mpv/options/m_config.h

364 lines
16 KiB
C
Raw Normal View History

/*
* This file is part of mpv.
*
* mpv is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* mpv 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 Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with mpv. If not, see <http://www.gnu.org/licenses/>.
*/
#ifndef MPLAYER_M_CONFIG_H
#define MPLAYER_M_CONFIG_H
#include <stddef.h>
#include <stdint.h>
#include <stdbool.h>
#include "misc/bstr.h"
// m_config provides an API to manipulate the config variables in MPlayer.
// It makes use of the Options API to provide a context stack that
// allows saving and later restoring the state of all variables.
typedef struct m_profile m_profile_t;
struct m_option;
struct m_option_type;
struct m_sub_options;
struct m_obj_desc;
struct m_obj_settings;
struct mp_log;
struct mp_dispatch_queue;
// Config option
struct m_config_option {
bool is_hidden : 1; // Does not show up in help
bool is_set_from_cmdline : 1; // Set by user from command line
bool is_set_from_config : 1; // Set by a config file
bool is_set_locally : 1; // Has a backup entry
bool warning_was_printed : 1;
int16_t group_index; // Index into m_config.groups
const char *name; // Full name (ie option-subopt)
const struct m_option *opt; // Option description
void *data; // Raw value of the option
};
// Config object
/** \ingroup Config */
typedef struct m_config {
struct mp_log *log;
struct mpv_global *global; // can be NULL
// Registered options.
struct m_config_option *opts; // all options, even suboptions
int num_opts;
// List of defined profiles.
struct m_profile *profiles;
// Depth when recursively including profiles.
int profile_depth;
struct m_opt_backup *backup_opts;
bool use_profiles;
bool is_toplevel;
int (*includefunc)(void *ctx, char *filename, int flags);
void *includefunc_ctx;
// Notification after an option was successfully written to.
// Uses flags as set in UPDATE_OPTS_MASK.
player: change m_config to use new option handling mechanisms Instead of making m_config a special-case, it more or less uses the underlying m_config_cache/m_config_shadow APIs properly. This makes the player core a (relatively) equivalent user of the core option API. In particular, this means that other threads can change core options with m_config_cache_write_opt() calls (before this commit, this merely led to diverging option values). An important change is that before this commit, mpctx->opts contained the "master copy" of all option data. Now it's just another copy of the option data, and the shadow copy is considered the master. This is why whenever mpctx->opts is written, the change needs to be copied to the master (thus why this commits add a bunch of m_config_notify... calls). If another thread (e.g. a VO) changes an option, async_change_cb is now invoked, which funnels the change notification through the player's layers. The new self_notification parameter on mp_option_change_callback is so that m_config_notify... doesn't trigger recursion, and it's used in cases where the change was already "processed". It's still needed to trigger libmpv property updates. (I considered using an extra m_config_cache for that, but it'd only cause problems with no advantages.) I think the recent changes actually forgot to send libmpv property updates in some cases. This should fix this anyway. In some cases, property updates are reworked, and the potential for bugs should be lower (probably). The primary point of this change is to allow external updates, for example by a VO writing the fullscreen option if the window state is changed by the window manager (rather than mpv changing it). This is not used yet, but the following commits will.
2019-11-29 11:49:15 +00:00
// self_update==true means the update was caused by a call to
// m_config_notify_change_opt_ptr(). If false, it's caused either by
// m_config_set_option_*() (and similar) calls or external updates.
void (*option_change_callback)(void *ctx, struct m_config_option *co,
player: change m_config to use new option handling mechanisms Instead of making m_config a special-case, it more or less uses the underlying m_config_cache/m_config_shadow APIs properly. This makes the player core a (relatively) equivalent user of the core option API. In particular, this means that other threads can change core options with m_config_cache_write_opt() calls (before this commit, this merely led to diverging option values). An important change is that before this commit, mpctx->opts contained the "master copy" of all option data. Now it's just another copy of the option data, and the shadow copy is considered the master. This is why whenever mpctx->opts is written, the change needs to be copied to the master (thus why this commits add a bunch of m_config_notify... calls). If another thread (e.g. a VO) changes an option, async_change_cb is now invoked, which funnels the change notification through the player's layers. The new self_notification parameter on mp_option_change_callback is so that m_config_notify... doesn't trigger recursion, and it's used in cases where the change was already "processed". It's still needed to trigger libmpv property updates. (I considered using an extra m_config_cache for that, but it'd only cause problems with no advantages.) I think the recent changes actually forgot to send libmpv property updates in some cases. This should fix this anyway. In some cases, property updates are reworked, and the potential for bugs should be lower (probably). The primary point of this change is to allow external updates, for example by a VO writing the fullscreen option if the window state is changed by the window manager (rather than mpv changing it). This is not used yet, but the following commits will.
2019-11-29 11:49:15 +00:00
int flags, bool self_update);
void *option_change_callback_ctx;
// For the command line parser
int recursion_depth;
void *optstruct; // struct mpopts or other
// Private. Non-NULL if data was allocated. m_config_option.data uses it.
player: change m_config to use new option handling mechanisms Instead of making m_config a special-case, it more or less uses the underlying m_config_cache/m_config_shadow APIs properly. This makes the player core a (relatively) equivalent user of the core option API. In particular, this means that other threads can change core options with m_config_cache_write_opt() calls (before this commit, this merely led to diverging option values). An important change is that before this commit, mpctx->opts contained the "master copy" of all option data. Now it's just another copy of the option data, and the shadow copy is considered the master. This is why whenever mpctx->opts is written, the change needs to be copied to the master (thus why this commits add a bunch of m_config_notify... calls). If another thread (e.g. a VO) changes an option, async_change_cb is now invoked, which funnels the change notification through the player's layers. The new self_notification parameter on mp_option_change_callback is so that m_config_notify... doesn't trigger recursion, and it's used in cases where the change was already "processed". It's still needed to trigger libmpv property updates. (I considered using an extra m_config_cache for that, but it'd only cause problems with no advantages.) I think the recent changes actually forgot to send libmpv property updates in some cases. This should fix this anyway. In some cases, property updates are reworked, and the potential for bugs should be lower (probably). The primary point of this change is to allow external updates, for example by a VO writing the fullscreen option if the window state is changed by the window manager (rather than mpv changing it). This is not used yet, but the following commits will.
2019-11-29 11:49:15 +00:00
// API users call m_config_set_update_dispatch_queue() to get async updates.
struct m_config_cache *cache;
// Private. Thread-safe shadow memory; only set for the main m_config.
struct m_config_shadow *shadow;
} m_config_t;
// Create a new config object.
// talloc_ctx: talloc parent context for the m_config allocation
// root: description of all options
// Note that the m_config object will keep pointers to root and log.
struct m_config *m_config_new(void *talloc_ctx, struct mp_log *log,
const struct m_sub_options *root);
options: support parsing values into substructs Add an alternate mode for option parser objects (struct m_config) which is not inherently tied to any particular instance of an option value struct. Instead, this type or parsers can be used to initialize defaults in or parse values into a struct given as a parameter. They do not have the save slot functionality used for main player configuration. The new functionality will be used to replace the separate subopt_helper.c parsing code that is currently used to parse per-object suboptions in VOs etc. Previously, option default values were handled by initializing them in external code before creating a parser. This initialization was done with constants even for dynamically-allocated types like strings. Because trying to free a pointer to a constant would cause a crash when trying to replace the default with another value, parser initialization code then replaced all the original defaults with dynamically-allocated copies. This replace-with-copy behavior is no longer supported for new-style options; instead the option definition itself may contain a default value (new OPTDEF macros), and the new function m_config_initialize() is used to set all options to their default values. Convert the existing initialized dynamically allocated options in main config (the string options --dumpfile, --term-osd-esc, --input=conf) to use this. Other non-dynamic ones could be later converted to use this style of initialization too. There's currently no public call to free all dynamically allocated options in a given option struct because I intend to use talloc functionality for that (make them children of the struct and free with it).
2012-05-17 00:31:11 +00:00
// Create a m_config for the given desc. This is for --af/--vf, which have
// different sub-options for every filter (represented by separate desc
// structs).
// args is an array of key/value pairs (args=[k0, v0, k1, v1, ..., NULL]).
// name/defaults is only needed for the legacy af-defaults/vf-defaults options.
struct m_config *m_config_from_obj_desc_and_args(void *ta_parent,
struct mp_log *log, struct mpv_global *global, struct m_obj_desc *desc,
const char *name, struct m_obj_settings *defaults, char **args);
// Like m_config_from_obj_desc_and_args(), but don't allocate option the
// struct, i.e. m_config.optstruct==NULL. This is used by the sub-option
// parser (--af/--vf, to a lesser degree --ao/--vo) to check sub-option names
// and types.
struct m_config *m_config_from_obj_desc_noalloc(void *talloc_ctx,
struct mp_log *log,
struct m_obj_desc *desc);
// Allocate a priv struct that is backed by global options (like AOs and VOs,
// anything that uses m_obj_list.use_global_options == true).
// The result contains a snapshot of the current option values of desc->options.
// For convenience, desc->options can be NULL; then priv struct is allocated
// with just zero (or priv_defaults if set).
void *m_config_group_from_desc(void *ta_parent, struct mp_log *log,
struct mpv_global *global, struct m_obj_desc *desc, const char *name);
// Make sure the option is backed up. If it's already backed up, do nothing.
// All backed up options can be restored with m_config_restore_backups().
void m_config_backup_opt(struct m_config *config, const char *opt);
// Call m_config_backup_opt() on all options.
void m_config_backup_all_opts(struct m_config *config);
// Restore all options backed up with m_config_backup_opt(), and delete the
// backups afterwards.
void m_config_restore_backups(struct m_config *config);
enum {
M_SETOPT_PRE_PARSE_ONLY = 1, // Silently ignore non-M_OPT_PRE_PARSE opt.
M_SETOPT_CHECK_ONLY = 2, // Don't set, just check name/value
M_SETOPT_FROM_CONFIG_FILE = 4, // Reject M_OPT_NOCFG opt. (print error)
M_SETOPT_FROM_CMDLINE = 8, // Mark as set by command line
M_SETOPT_BACKUP = 16, // Call m_config_backup_opt() before
M_SETOPT_PRESERVE_CMDLINE = 32, // Don't set if already marked as FROM_CMDLINE
M_SETOPT_NO_PRE_PARSE = 128, // Reject M_OPT_PREPARSE options
M_SETOPT_NO_OVERWRITE = 256, // Skip options marked with FROM_*
};
// Set the named option to the given string. This is for command line and config
// file use only.
// flags: combination of M_SETOPT_* flags (0 for normal operation)
// Returns >= 0 on success, otherwise see OptionParserReturn.
int m_config_set_option_cli(struct m_config *config, struct bstr name,
struct bstr param, int flags);
// Similar to m_config_set_option_cli(), but set as data in its native format.
// This takes care of some details like sending change notifications.
// The type data points to is as in: co->opt
int m_config_set_option_raw(struct m_config *config, struct m_config_option *co,
void *data, int flags);
void m_config_mark_co_flags(struct m_config_option *co, int flags);
// Convert the mpv_node to raw option data, then call m_config_set_option_raw().
struct mpv_node;
int m_config_set_option_node(struct m_config *config, bstr name,
struct mpv_node *data, int flags);
// Return option descriptor. You shouldn't use this.
struct m_config_option *m_config_get_co(const struct m_config *config,
struct bstr name);
// Same as above, but does not resolve aliases or trigger warning messages.
struct m_config_option *m_config_get_co_raw(const struct m_config *config,
struct bstr name);
// Special uses only. Look away.
int m_config_get_co_count(struct m_config *config);
struct m_config_option *m_config_get_co_index(struct m_config *config, int index);
const void *m_config_get_co_default(const struct m_config *config,
struct m_config_option *co);
// Return the n-th option by position. n==0 is the first option. If there are
// less than (n + 1) options, return NULL.
const char *m_config_get_positional_option(const struct m_config *config, int n);
// Return a hint to the option parser whether a parameter is/may be required.
// The option may still accept empty/non-empty parameters independent from
// this, and this function is useful only for handling ambiguous options like
// flags (e.g. "--a" is ok, "--a=yes" is also ok).
// Returns: error code (<0), or number of expected params (0, 1)
int m_config_option_requires_param(struct m_config *config, bstr name);
// Notify m_config_cache users that the option has (probably) changed its value.
player: change m_config to use new option handling mechanisms Instead of making m_config a special-case, it more or less uses the underlying m_config_cache/m_config_shadow APIs properly. This makes the player core a (relatively) equivalent user of the core option API. In particular, this means that other threads can change core options with m_config_cache_write_opt() calls (before this commit, this merely led to diverging option values). An important change is that before this commit, mpctx->opts contained the "master copy" of all option data. Now it's just another copy of the option data, and the shadow copy is considered the master. This is why whenever mpctx->opts is written, the change needs to be copied to the master (thus why this commits add a bunch of m_config_notify... calls). If another thread (e.g. a VO) changes an option, async_change_cb is now invoked, which funnels the change notification through the player's layers. The new self_notification parameter on mp_option_change_callback is so that m_config_notify... doesn't trigger recursion, and it's used in cases where the change was already "processed". It's still needed to trigger libmpv property updates. (I considered using an extra m_config_cache for that, but it'd only cause problems with no advantages.) I think the recent changes actually forgot to send libmpv property updates in some cases. This should fix this anyway. In some cases, property updates are reworked, and the potential for bugs should be lower (probably). The primary point of this change is to allow external updates, for example by a VO writing the fullscreen option if the window state is changed by the window manager (rather than mpv changing it). This is not used yet, but the following commits will.
2019-11-29 11:49:15 +00:00
// This will force a self-notification back to config->option_change_callback.
void m_config_notify_change_opt_ptr(struct m_config *config, void *ptr);
// Return all (visible) option names as NULL terminated string list.
char **m_config_list_options(void *ta_parent, const struct m_config *config);
void m_config_print_option_list(const struct m_config *config, const char *name);
/* Find the profile with the given name.
* \param config The config object.
* \param arg The profile's name.
* \return The profile object or NULL.
*/
struct m_profile *m_config_get_profile0(const struct m_config *config,
char *name);
struct m_profile *m_config_get_profile(const struct m_config *config, bstr name);
// Apply and clear the default profile - it's the only profile that new config
// files do not simply append to (for configfile parser).
void m_config_finish_default_profile(struct m_config *config, int flags);
/* Get the profile with the given name, creating it if necessary.
* \param config The config object.
* \param arg The profile's name.
* \return The profile object.
*/
struct m_profile *m_config_add_profile(struct m_config *config, char *name);
/* Set the description of a profile.
* Used by the config file parser when defining a profile.
*
* \param p The profile object.
* \param arg The profile's name.
*/
void m_profile_set_desc(struct m_profile *p, bstr desc);
/* Add an option to a profile.
* Used by the config file parser when defining a profile.
*
* \param config The config object.
* \param p The profile object.
* \param name The option's name.
* \param val The option's value.
*/
int m_config_set_profile_option(struct m_config *config, struct m_profile *p,
bstr name, bstr val);
/* Enables profile usage
* Used by the config file parser when loading a profile.
*
* \param config The config object.
* \param p The profile object.
* \param flags M_SETOPT_* bits
* Returns error code (<0) or 0 on success
*/
int m_config_set_profile(struct m_config *config, char *name, int flags);
struct mpv_node m_config_get_profiles(struct m_config *config);
player: change m_config to use new option handling mechanisms Instead of making m_config a special-case, it more or less uses the underlying m_config_cache/m_config_shadow APIs properly. This makes the player core a (relatively) equivalent user of the core option API. In particular, this means that other threads can change core options with m_config_cache_write_opt() calls (before this commit, this merely led to diverging option values). An important change is that before this commit, mpctx->opts contained the "master copy" of all option data. Now it's just another copy of the option data, and the shadow copy is considered the master. This is why whenever mpctx->opts is written, the change needs to be copied to the master (thus why this commits add a bunch of m_config_notify... calls). If another thread (e.g. a VO) changes an option, async_change_cb is now invoked, which funnels the change notification through the player's layers. The new self_notification parameter on mp_option_change_callback is so that m_config_notify... doesn't trigger recursion, and it's used in cases where the change was already "processed". It's still needed to trigger libmpv property updates. (I considered using an extra m_config_cache for that, but it'd only cause problems with no advantages.) I think the recent changes actually forgot to send libmpv property updates in some cases. This should fix this anyway. In some cases, property updates are reworked, and the potential for bugs should be lower (probably). The primary point of this change is to allow external updates, for example by a VO writing the fullscreen option if the window state is changed by the window manager (rather than mpv changing it). This is not used yet, but the following commits will.
2019-11-29 11:49:15 +00:00
// Run async option updates here. This will call option_change_callback() on it.
void m_config_set_update_dispatch_queue(struct m_config *config,
struct mp_dispatch_queue *dispatch);
// This can be used to create and synchronize per-thread option structs,
// which then can be read without synchronization. No concurrent access to
// the cache itself is allowed.
struct m_config_cache {
// The struct as indicated by m_config_cache_alloc's group parameter.
// (Internally the same as internal->gdata[0]->udata.)
void *opts;
2019-11-28 20:19:41 +00:00
// Accumulated change flags. The user can set this to 0 to unset all flags.
// They are set when calling any of the update functions. A flag is only set
// once the new value is visible in ->opts.
uint64_t change_flags;
// Set to non-NULL for logging all option changes as they are retrieved
// with one of the update functions (like m_config_cache_update()).
struct mp_log *debug;
// Do not access.
struct config_cache *internal;
};
// Create a mirror copy from the global options.
options: add a thread-safe way to notify option updates So far, we had a thread-safe way to read options, but no option update notification mechanism. Everything was funneled though the main thread's central mp_option_change_callback() function. For example, if the panscan options were changed, the function called vo_control() with VOCTRL_SET_PANSCAN to manually notify the VO thread of updates. This worked, but's pretty inconvenient. Most of these problems come from the fact that MPlayer was written as a single-threaded program. This commit works towards a more flexible mechanism. It adds an update callback to m_config_cache (the thing that is already used for thread-safe access of global options). This alone would still be rather inconvenient, at least in context of VOs. Add another mechanism on top of it that uses mp_dispatch_queue, and takes care of some annoying synchronization issues. We extend mp_dispatch_queue itself to make this easier and slightly more efficient. As a first application, use this to reimplement certain VO scaling and renderer options. The update_opts() function translates these to the "old" VOCTRLs, though. An annoyingly subtle issue is that m_config_cache's destructor now releases pending notifications, and must be released before the associated dispatch queue. Otherwise, it could happen that option updates during e.g. VO destruction queue or run stale entries, which is not expected. Rather untested. The singly-linked list code in dispatch.c is probably buggy, and I bet some aspects about synchronization are not entirely sane.
2017-08-22 13:50:33 +00:00
// Keep in mind that a m_config_cache object is not thread-safe; it merely
// provides thread-safe access to the global options. All API functions for
// the same m_config_cache object must synchronized, unless otherwise noted.
2019-11-28 20:19:41 +00:00
// This does not create an initial change event (m_config_cache_update() will
// return false), but note that a change might be asynchronously signaled at any
// time.
// ta_parent: parent for the returned allocation
// global: option data source
// group: the option group to return
struct m_config_cache *m_config_cache_alloc(void *ta_parent,
struct mpv_global *global,
const struct m_sub_options *group);
options: add a thread-safe way to notify option updates So far, we had a thread-safe way to read options, but no option update notification mechanism. Everything was funneled though the main thread's central mp_option_change_callback() function. For example, if the panscan options were changed, the function called vo_control() with VOCTRL_SET_PANSCAN to manually notify the VO thread of updates. This worked, but's pretty inconvenient. Most of these problems come from the fact that MPlayer was written as a single-threaded program. This commit works towards a more flexible mechanism. It adds an update callback to m_config_cache (the thing that is already used for thread-safe access of global options). This alone would still be rather inconvenient, at least in context of VOs. Add another mechanism on top of it that uses mp_dispatch_queue, and takes care of some annoying synchronization issues. We extend mp_dispatch_queue itself to make this easier and slightly more efficient. As a first application, use this to reimplement certain VO scaling and renderer options. The update_opts() function translates these to the "old" VOCTRLs, though. An annoyingly subtle issue is that m_config_cache's destructor now releases pending notifications, and must be released before the associated dispatch queue. Otherwise, it could happen that option updates during e.g. VO destruction queue or run stale entries, which is not expected. Rather untested. The singly-linked list code in dispatch.c is probably buggy, and I bet some aspects about synchronization are not entirely sane.
2017-08-22 13:50:33 +00:00
// If any of the options in the group possibly changes, call this callback. The
// callback must not actually access the cache or anything option related.
// Instead, it must wake up the thread that normally accesses the cache.
void m_config_cache_set_wakeup_cb(struct m_config_cache *cache,
void (*cb)(void *ctx), void *cb_ctx);
// If any of the options in the group change, call this callback on the given
// dispatch queue. This is higher level than m_config_cache_set_wakeup_cb(),
// and you can do anything you want in the callback (assuming the dispatch
// queue is processed in the same thread that accesses m_config_cache API).
// To ensure clean shutdown, you must destroy the m_config_cache (or unset the
// callback) before the dispatch queue is destroyed.
void m_config_cache_set_dispatch_change_cb(struct m_config_cache *cache,
struct mp_dispatch_queue *dispatch,
void (*cb)(void *ctx), void *cb_ctx);
// Update the options in cache->opts to current global values. Return whether
// there was an update notification at all (which may or may not indicate that
// some options have changed).
// Keep in mind that while the cache->opts pointer does not change, the option
// data itself will (e.g. string options might be reallocated).
2019-11-28 20:19:41 +00:00
// New change flags are or-ed into cache->change_flags with this call (if you
// use them, you should probably do cache->change_flags=0 before this call).
bool m_config_cache_update(struct m_config_cache *cache);
2019-11-28 20:19:41 +00:00
// Check for changes and return fine grained change information.
// Warning: this conflicts with m_config_cache_update(). If you call
// m_config_cache_update(), all options will be marked as "not changed",
// and this function will return false. Also, calling this function and
// then m_config_cache_update() is not supported, and may skip updating
// some fields.
// This returns true as long as there is a changed option, and false if all
// changed options have been returned.
// If multiple options have changed, the new option value is visible only once
// this function has returned the change for it.
// out_ptr: pointer to a void*, which is set to the cache->opts field associated
// with the changed option if the function returns true; set to NULL
// if no option changed.
// returns: *out_ptr!=NULL (true if there was a changed option)
bool m_config_cache_get_next_changed(struct m_config_cache *cache, void **out_ptr);
// Copy the option field pointed to by ptr to the global option storage. This
// is sort of similar to m_config_set_option_raw(), except doesn't require
// access to the main thread. (And you can't pass any flags.)
// You write the new value to the option struct, and then call this function
// with the pointer to it. You will not get a change notification for it (though
// you might still get a redundant wakeup callback).
// Changing the option struct and not calling this function before any update
// function (like m_config_cache_update()) will leave the value inconsistent,
// and will possibly (but not necessarily) overwrite it with the next update
// call.
// ptr: points to any field in cache->opts that is managed by an option. If
// this is not the case, the function crashes for your own good.
// returns: if true, this was an update; if false, shadow had same value
bool m_config_cache_write_opt(struct m_config_cache *cache, void *ptr);
// Like m_config_cache_alloc(), but return the struct (m_config_cache->opts)
options: add a thread-safe way to notify option updates So far, we had a thread-safe way to read options, but no option update notification mechanism. Everything was funneled though the main thread's central mp_option_change_callback() function. For example, if the panscan options were changed, the function called vo_control() with VOCTRL_SET_PANSCAN to manually notify the VO thread of updates. This worked, but's pretty inconvenient. Most of these problems come from the fact that MPlayer was written as a single-threaded program. This commit works towards a more flexible mechanism. It adds an update callback to m_config_cache (the thing that is already used for thread-safe access of global options). This alone would still be rather inconvenient, at least in context of VOs. Add another mechanism on top of it that uses mp_dispatch_queue, and takes care of some annoying synchronization issues. We extend mp_dispatch_queue itself to make this easier and slightly more efficient. As a first application, use this to reimplement certain VO scaling and renderer options. The update_opts() function translates these to the "old" VOCTRLs, though. An annoyingly subtle issue is that m_config_cache's destructor now releases pending notifications, and must be released before the associated dispatch queue. Otherwise, it could happen that option updates during e.g. VO destruction queue or run stale entries, which is not expected. Rather untested. The singly-linked list code in dispatch.c is probably buggy, and I bet some aspects about synchronization are not entirely sane.
2017-08-22 13:50:33 +00:00
// directly, with no way to update the config. Basically this returns a copy
// with a snapshot of the current option values.
void *mp_get_config_group(void *ta_parent, struct mpv_global *global,
const struct m_sub_options *group);
// Read a single global option in a thread-safe way. For multiple options,
// use m_config_cache. The option must exist and match the provided type (the
// type is used as a sanity check only). Performs semi-expensive lookup.
// Warning: new code must not use this.
void mp_read_option_raw(struct mpv_global *global, const char *name,
const struct m_option_type *type, void *dst);
#endif /* MPLAYER_M_CONFIG_H */