry about some option implementation
details rather than leaving it to the parser, but all in all the new
code is simpler and there are less weird corner cases to worry about.
2013-02-16 18:57:57 +00:00
|
|
|
.new = opt->new,
|
|
|
|
.p = opt->p,
|
|
|
|
.offset = opt->offset,
|
|
|
|
.max = value,
|
|
|
|
};
|
2013-07-27 19:30:00 +00:00
|
|
|
struct m_config_option *co = m_config_add_option(config, "", parent, no_opt);
|
|
|
|
co->is_generated = true;
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
// Consider a parent option "--sub" and a subopt "opt". Then the above
|
|
|
|
// call will add "no-opt". Add "--no-sub-opt" too. (This former call will
|
|
|
|
// also generate "--sub-no-opt", which is not really needed or wanted, but
|
|
|
|
// is a consequence of supporting "--sub=...:no-opt".)
|
|
|
|
if (parent && parent->name && strlen(parent->name)) {
|
|
|
|
no_opt = talloc_memdup(config, no_opt, sizeof(*no_opt));
|
|
|
|
no_opt->name = opt->name;
|
2013-07-27 19:30:00 +00:00
|
|
|
co = m_config_add_option(config, "no-", parent, no_opt);
|
|
|
|
co->is_generated = true;
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
}
|
options: change handling of "no-" options yet again
Commit 4a40eed "options: change handling of "no-" options" generally
improved the handling of automatically added negation options
(recognizing "--no-opt", even though only "--opt" is declared in the
option list).
Unfortunately, one corner case was missed, which broke the option
"--input=no-default-bindings" (other suboptions, e.g. VO suboptions,
were not affected, and this is the only option where this mattered).
Instead of increasing the complexity further, use a completely different
approach: add the "--no-" options at runtime, and make them behave like
real options. This approach could be considered slightly less elegant,
because the code now has to worry about some option implementation
details rather than leaving it to the parser, but all in all the new
code is simpler and there are less weird corner cases to worry about.
2013-02-16 18:57:57 +00:00
|
|
|
}
|
|
|
|
|
2012-08-06 15:42:53 +00:00
|
|
|
static void add_options(struct m_config *config,
|
|
|
|
struct m_config_option *parent,
|
|
|
|
const struct m_option *defs)
|
2011-09-04 11:34:45 +00:00
|
|
|
{
|
2012-08-05 22:10:13 +00:00
|
|
|
for (int i = 0; defs[i].name; i++)
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
m_config_add_option(config, "", parent, defs + i);
|
|
|
|
}
|
|
|
|
|
|
|
|
// Sub-config that adds all its children to the parent.
|
|
|
|
static bool is_merge_opt(const struct m_option *opt)
|
|
|
|
{
|
|
|
|
return (opt->type->flags & M_OPT_TYPE_HAS_CHILD) && strlen(opt->name) == 0;
|
2011-09-04 11:34:45 +00:00
|
|
|
}
|
|
|
|
|
2013-07-27 19:30:00 +00:00
|
|
|
static struct m_config_option *m_config_add_option(struct m_config *config,
|
|
|
|
const char *prefix,
|
|
|
|
struct m_config_option *parent,
|
|
|
|
const struct m_option *arg)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
|
|
|
assert(config != NULL);
|
|
|
|
assert(arg != NULL);
|
|
|
|
|
|
|
|
// Allocate a new entry for this option
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
struct m_config_option *co = talloc_zero(config, struct m_config_option);
|
2011-07-05 19:31:44 +00:00
|
|
|
co->opt = arg;
|
|
|
|
|
2012-08-06 15:45:17 +00:00
|
|
|
void *optstruct = config->optstruct;
|
|
|
|
if (parent && (parent->opt->type->flags & M_OPT_TYPE_USE_SUBSTRUCT))
|
|
|
|
optstruct = substruct_read_ptr(parent->data);
|
|
|
|
co->data = arg->new ? (char *)optstruct + arg->offset : arg->p;
|
|
|
|
|
2012-08-06 15:42:53 +00:00
|
|
|
if (parent) {
|
|
|
|
// Merge case: pretend it has no parent (note that we still must follow
|
|
|
|
// the "real" parent for accessing struct fields)
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
co->parent = is_merge_opt(parent->opt) ? parent->parent : parent;
|
2012-08-06 15:42:53 +00:00
|
|
|
}
|
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
// Fill in the full name
|
2012-08-06 15:42:53 +00:00
|
|
|
if (co->parent) {
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
co->name = talloc_asprintf(co, "%s-%s", co->parent->name, arg->name);
|
|
|
|
} else {
|
2011-07-05 19:31:44 +00:00
|
|
|
co->name = (char *)arg->name;
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
}
|
|
|
|
co->name = talloc_asprintf(co, "%s%s", prefix, co->name);
|
2011-07-05 19:31:44 +00:00
|
|
|
|
|
|
|
// Option with children -> add them
|
|
|
|
if (arg->type->flags & M_OPT_TYPE_HAS_CHILD) {
|
2012-08-06 15:45:17 +00:00
|
|
|
if (arg->type->flags & M_OPT_TYPE_USE_SUBSTRUCT) {
|
|
|
|
const struct m_sub_options *subopts = arg->priv;
|
|
|
|
if (!substruct_read_ptr(co->data)) {
|
|
|
|
void *subdata = m_config_alloc_struct(config, subopts);
|
|
|
|
substruct_write_ptr(co->data, subdata);
|
|
|
|
}
|
|
|
|
add_options(config, co, subopts->opts);
|
|
|
|
} else {
|
|
|
|
const struct m_option *sub = arg->p;
|
|
|
|
add_options(config, co, sub);
|
|
|
|
}
|
2011-07-05 19:31:44 +00:00
|
|
|
} else {
|
2013-07-31 18:44:34 +00:00
|
|
|
// Initialize options
|
2012-08-06 15:45:17 +00:00
|
|
|
if (co->data) {
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
if (arg->defval) {
|
|
|
|
// Target data in optstruct is supposed to be cleared (consider
|
|
|
|
// m_option freeing previously set dynamic data).
|
2012-08-06 15:45:17 +00:00
|
|
|
m_option_copy(arg, co->data, arg->defval);
|
2012-08-02 00:26:55 +00:00
|
|
|
} else if (arg->type->flags & M_OPT_TYPE_DYNAMIC) {
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
// Initialize dynamically managed fields from static data (like
|
|
|
|
// string options): copy the option into temporary memory,
|
2012-09-18 14:08:17 +00:00
|
|
|
// clear the original option (to stop m_option from freeing the
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
// static data), copy it back.
|
2013-07-31 18:44:34 +00:00
|
|
|
// This would leak memory when done on aliased options.
|
|
|
|
bool aliased = false;
|
|
|
|
for (struct m_config_option *i = config->opts; i; i = i->next) {
|
|
|
|
if (co->data == i->data) {
|
|
|
|
aliased = true;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (!aliased) {
|
2012-09-18 14:08:17 +00:00
|
|
|
union m_option_value temp = {0};
|
|
|
|
m_option_copy(arg, &temp, co->data);
|
2012-08-06 15:45:17 +00:00
|
|
|
memset(co->data, 0, arg->type->size);
|
2012-09-18 14:08:17 +00:00
|
|
|
m_option_copy(arg, co->data, &temp);
|
|
|
|
m_option_free(arg, &temp);
|
2011-07-05 19:31:44 +00:00
|
|
|
}
|
2008-04-30 16:34:26 +00:00
|
|
|
}
|
2008-04-30 16:20:01 +00:00
|
|
|
}
|
|
|
|
}
|
2012-08-06 15:42:53 +00:00
|
|
|
|
|
|
|
// pretend that merge options don't exist (only their children matter)
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
if (!is_merge_opt(co->opt)) {
|
options: use m_config for options instead of m_struct
For some reason, both m_config and m_struct are somewhat similar, except
that m_config is much more powerful. m_config is used for VOs and some
other things, so to unify them. We plan to kick out m_struct and use
m_config for everything. (Unfortunately, m_config is also a bit more
bloated, so this commit isn't all that great, but it will allow to
reduce the option parser mess somewhat.)
This commit also switches all video filters to use the option macros.
One reason is that m_struct and m_config, even though they both use
m_option, store the offsets of the option fields differently (sigh...),
meaning the options defined for either are incompatible. It's easier to
switch everything in one go.
This commit will allow using the -vf option parser for other things,
like VOs and AOs.
2013-07-21 17:33:08 +00:00
|
|
|
struct m_config_option **last = &config->opts;
|
|
|
|
while (*last)
|
|
|
|
last = &(*last)->next;
|
|
|
|
*last = co;
|
2012-08-06 15:42:53 +00:00
|
|
|
}
|
options: change handling of "no-" options yet again
Commit 4a40eed "options: change handling of "no-" options" generally
improved the handling of automatically added negation options
(recognizing "--no-opt", even though only "--opt" is declared in the
option list).
Unfortunately, one corner case was missed, which broke the option
"--input=no-default-bindings" (other suboptions, e.g. VO suboptions,
were not affected, and this is the only option where this mattered).
Instead of increasing the complexity further, use a completely different
approach: add the "--no-" options at runtime, and make them behave like
real options. This approach could be considered slightly less elegant,
because the code now has to worry about some option implementation
details rather than leaving it to the parser, but all in all the new
code is simpler and there are less weird corner cases to worry about.
2013-02-16 18:57:57 +00:00
|
|
|
|
|
|
|
add_negation_option(config, parent, arg);
|
2013-07-27 19:30:00 +00:00
|
|
|
|
|
|
|
return co;
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
|
|
|
|
2012-09-18 18:07:24 +00:00
|
|
|
struct m_config_option *m_config_get_co(const struct m_config *config,
|
|
|
|
struct bstr name)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
|
|
|
struct m_config_option *co;
|
|
|
|
|
|
|
|
for (co = config->opts; co; co = co->next) {
|
2012-07-28 21:47:42 +00:00
|
|
|
struct bstr coname = bstr0(co->name);
|
2011-07-28 08:07:47 +00:00
|
|
|
if ((co->opt->type->flags & M_OPT_TYPE_ALLOW_WILDCARD)
|
|
|
|
&& bstr_endswith0(coname, "*")) {
|
|
|
|
coname.len--;
|
2013-05-15 13:28:29 +00:00
|
|
|
if (bstrcmp(bstr_splice(name, 0, coname.len), coname) == 0)
|
2011-07-05 19:31:44 +00:00
|
|
|
return co;
|
2013-05-15 13:28:29 +00:00
|
|
|
} else if (bstrcmp(coname, name) == 0)
|
2011-07-05 19:31:44 +00:00
|
|
|
return co;
|
|
|
|
}
|
|
|
|
return NULL;
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
|
|
|
|
options: use m_config for options instead of m_struct
For some reason, both m_config and m_struct are somewhat similar, except
that m_config is much more powerful. m_config is used for VOs and some
other things, so to unify them. We plan to kick out m_struct and use
m_config for everything. (Unfortunately, m_config is also a bit more
bloated, so this commit isn't all that great, but it will allow to
reduce the option parser mess somewhat.)
This commit also switches all video filters to use the option macros.
One reason is that m_struct and m_config, even though they both use
m_option, store the offsets of the option fields differently (sigh...),
meaning the options defined for either are incompatible. It's easier to
switch everything in one go.
This commit will allow using the -vf option parser for other things,
like VOs and AOs.
2013-07-21 17:33:08 +00:00
|
|
|
const char *m_config_get_positional_option(const struct m_config *config, int n)
|
|
|
|
{
|
2013-07-27 19:30:00 +00:00
|
|
|
int pos = 0;
|
options: use m_config for options instead of m_struct
For some reason, both m_config and m_struct are somewhat similar, except
that m_config is much more powerful. m_config is used for VOs and some
other things, so to unify them. We plan to kick out m_struct and use
m_config for everything. (Unfortunately, m_config is also a bit more
bloated, so this commit isn't all that great, but it will allow to
reduce the option parser mess somewhat.)
This commit also switches all video filters to use the option macros.
One reason is that m_struct and m_config, even though they both use
m_option, store the offsets of the option fields differently (sigh...),
meaning the options defined for either are incompatible. It's easier to
switch everything in one go.
This commit will allow using the -vf option parser for other things,
like VOs and AOs.
2013-07-21 17:33:08 +00:00
|
|
|
for (struct m_config_option *co = config->opts; co; co = co->next) {
|
2013-07-27 19:30:00 +00:00
|
|
|
if (!co->is_generated) {
|
|
|
|
if (pos == n)
|
|
|
|
return co->name;
|
|
|
|
pos++;
|
|
|
|
}
|
options: use m_config for options instead of m_struct
For some reason, both m_config and m_struct are somewhat similar, except
that m_config is much more powerful. m_config is used for VOs and some
other things, so to unify them. We plan to kick out m_struct and use
m_config for everything. (Unfortunately, m_config is also a bit more
bloated, so this commit isn't all that great, but it will allow to
reduce the option parser mess somewhat.)
This commit also switches all video filters to use the option macros.
One reason is that m_struct and m_config, even though they both use
m_option, store the offsets of the option fields differently (sigh...),
meaning the options defined for either are incompatible. It's easier to
switch everything in one go.
This commit will allow using the -vf option parser for other things,
like VOs and AOs.
2013-07-21 17:33:08 +00:00
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2012-05-17 00:31:11 +00:00
|
|
|
static int parse_subopts(struct m_config *config, void *optstruct, char *name,
|
2012-09-21 07:22:25 +00:00
|
|
|
char *prefix, struct bstr param, int flags);
|
2012-05-17 00:31:11 +00:00
|
|
|
|
|
|
|
static int m_config_parse_option(struct m_config *config, void *optstruct,
|
2012-09-21 07:22:25 +00:00
|
|
|
struct bstr name, struct bstr param, int flags)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
|
|
|
assert(config != NULL);
|
2011-07-28 08:07:47 +00:00
|
|
|
assert(name.len != 0);
|
2012-09-21 07:22:25 +00:00
|
|
|
bool set = !(flags & M_SETOPT_CHECK_ONLY);
|
2011-07-05 19:31:44 +00:00
|
|
|
|
2012-05-07 23:10:39 +00:00
|
|
|
struct m_config_option *co = m_config_get_co(config, name);
|
2011-07-28 08:07:47 +00:00
|
|
|
if (!co)
|
2011-07-05 19:31:44 +00:00
|
|
|
return M_OPT_UNKNOWN;
|
2002-11-12 01:56:42 +00:00
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
// This is the only mandatory function
|
|
|
|
assert(co->opt->type->parse);
|
2002-11-12 01:56:42 +00:00
|
|
|
|
2012-09-21 07:22:25 +00:00
|
|
|
if ((flags & M_SETOPT_PRE_PARSE_ONLY) && !(co->opt->flags & M_OPT_PRE_PARSE))
|
|
|
|
return 0;
|
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
// Check if this option isn't forbidden in the current mode
|
2012-09-21 07:22:25 +00:00
|
|
|
if ((flags & M_SETOPT_FROM_CONFIG_FILE) && (co->opt->flags & M_OPT_NOCFG)) {
|
2011-07-05 19:31:44 +00:00
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_ERR,
|
2011-07-28 08:07:47 +00:00
|
|
|
"The %.*s option can't be used in a config file.\n",
|
|
|
|
BSTR_P(name));
|
2011-07-05 19:31:44 +00:00
|
|
|
return M_OPT_INVALID;
|
|
|
|
}
|
2012-08-04 10:07:35 +00:00
|
|
|
if (config->file_local_mode && (co->opt->flags & M_OPT_GLOBAL)) {
|
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_ERR,
|
|
|
|
"The %.*s option is global and can't be set per-file.\n",
|
|
|
|
BSTR_P(name));
|
|
|
|
return M_OPT_INVALID;
|
|
|
|
}
|
2011-07-05 19:31:44 +00:00
|
|
|
|
2013-07-27 19:26:00 +00:00
|
|
|
if (config->includefunc && bstr_equals0(name, "include"))
|
2012-05-07 23:10:39 +00:00
|
|
|
return parse_include(config, param, set);
|
2013-07-27 19:26:00 +00:00
|
|
|
if (config->use_profiles && bstr_equals0(name, "profile"))
|
2012-05-07 23:10:39 +00:00
|
|
|
return parse_profile(config, co->opt, name, param, set);
|
2013-07-27 19:26:00 +00:00
|
|
|
if (config->use_profiles && bstr_equals0(name, "show-profile"))
|
|
|
|
return show_profile(config, param);
|
|
|
|
if (bstr_equals0(name, "list-options"))
|
|
|
|
return list_options(config);
|
2012-05-07 23:10:39 +00:00
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
// Option with children are a bit different to parse
|
|
|
|
if (co->opt->type->flags & M_OPT_TYPE_HAS_CHILD) {
|
2012-05-17 00:31:11 +00:00
|
|
|
char prefix[110];
|
|
|
|
assert(strlen(co->name) < 100);
|
options: drop --opt:subopt option names
For all suboptions, "flat" options were available by separating the
parent option and the sub option with ":", e.g. "--rawvideo:w=123". Drop
this syntax and use "-" as separator. This means even suboptions are
available as normal options now, e.g. "--rawvideo-w=123". The old syntax
doesn't work anymore.
Note that this is completely separate from actual suboptions. For
example, "-rawvideo w=123:h=123" still works. (Not that this syntax is
worth supporting, but it's needed anyway, for for other things like vf
and vo suboptions.)
As a consequence of this change, we also have to add new "no-" prefixed
options for flag suboptions, so that "--no-input-default-bindings"
works. ("--input-no-default-bindings" also works as a consequence of
allowing "-input no-default-bindings" - they are handled by the same
underlying option.)
For --input, always use the full syntax in the manpage. There exist
suboptions other than --input (like --tv, --rawvideo, etc.), but since
they might be handled differently in the future, don't touch these yet.
M_OPT_PREFIXED becomes the default, so remove it. As a minor unrelated
cleanup, get rid of M_OPT_MERGE too and use the OPT_SUBSTRUCT() macro in
some places.
Unrelated: remove the duplicated --tv:buffersize option, fix a typo in
changes.rst.
2013-02-21 21:10:21 +00:00
|
|
|
sprintf(prefix, "%s-", co->name);
|
2012-09-21 07:22:25 +00:00
|
|
|
return parse_subopts(config, optstruct, co->name, prefix, param, flags);
|
2012-05-07 23:10:39 +00:00
|
|
|
}
|
2011-07-05 19:31:44 +00:00
|
|
|
|
mplayer: turn playtree into a list, and change per-file option handling
Summary:
- There is no playtree anymore. It's reduced to a simple list.
- Options are now always global. You can still have per-file options,
but these are optional and require special syntax.
- The slave command pt_step has been removed, and playlist_next
and playlist_prev added. (See etc/input.conf changes.)
This is a user visible incompatible change, and will break slave-mode
applications.
- The pt_clear slave command is renamed to playlist_clear.
- Playtree entries could have multiple files. This is not the case
anymore, and playlist entries have always exactly one entry. Whenever
something adds more than one file (like ASX playlists or dvd:// or
dvdnav:// on the command line), all files are added as separate
playlist entries.
Note that some of the changes are quite deep and violent. Expect
regressions.
The playlist parsing code in particular is of low quality. I didn't try
to improve it, and merely spent to least effort necessary to keep it
somehow working. (Especially ASX playlist handling.)
The playtree code was complicated and bloated. It was also barely used.
Most users don't even know that mplayer manages the playlist as tree,
or how to use it. The most obscure features was probably specifying a
tree on command line (with '{' and '}' to create/close tree nodes). It
filled the player code with complexity and confused users with weird
slave commands like pt_up.
Replace the playtree with a simple flat playlist. Playlist parsers that
actually return trees are changed to append all files to the playlist
pre-order.
It used to be the responsibility of the playtree code to change per-file
config options. Now this is done by the player core, and the playlist
code is free of such details.
Options are not per-file by default anymore. This was a very obscure and
complicated feature that confused even experienced users. Consider the
following command line:
mplayer file1.mkv file2.mkv --no-audio file3.mkv
This will disable the audio for file2.mkv only, because options are
per-file by default. To make the option affect all files, you're
supposed to put it before the first file.
This is bad, because normally you don't need per-file options. They are
very rarely needed, and the only reasonable use cases I can imagine are
use of the encode backend (mplayer encode branch), or for debugging. The
normal use case is made harder, and the feature is perceived as bug.
Even worse, correct usage is hard to explain for users.
Make all options global by default. The position of an option isn't
significant anymore (except for options that compensate each other,
consider --shuffle --no-shuffle).
One other important change is that no options are reset anymore if a
new file is started. If you change settings with slave mode commands,
they will not be changed by playing a new file. (Exceptions include
settings that are too file specific, like audio/subtitle stream
selection.)
There is still some need for per-file options. Debugging and encoding
are use cases that profit from per-file options. Per-file profiles (as
well as per-protocol and per-VO/AO options) need the implementation
related mechanisms to backup and restore options when the playback file
changes.
Simplify the save-slot stuff, which is possible because there is no
hierarchical play tree anymore. Now there's a simple backup field.
Add a way to specify per-file options on command line. Example:
mplayer f1.mkv -o0 --{ -o1 f2.mkv -o2 f3.mkv --} f4.mkv -o3
will have the following options per file set:
f1.mkv, f4.mkv: -o0 -o3
f2.mkv, f3.mkv: -o0 -o3 -o1 -o2
The options --{ and --} start and end per-file options. All files inside
the { } will be affected by the options equally (similar to how global
options and multiple files are handled). When playback of a file starts,
the per-file options are set according to the command line. When
playback ends, the per-file options are restored to the values when
playback started.
2012-07-31 19:33:26 +00:00
|
|
|
if (set)
|
|
|
|
ensure_backup(config, co);
|
|
|
|
|
2012-08-06 15:45:17 +00:00
|
|
|
return m_option_parse(co->opt, name, param, set ? co->data : NULL);
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
|
|
|
|
2012-05-17 00:31:11 +00:00
|
|
|
static int parse_subopts(struct m_config *config, void *optstruct, char *name,
|
2012-09-21 07:22:25 +00:00
|
|
|
char *prefix, struct bstr param, int flags)
|
2012-05-17 00:31:11 +00:00
|
|
|
{
|
|
|
|
char **lst = NULL;
|
|
|
|
// Split the argument into child options
|
2012-08-05 21:34:28 +00:00
|
|
|
int r = m_option_type_subconfig.parse(NULL, bstr0(""), param, &lst);
|
2012-05-17 00:31:11 +00:00
|
|
|
if (r < 0)
|
|
|
|
return r;
|
|
|
|
// Parse the child options
|
|
|
|
for (int i = 0; lst && lst[2 * i]; i++) {
|
|
|
|
// Build the full name
|
|
|
|
char n[110];
|
|
|
|
if (snprintf(n, 110, "%s%s", prefix, lst[2 * i]) > 100)
|
|
|
|
abort();
|
2012-09-21 07:22:25 +00:00
|
|
|
r = m_config_parse_option(config, optstruct, bstr0(n),
|
|
|
|
bstr0(lst[2 * i + 1]), flags);
|
|
|
|
if (r < 0) {
|
|
|
|
if (r > M_OPT_EXIT) {
|
2012-05-17 00:31:11 +00:00
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_ERR,
|
2012-09-21 07:22:25 +00:00
|
|
|
"Error parsing suboption %s/%s (%s)\n",
|
|
|
|
name, lst[2 * i], m_option_strerror(r));
|
2012-05-17 00:31:11 +00:00
|
|
|
r = M_OPT_INVALID;
|
2012-09-21 07:22:25 +00:00
|
|
|
}
|
2012-05-17 00:31:11 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
talloc_free(lst);
|
|
|
|
return r;
|
|
|
|
}
|
|
|
|
|
2012-09-21 07:22:25 +00:00
|
|
|
int m_config_parse_suboptions(struct m_config *config, char *name,
|
|
|
|
char *subopts)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
2012-09-21 07:22:25 +00:00
|
|
|
if (!subopts || !*subopts)
|
|
|
|
return 0;
|
|
|
|
int r = parse_subopts(config, config->optstruct, name, "", bstr0(subopts), 0);
|
|
|
|
if (r < 0 && r > M_OPT_EXIT) {
|
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_ERR, "Error parsing suboption %s (%s)\n",
|
|
|
|
name, m_option_strerror(r));
|
|
|
|
r = M_OPT_INVALID;
|
|
|
|
}
|
|
|
|
return r;
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
|
|
|
|
2012-09-21 07:22:25 +00:00
|
|
|
int m_config_set_option_ext(struct m_config *config, struct bstr name,
|
|
|
|
struct bstr param, int flags)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
2012-09-21 07:22:25 +00:00
|
|
|
int r = m_config_parse_option(config, config->optstruct, name, param, flags);
|
|
|
|
if (r < 0 && r > M_OPT_EXIT) {
|
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_ERR, "Error parsing option %.*s (%s)\n",
|
|
|
|
BSTR_P(name), m_option_strerror(r));
|
|
|
|
r = M_OPT_INVALID;
|
2011-07-05 19:31:44 +00:00
|
|
|
}
|
|
|
|
return r;
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
|
|
|
|
2012-09-21 07:22:25 +00:00
|
|
|
int m_config_set_option(struct m_config *config, struct bstr name,
|
|
|
|
struct bstr param)
|
2012-05-17 00:31:11 +00:00
|
|
|
{
|
2012-09-21 07:22:25 +00:00
|
|
|
return m_config_set_option_ext(config, name, param, 0);
|
2012-05-17 00:31:11 +00:00
|
|
|
}
|
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
const struct m_option *m_config_get_option(const struct m_config *config,
|
2011-07-28 08:07:47 +00:00
|
|
|
struct bstr name)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
|
|
|
struct m_config_option *co;
|
2002-11-12 01:56:42 +00:00
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
assert(config != NULL);
|
2002-11-12 01:56:42 +00:00
|
|
|
|
2011-07-28 08:07:47 +00:00
|
|
|
co = m_config_get_co(config, name);
|
2011-07-05 19:31:44 +00:00
|
|
|
if (co)
|
|
|
|
return co->opt;
|
|
|
|
else
|
|
|
|
return NULL;
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
|
|
|
|
options: change handling of "no-" options yet again
Commit 4a40eed "options: change handling of "no-" options" generally
improved the handling of automatically added negation options
(recognizing "--no-opt", even though only "--opt" is declared in the
option list).
Unfortunately, one corner case was missed, which broke the option
"--input=no-default-bindings" (other suboptions, e.g. VO suboptions,
were not affected, and this is the only option where this mattered).
Instead of increasing the complexity further, use a completely different
approach: add the "--no-" options at runtime, and make them behave like
real options. This approach could be considered slightly less elegant,
because the code now has to worry about some option implementation
details rather than leaving it to the parser, but all in all the new
code is simpler and there are less weird corner cases to worry about.
2013-02-16 18:57:57 +00:00
|
|
|
int m_config_option_requires_param(struct m_config *config, bstr name)
|
2012-09-20 01:32:01 +00:00
|
|
|
{
|
options: change handling of "no-" options yet again
Commit 4a40eed "options: change handling of "no-" options" generally
improved the handling of automatically added negation options
(recognizing "--no-opt", even though only "--opt" is declared in the
option list).
Unfortunately, one corner case was missed, which broke the option
"--input=no-default-bindings" (other suboptions, e.g. VO suboptions,
were not affected, and this is the only option where this mattered).
Instead of increasing the complexity further, use a completely different
approach: add the "--no-" options at runtime, and make them behave like
real options. This approach could be considered slightly less elegant,
because the code now has to worry about some option implementation
details rather than leaving it to the parser, but all in all the new
code is simpler and there are less weird corner cases to worry about.
2013-02-16 18:57:57 +00:00
|
|
|
const struct m_option *opt = m_config_get_option(config, name);
|
2012-09-21 07:22:25 +00:00
|
|
|
if (opt) {
|
options: change handling of "no-" options yet again
Commit 4a40eed "options: change handling of "no-" options" generally
improved the handling of automatically added negation options
(recognizing "--no-opt", even though only "--opt" is declared in the
option list).
Unfortunately, one corner case was missed, which broke the option
"--input=no-default-bindings" (other suboptions, e.g. VO suboptions,
were not affected, and this is the only option where this mattered).
Instead of increasing the complexity further, use a completely different
approach: add the "--no-" options at runtime, and make them behave like
real options. This approach could be considered slightly less elegant,
because the code now has to worry about some option implementation
details rather than leaving it to the parser, but all in all the new
code is simpler and there are less weird corner cases to worry about.
2013-02-16 18:57:57 +00:00
|
|
|
if (bstr_endswith0(name, "-clr"))
|
|
|
|
return 0;
|
2013-04-21 01:18:32 +00:00
|
|
|
return m_option_required_params(opt);
|
2012-09-21 07:22:25 +00:00
|
|
|
}
|
options: change handling of "no-" options yet again
Commit 4a40eed "options: change handling of "no-" options" generally
improved the handling of automatically added negation options
(recognizing "--no-opt", even though only "--opt" is declared in the
option list).
Unfortunately, one corner case was missed, which broke the option
"--input=no-default-bindings" (other suboptions, e.g. VO suboptions,
were not affected, and this is the only option where this mattered).
Instead of increasing the complexity further, use a completely different
approach: add the "--no-" options at runtime, and make them behave like
real options. This approach could be considered slightly less elegant,
because the code now has to worry about some option implementation
details rather than leaving it to the parser, but all in all the new
code is simpler and there are less weird corner cases to worry about.
2013-02-16 18:57:57 +00:00
|
|
|
return M_OPT_UNKNOWN;
|
2012-09-20 01:32:01 +00:00
|
|
|
}
|
|
|
|
|
2013-07-27 19:27:08 +00:00
|
|
|
static struct m_config *get_defaults(const struct m_config *config)
|
|
|
|
{
|
|
|
|
return m_config_new(NULL, config->optstruct_size,
|
2013-07-31 18:45:06 +00:00
|
|
|
config->optstruct_defaults, config->options,
|
|
|
|
config->suboptinit);
|
2013-07-27 19:27:08 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static char *get_option_value_string(const struct m_config *config,
|
|
|
|
const char *name)
|
|
|
|
{
|
|
|
|
struct m_config_option *co = m_config_get_co(config, bstr0(name));
|
|
|
|
if (!co || !co->data)
|
|
|
|
return NULL;
|
|
|
|
return m_option_print(co->opt, co->data);
|
|
|
|
}
|
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
void m_config_print_option_list(const struct m_config *config)
|
|
|
|
{
|
|
|
|
char min[50], max[50];
|
|
|
|
struct m_config_option *co;
|
|
|
|
int count = 0;
|
2002-11-12 01:56:42 +00:00
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
if (!config->opts)
|
|
|
|
return;
|
2002-11-12 01:56:42 +00:00
|
|
|
|
2013-07-27 19:27:08 +00:00
|
|
|
struct m_config *defaults = get_defaults(config);
|
|
|
|
|
2013-07-21 18:16:46 +00:00
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_INFO, "Options:\n\n");
|
2011-07-05 19:31:44 +00:00
|
|
|
for (co = config->opts; co; co = co->next) {
|
|
|
|
const struct m_option *opt = co->opt;
|
|
|
|
if (opt->type->flags & M_OPT_TYPE_HAS_CHILD)
|
|
|
|
continue;
|
2013-07-27 19:30:00 +00:00
|
|
|
if (co->is_generated)
|
|
|
|
continue;
|
2013-07-21 18:16:46 +00:00
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " %-30.30s", co->name);
|
|
|
|
if (opt->type == &m_option_type_choice) {
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " Choices:");
|
|
|
|
struct m_opt_choice_alternatives *alt = opt->priv;
|
|
|
|
for (int n = 0; alt[n].name; n++)
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " %s", alt[n].name);
|
|
|
|
if (opt->flags & (M_OPT_MIN | M_OPT_MAX))
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " (or an integer)");
|
|
|
|
} else {
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " %s", co->opt->type->name);
|
|
|
|
}
|
|
|
|
if (opt->flags & (M_OPT_MIN | M_OPT_MAX)) {
|
|
|
|
snprintf(min, sizeof(min), "any");
|
|
|
|
snprintf(max, sizeof(max), "any");
|
|
|
|
if (opt->flags & M_OPT_MIN)
|
2013-07-27 19:31:04 +00:00
|
|
|
snprintf(min, sizeof(min), "%.14g", opt->min);
|
2013-07-21 18:16:46 +00:00
|
|
|
if (opt->flags & M_OPT_MAX)
|
2013-07-27 19:31:04 +00:00
|
|
|
snprintf(max, sizeof(max), "%.14g", opt->max);
|
2013-07-21 18:16:46 +00:00
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " (%s to %s)", min, max);
|
|
|
|
}
|
2013-07-27 19:27:08 +00:00
|
|
|
char *def = get_option_value_string(defaults, co->name);
|
|
|
|
if (def) {
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " (default: %s)", def);
|
|
|
|
talloc_free(def);
|
|
|
|
}
|
2013-07-21 18:16:46 +00:00
|
|
|
if (opt->flags & CONF_GLOBAL)
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " [global]");
|
|
|
|
if (opt->flags & CONF_NOCFG)
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, " [nocfg]");
|
|
|
|
mp_msg(MSGT_CFGPARSER, MSGL_INFO, "\n");
|
2011-07-05 19:31:44 +00:00
|
|
|
count++;
|
|
|
|
}
|
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_INFO, "\nTotal: %d options\n", count);
|
2013-07-27 19:27:08 +00:00
|
|
|
|
|
|
|
talloc_free(defaults);
|
2002-11-12 01:56:42 +00:00
|
|
|
}
|
2006-01-24 11:16:13 +00:00
|
|
|
|
2013-07-27 19:26:00 +00:00
|
|
|
struct m_profile *m_config_get_profile(const struct m_config *config, bstr name)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
2013-07-27 19:26:00 +00:00
|
|
|
for (struct m_profile *p = config->profiles; p; p = p->next) {
|
|
|
|
if (bstr_equals0(name, p->name))
|
2011-07-05 19:31:44 +00:00
|
|
|
return p;
|
2013-07-27 19:26:00 +00:00
|
|
|
}
|
2011-07-05 19:31:44 +00:00
|
|
|
return NULL;
|
2006-01-24 11:16:13 +00:00
|
|
|
}
|
|
|
|
|
2013-07-27 19:26:00 +00:00
|
|
|
struct m_profile *m_config_get_profile0(const struct m_config *config,
|
|
|
|
char *name)
|
|
|
|
{
|
|
|
|
return m_config_get_profile(config, bstr0(name));
|
|
|
|
}
|
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
struct m_profile *m_config_add_profile(struct m_config *config, char *name)
|
|
|
|
{
|
2013-07-27 19:26:00 +00:00
|
|
|
struct m_profile *p = m_config_get_profile0(config, name);
|
2011-07-05 19:31:44 +00:00
|
|
|
if (p)
|
|
|
|
return p;
|
|
|
|
p = talloc_zero(config, struct m_profile);
|
|
|
|
p->name = talloc_strdup(p, name);
|
|
|
|
p->next = config->profiles;
|
|
|
|
config->profiles = p;
|
|
|
|
return p;
|
2006-01-24 11:16:13 +00:00
|
|
|
}
|
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
void m_profile_set_desc(struct m_profile *p, char *desc)
|
|
|
|
{
|
2008-04-25 12:59:30 +00:00
|
|
|
talloc_free(p->desc);
|
|
|
|
p->desc = talloc_strdup(p, desc);
|
2006-01-24 11:16:13 +00:00
|
|
|
}
|
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
int m_config_set_profile_option(struct m_config *config, struct m_profile *p,
|
2012-09-21 07:22:25 +00:00
|
|
|
bstr name, bstr val)
|
2011-07-05 19:31:44 +00:00
|
|
|
{
|
2012-09-21 07:22:25 +00:00
|
|
|
int i = m_config_set_option_ext(config, name, val,
|
|
|
|
M_SETOPT_CHECK_ONLY |
|
|
|
|
M_SETOPT_FROM_CONFIG_FILE);
|
2011-07-05 19:31:44 +00:00
|
|
|
if (i < 0)
|
|
|
|
return i;
|
|
|
|
p->opts = talloc_realloc(p, p->opts, char *, 2 * (p->num_opts + 2));
|
2012-09-21 07:22:25 +00:00
|
|
|
p->opts[p->num_opts * 2] = bstrdup0(p, name);
|
|
|
|
p->opts[p->num_opts * 2 + 1] = bstrdup0(p, val);
|
2011-07-05 19:31:44 +00:00
|
|
|
p->num_opts++;
|
|
|
|
p->opts[p->num_opts * 2] = p->opts[p->num_opts * 2 + 1] = NULL;
|
|
|
|
return 1;
|
2006-01-24 11:16:13 +00:00
|
|
|
}
|
|
|
|
|
2011-07-05 19:31:44 +00:00
|
|
|
void m_config_set_profile(struct m_config *config, struct m_profile *p)
|
|
|
|
{
|
|
|
|
if (config->profile_depth > MAX_PROFILE_DEPTH) {
|
|
|
|
mp_tmsg(MSGT_CFGPARSER, MSGL_WARN,
|
|
|
|
"WARNING: Profile inclusion too deep.\n");
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
config->profile_depth++;
|
2012-09-21 07:22:25 +00:00
|
|
|
for (int i = 0; i < p->num_opts; i++) {
|
|
|
|
m_config_set_option_ext(config,
|
|
|
|
bstr0(p->opts[2 * i]),
|
|
|
|
bstr0(p->opts[2 * i + 1]),
|
|
|
|
M_SETOPT_FROM_CONFIG_FILE);
|
|
|
|
}
|
2011-07-05 19:31:44 +00:00
|
|
|
config->profile_depth--;
|
2006-01-24 11:16:13 +00:00
|
|
|
}
|
2012-08-06 15:45:17 +00:00
|
|
|
|
|
|
|
void *m_config_alloc_struct(void *talloc_parent,
|
|
|
|
const struct m_sub_options *subopts)
|
|
|
|
{
|
|
|
|
void *substruct = talloc_zero_size(talloc_parent, subopts->size);
|
|
|
|
if (subopts->defaults)
|
|
|
|
memcpy(substruct, subopts->defaults, subopts->size);
|
|
|
|
return substruct;
|
|
|
|
}
|