mpv/options/m_option.h

766 lines
28 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_OPTION_H
#define MPLAYER_M_OPTION_H
#include <float.h>
#include <string.h>
#include <stddef.h>
#include <stdbool.h>
#include "misc/bstr.h"
#include "audio/chmap.h"
#include "common/common.h"
// m_option allows to parse, print and copy data of various types.
typedef struct m_option_type m_option_type_t;
typedef struct m_option m_option_t;
struct m_config;
struct mp_log;
struct mpv_node;
struct mpv_global;
///////////////////////////// Options types declarations ////////////////////
// Simple types
options: introduce bool option type, use it for --fullscreen The option code is very old and was added to MPlayer in the early 2000s, when C99 was still new. MPlayer did not use the "bool" type anywhere,l and the logical option equivalent to bool, the "flag" option type, used int, with the convention that only the values 0 and 1 are allowed. mpv may have hammered many, many additional tentacles to the option code, but some of the basics never changed, and m_option_type_flag still uses int. This seems a bit weird, since mpv uses bool for booleans. So finally introduce an m_option_type_bool. To avoid duplicating too much code, change the flag code to bool, and "reimplement" m_option_type_flag on top of m_option_type_bool. As a "demonstration", change the --fullscreen option to this new type. Ideally, all options would be changed too bool, and m_option_type_flag would be removed. But that is a lot of monotonous thankless work, so I'm not doing it, and making it a painful years long transition. At the same time, I'm introducing a new concept for option declarations. Instead of OPT_BOOL(), which define the full m_option struct contents, there's OPTF_BOOL(), which only takes the option field name itself. The name is provided via a normal struct field initializer. Other fields (such as flags) can be provided via designated initializers. The advantage of this is that we don't need tons of nested vararg macros. We also don't need to deal with 0-sized varargs being a pain (and in fact they are not a thing in standard C99 and probably C11). There is no need to provide a mandatory flags argument either, which is the reason why so many OPT_ macros are used with a "0" argument. (The flag argument seems to confuse other developers; they either don't immediately recognize what it is, and sometimes it's supposed to be the option's default value.) Not having to mess with the flag argument in such option macros is also a reason for the removal of M_OPT_RANGE etc., for the better or worse. The only place that special-cased the _flag option type was in command.c; change it to use something effectively very similar that automatically includes the new _bool option type. Everything else should be transparent to the change. The fullscreen option change should be transparent too, as C99 bool is basically an integer type that is clamped to 0/1 (except in Swift, Swift sucks).
2020-03-14 01:07:35 +00:00
extern const m_option_type_t m_option_type_bool;
extern const m_option_type_t m_option_type_flag;
extern const m_option_type_t m_option_type_dummy_flag;
extern const m_option_type_t m_option_type_int;
extern const m_option_type_t m_option_type_int64;
extern const m_option_type_t m_option_type_byte_size;
extern const m_option_type_t m_option_type_float;
extern const m_option_type_t m_option_type_double;
extern const m_option_type_t m_option_type_string;
extern const m_option_type_t m_option_type_string_list;
extern const m_option_type_t m_option_type_string_append_list;
extern const m_option_type_t m_option_type_keyvalue_list;
extern const m_option_type_t m_option_type_time;
extern const m_option_type_t m_option_type_rel_time;
extern const m_option_type_t m_option_type_choice;
2015-03-04 16:20:47 +00:00
extern const m_option_type_t m_option_type_flags;
extern const m_option_type_t m_option_type_msglevels;
extern const m_option_type_t m_option_type_print_fn;
extern const m_option_type_t m_option_type_imgfmt;
video: decouple internal pixel formats from FourCCs mplayer's video chain traditionally used FourCCs for pixel formats. For example, it used IMGFMT_YV12 for 4:2:0 YUV, which was defined to the string 'YV12' interpreted as unsigned int. Additionally, it used to encode information into the numeric values of some formats. The RGB formats had their bit depth and endian encoded into the least significant byte. Extended planar formats (420P10 etc.) had chroma shift, endian, and component bit depth encoded. (This has been removed in recent commits.) Replace the FourCC mess with a simple enum. Remove all the redundant formats like YV12/I420/IYUV. Replace some image format names by something more intuitive, most importantly IMGFMT_YV12 -> IMGFMT_420P. Add img_fourcc.h, which contains the old IDs for code that actually uses FourCCs. Change the way demuxers, that output raw video, identify the video format: they set either MP_FOURCC_RAWVIDEO or MP_FOURCC_IMGFMT to request the rawvideo decoder, and sh_video->imgfmt specifies the pixel format. Like the previous hack, this is supposed to avoid the need for a complete codecs.cfg entry per format, or other lookup tables. (Note that the RGB raw video FourCCs mostly rely on ffmpeg's mappings for NUT raw video, but this is still considered better than adding a raw video decoder - even if trivial, it would be full of annoying lookup tables.) The TV code has not been tested. Some corrective changes regarding endian and other image format flags creep in.
2012-12-23 19:03:30 +00:00
extern const m_option_type_t m_option_type_fourcc;
extern const m_option_type_t m_option_type_afmt;
extern const m_option_type_t m_option_type_color;
extern const m_option_type_t m_option_type_geometry;
extern const m_option_type_t m_option_type_size_box;
extern const m_option_type_t m_option_type_channels;
extern const m_option_type_t m_option_type_aspect;
extern const m_option_type_t m_option_type_obj_settings_list;
extern const m_option_type_t m_option_type_node;
extern const m_option_type_t m_option_type_rect;
// Used internally by m_config.c
extern const m_option_type_t m_option_type_alias;
extern const m_option_type_t m_option_type_cli_alias;
extern const m_option_type_t m_option_type_removed;
extern const m_option_type_t m_option_type_subconfig;
// Callback used by m_option_type_print_fn options.
typedef void (*m_opt_print_fn)(struct mp_log *log);
enum m_rel_time_type {
REL_TIME_NONE,
REL_TIME_ABSOLUTE,
REL_TIME_RELATIVE,
REL_TIME_PERCENT,
REL_TIME_CHAPTER,
};
struct m_rel_time {
double pos;
enum m_rel_time_type type;
};
struct m_color {
uint8_t r, g, b, a;
};
struct m_geometry {
int x, y, w, h;
bool xy_valid : 1, wh_valid : 1;
bool w_per : 1, h_per : 1;
bool x_sign : 1, y_sign : 1, x_per : 1, y_per : 1;
int ws; // workspace; valid if !=0
};
void m_geometry_apply(int *xpos, int *ypos, int *widw, int *widh,
int scrw, int scrh, struct m_geometry *gm);
void m_rect_apply(struct mp_rect *rc, int w, int h, struct m_geometry *gm);
struct m_channels {
bool set : 1;
bool auto_safe : 1;
struct mp_chmap *chmaps;
int num_chmaps;
};
struct m_obj_desc {
// Name which will be used in the option string
const char *name;
// Will be printed when "help" is passed
const char *description;
// Size of the private struct
int priv_size;
// If not NULL, default values for private struct
const void *priv_defaults;
// Options which refer to members in the private struct
const struct m_option *options;
// Prefix for each of the above options (none if NULL).
const char *options_prefix;
// For free use by the implementer of m_obj_list.get_desc
const void *p;
// Don't list entry with "help"
bool hidden;
// Callback to print custom help if "vf=entry=help" is passed
void (*print_help)(struct mp_log *log);
// Set by m_obj_list_find(). If the requested name is an old alias, this
// is set to the old name (while the name field uses the new name).
const char *replaced_name;
// For convenience: these are added as global command-line options.
const struct m_sub_options *global_opts;
};
// Extra definition needed for \ref m_option_type_obj_settings_list options.
struct m_obj_list {
bool (*get_desc)(struct m_obj_desc *dst, int index);
const char *description;
// Can be set to a NULL terminated array of aliases
const char *aliases[5][2];
// Allow a trailing ",", which adds an entry with name=""
bool allow_trailer;
// Callback to test whether an unknown entry should be allowed. (This can
// be useful if adding them as explicit entries is too much work.)
bool (*check_unknown_entry)(const char *name);
// Allow syntax for disabling entries.
bool allow_disable_entries;
// This helps with confusing error messages if unknown flag options are used.
bool disallow_positional_parameters;
// Each sub-item is backed by global options (for AOs and VOs).
bool use_global_options;
// Callback to print additional custom help if "vf=help" is passed
void (*print_help_list)(struct mp_log *log);
// Callback to print help for _unknown_ entries with "vf=entry=help"
void (*print_unknown_entry_help)(struct mp_log *log, const char *name);
};
// Find entry by name
bool m_obj_list_find(struct m_obj_desc *dst, const struct m_obj_list *list,
bstr name);
// The data type used by \ref m_option_type_obj_settings_list.
typedef struct m_obj_settings {
// Type of the object.
char *name;
// Optional user-defined name.
char *label;
// User enable flag.
bool enabled;
// NULL terminated array of parameter/value pairs.
char **attribs;
} m_obj_settings_t;
bool m_obj_settings_equal(struct m_obj_settings *a, struct m_obj_settings *b);
struct m_opt_choice_alternatives {
char *name;
int value;
};
const char *m_opt_choice_str(const struct m_opt_choice_alternatives *choices,
int value);
options: Make validation and help possible for all option types Today, validation is only possible for string type options. But there's no particular reason why it needs to be restricted in this way, and there are potential uses, to allow other options to be validated without forcing the option to have to reimplement parsing from scratch. The first part, simply making the validation function an explicit field instead of overloading priv is simple enough. But if we only do that, then the validation function still needs to deal with the raw pre-parsed string. Instead, we want to allow the value to be parsed before it is validated. That in turn leads to us having validator functions that should be type aware. Unfortunately, that means we need to keep the explicit macro like OPT_STRING_VALIDATE() as a way to enforce the correct typing of the function. Otherwise, we'd have to have the validator take a void * and hope the implementation can cast it correctly. For help, we don't have this problem, as help doesn't look at the value. Then, we turn validators that are really help generators into explicit help functions and where a validator is help + validation, we split them into two parts. I have, however, left functions that need to query information for both help and validation as single functions to avoid code duplication. In this change, I have not added an other OPT_FOO_VALIDATE() macros as they are not needed, but I will add some in a separate change to illustrate the pattern.
2021-02-21 00:41:44 +00:00
// Validator function signatures. Required to properly type the param value.
typedef int (*m_opt_generic_validate_fn)(struct mp_log *log, const m_option_t *opt,
struct bstr name, void *value);
typedef int (*m_opt_string_validate_fn)(struct mp_log *log, const m_option_t *opt,
options: Make validation and help possible for all option types Today, validation is only possible for string type options. But there's no particular reason why it needs to be restricted in this way, and there are potential uses, to allow other options to be validated without forcing the option to have to reimplement parsing from scratch. The first part, simply making the validation function an explicit field instead of overloading priv is simple enough. But if we only do that, then the validation function still needs to deal with the raw pre-parsed string. Instead, we want to allow the value to be parsed before it is validated. That in turn leads to us having validator functions that should be type aware. Unfortunately, that means we need to keep the explicit macro like OPT_STRING_VALIDATE() as a way to enforce the correct typing of the function. Otherwise, we'd have to have the validator take a void * and hope the implementation can cast it correctly. For help, we don't have this problem, as help doesn't look at the value. Then, we turn validators that are really help generators into explicit help functions and where a validator is help + validation, we split them into two parts. I have, however, left functions that need to query information for both help and validation as single functions to avoid code duplication. In this change, I have not added an other OPT_FOO_VALIDATE() macros as they are not needed, but I will add some in a separate change to illustrate the pattern.
2021-02-21 00:41:44 +00:00
struct bstr name, const char **value);
typedef int (*m_opt_int_validate_fn)(struct mp_log *log, const m_option_t *opt,
struct bstr name, const int *value);
// m_option.priv points to this if OPT_SUBSTRUCT is used
struct m_sub_options {
const char *prefix;
const struct m_option *opts;
size_t size;
const void *defaults;
// Change flags passed to mp_option_change_callback() if any option that is
// directly or indirectly part of this group is changed.
int change_flags;
// Return further sub-options, for example for optional components. If set,
// this is called with increasing index (starting from 0), as long as true
// is returned. If true is returned and *sub is set in any of these calls,
// they are added as options.
bool (*get_sub_options)(int index, const struct m_sub_options **sub);
};
#define CONF_TYPE_BOOL (&m_option_type_bool)
#define CONF_TYPE_FLAG (&m_option_type_flag)
#define CONF_TYPE_INT (&m_option_type_int)
#define CONF_TYPE_INT64 (&m_option_type_int64)
#define CONF_TYPE_FLOAT (&m_option_type_float)
#define CONF_TYPE_DOUBLE (&m_option_type_double)
#define CONF_TYPE_STRING (&m_option_type_string)
#define CONF_TYPE_STRING_LIST (&m_option_type_string_list)
#define CONF_TYPE_IMGFMT (&m_option_type_imgfmt)
video: decouple internal pixel formats from FourCCs mplayer's video chain traditionally used FourCCs for pixel formats. For example, it used IMGFMT_YV12 for 4:2:0 YUV, which was defined to the string 'YV12' interpreted as unsigned int. Additionally, it used to encode information into the numeric values of some formats. The RGB formats had their bit depth and endian encoded into the least significant byte. Extended planar formats (420P10 etc.) had chroma shift, endian, and component bit depth encoded. (This has been removed in recent commits.) Replace the FourCC mess with a simple enum. Remove all the redundant formats like YV12/I420/IYUV. Replace some image format names by something more intuitive, most importantly IMGFMT_YV12 -> IMGFMT_420P. Add img_fourcc.h, which contains the old IDs for code that actually uses FourCCs. Change the way demuxers, that output raw video, identify the video format: they set either MP_FOURCC_RAWVIDEO or MP_FOURCC_IMGFMT to request the rawvideo decoder, and sh_video->imgfmt specifies the pixel format. Like the previous hack, this is supposed to avoid the need for a complete codecs.cfg entry per format, or other lookup tables. (Note that the RGB raw video FourCCs mostly rely on ffmpeg's mappings for NUT raw video, but this is still considered better than adding a raw video decoder - even if trivial, it would be full of annoying lookup tables.) The TV code has not been tested. Some corrective changes regarding endian and other image format flags creep in.
2012-12-23 19:03:30 +00:00
#define CONF_TYPE_FOURCC (&m_option_type_fourcc)
#define CONF_TYPE_AFMT (&m_option_type_afmt)
#define CONF_TYPE_OBJ_SETTINGS_LIST (&m_option_type_obj_settings_list)
#define CONF_TYPE_TIME (&m_option_type_time)
#define CONF_TYPE_CHOICE (&m_option_type_choice)
#define CONF_TYPE_NODE (&m_option_type_node)
// Possible option values. Code is allowed to access option data without going
// through this union. It serves for self-documentation and to get minimal
// size/alignment requirements for option values in general.
union m_option_value {
options: introduce bool option type, use it for --fullscreen The option code is very old and was added to MPlayer in the early 2000s, when C99 was still new. MPlayer did not use the "bool" type anywhere,l and the logical option equivalent to bool, the "flag" option type, used int, with the convention that only the values 0 and 1 are allowed. mpv may have hammered many, many additional tentacles to the option code, but some of the basics never changed, and m_option_type_flag still uses int. This seems a bit weird, since mpv uses bool for booleans. So finally introduce an m_option_type_bool. To avoid duplicating too much code, change the flag code to bool, and "reimplement" m_option_type_flag on top of m_option_type_bool. As a "demonstration", change the --fullscreen option to this new type. Ideally, all options would be changed too bool, and m_option_type_flag would be removed. But that is a lot of monotonous thankless work, so I'm not doing it, and making it a painful years long transition. At the same time, I'm introducing a new concept for option declarations. Instead of OPT_BOOL(), which define the full m_option struct contents, there's OPTF_BOOL(), which only takes the option field name itself. The name is provided via a normal struct field initializer. Other fields (such as flags) can be provided via designated initializers. The advantage of this is that we don't need tons of nested vararg macros. We also don't need to deal with 0-sized varargs being a pain (and in fact they are not a thing in standard C99 and probably C11). There is no need to provide a mandatory flags argument either, which is the reason why so many OPT_ macros are used with a "0" argument. (The flag argument seems to confuse other developers; they either don't immediately recognize what it is, and sometimes it's supposed to be the option's default value.) Not having to mess with the flag argument in such option macros is also a reason for the removal of M_OPT_RANGE etc., for the better or worse. The only place that special-cased the _flag option type was in command.c; change it to use something effectively very similar that automatically includes the new _bool option type. Everything else should be transparent to the change. The fullscreen option change should be transparent too, as C99 bool is basically an integer type that is clamped to 0/1 (except in Swift, Swift sucks).
2020-03-14 01:07:35 +00:00
bool bool_;
int flag; // not the C type "bool"!
int int_;
int64_t int64;
float float_;
double double_;
char *string;
char **string_list;
char **keyvalue_list;
int imgfmt;
video: decouple internal pixel formats from FourCCs mplayer's video chain traditionally used FourCCs for pixel formats. For example, it used IMGFMT_YV12 for 4:2:0 YUV, which was defined to the string 'YV12' interpreted as unsigned int. Additionally, it used to encode information into the numeric values of some formats. The RGB formats had their bit depth and endian encoded into the least significant byte. Extended planar formats (420P10 etc.) had chroma shift, endian, and component bit depth encoded. (This has been removed in recent commits.) Replace the FourCC mess with a simple enum. Remove all the redundant formats like YV12/I420/IYUV. Replace some image format names by something more intuitive, most importantly IMGFMT_YV12 -> IMGFMT_420P. Add img_fourcc.h, which contains the old IDs for code that actually uses FourCCs. Change the way demuxers, that output raw video, identify the video format: they set either MP_FOURCC_RAWVIDEO or MP_FOURCC_IMGFMT to request the rawvideo decoder, and sh_video->imgfmt specifies the pixel format. Like the previous hack, this is supposed to avoid the need for a complete codecs.cfg entry per format, or other lookup tables. (Note that the RGB raw video FourCCs mostly rely on ffmpeg's mappings for NUT raw video, but this is still considered better than adding a raw video decoder - even if trivial, it would be full of annoying lookup tables.) The TV code has not been tested. Some corrective changes regarding endian and other image format flags creep in.
2012-12-23 19:03:30 +00:00
unsigned int fourcc;
int afmt;
m_obj_settings_t *obj_settings_list;
double time;
struct m_rel_time rel_time;
struct m_color color;
struct m_geometry geometry;
struct m_geometry size_box;
struct m_channels channels;
};
// Keep fully zeroed instance of m_option_value to use as a default value, before
// any specific union member is used. C standard says that `= {0}` activates and
// initializes only the first member of the union, leaving padding bits undefined.
static const union m_option_value m_option_value_default;
////////////////////////////////////////////////////////////////////////////
struct m_option_action {
// The name of the suffix, e.g. "add" for a list. If the option is named
// "foo", this will be available as "--foo-add". Note that no suffix (i.e.
// "--foo" is implicitly always available.
const char *name;
// One of M_OPT_TYPE*.
unsigned int flags;
};
// Option type description
struct m_option_type {
const char *name;
// Size needed for the data.
unsigned int size;
options: get rid of ambiguous option parsing Options parsing used to be ambiguous, as in the splitting into option and values pairs was ambiguous. Example: -option -something It wasn't clear whether -option actually takes an argument or not. The string "-something" could either be a separate option, or an argument to "-option". The code had to call the option specific parser function to resolve this. This made everything complicated and didn't even have a real use. There was only one case where this was actually used: string lists (m_option_type_string_list) and options based on it. That is because this option type actually turns a single option into a proxy for several real arguments, e.g. "vf*" can handle "-vf-add" and "-vf-clr". Options suffixed with "-clr" are the only options of this group which take no arguments. This is ambiguous only with the "old syntax" (as shown above). The "new" option syntax always puts option name and value into same argument. (E.g. "--option=--something" or "--option" "--something".) Simplify the code by making it statically known whether an option takes a parameter or not with the flag M_OPT_TYPE_OLD_SYNTAX_NO_PARAM. If it's set, the option parser assumes the option takes no argument. The only real ambiguity left, string list options that end on "-clr", are special cased in the parser. Remove some duplication of the logic in the command line parser by moving all argument splitting logic into split_opt(). (It's arguable whether that can be considered code duplication, but now the code is a bit simpler anyway. This might be subjective.) Remove the "ambiguous" parameter from all option parsing related code. Make m_config unaware of the pre-parsing concept. Make most CONF_NOCFG options also CONF_GLOBAL (except those explicitly usable as per-file options.)
2012-08-05 21:34:28 +00:00
// One of M_OPT_TYPE*.
unsigned int flags;
// Parse the data from a string.
/** It is the only required function, all others can be NULL.
options: Make validation and help possible for all option types Today, validation is only possible for string type options. But there's no particular reason why it needs to be restricted in this way, and there are potential uses, to allow other options to be validated without forcing the option to have to reimplement parsing from scratch. The first part, simply making the validation function an explicit field instead of overloading priv is simple enough. But if we only do that, then the validation function still needs to deal with the raw pre-parsed string. Instead, we want to allow the value to be parsed before it is validated. That in turn leads to us having validator functions that should be type aware. Unfortunately, that means we need to keep the explicit macro like OPT_STRING_VALIDATE() as a way to enforce the correct typing of the function. Otherwise, we'd have to have the validator take a void * and hope the implementation can cast it correctly. For help, we don't have this problem, as help doesn't look at the value. Then, we turn validators that are really help generators into explicit help functions and where a validator is help + validation, we split them into two parts. I have, however, left functions that need to query information for both help and validation as single functions to avoid code duplication. In this change, I have not added an other OPT_FOO_VALIDATE() macros as they are not needed, but I will add some in a separate change to illustrate the pattern.
2021-02-21 00:41:44 +00:00
* Generally should not be called directly outside of the options module,
* but instead through \ref m_option_parse which calls additional option
* specific callbacks during the process.
*
* \param log for outputting parser error or help messages
* \param opt The option that is parsed.
* \param name The full option name.
* \param param The parameter to parse.
* may not be an argument meant for this option
* \param dst Pointer to the memory where the data should be written.
* If NULL the parameter validity should still be checked.
* \return On error a negative value is returned, on success the number
* of arguments consumed. For details see \ref OptionParserReturn.
*/
int (*parse)(struct mp_log *log, const m_option_t *opt,
struct bstr name, struct bstr param, void *dst);
// Print back a value in string form.
/** \param opt The option to print.
* \param val Pointer to the memory holding the data to be printed.
* \return An allocated string containing the text value or (void*)-1
* on error.
*/
char *(*print)(const m_option_t *opt, const void *val);
// Print the value in a human readable form. Unlike print(), it doesn't
// necessarily return the exact value, and is generally not parseable with
// parse().
char *(*pretty_print)(const m_option_t *opt, const void *val);
// Copy data between two locations. Deep copy if the data has pointers.
// The implementation must free *dst if memory allocation is involved.
/** \param opt The option to copy.
* \param dst Pointer to the destination memory.
* \param src Pointer to the source memory.
*/
void (*copy)(const m_option_t *opt, void *dst, const void *src);
// Free the data allocated for a save slot.
/** This is only needed for dynamic types like strings.
* \param dst Pointer to the data, usually a pointer that should be freed and
* set to NULL.
*/
void (*free)(void *dst);
// Add the value add to the value in val. For types that are not numeric,
// add gives merely the direction. The wrap parameter determines whether
// the value is clipped, or wraps around to the opposite max/min.
void (*add)(const m_option_t *opt, void *val, double add, bool wrap);
// Multiply the value with the factor f. The callback must clip the result
// to the valid value range of the option.
void (*multiply)(const m_option_t *opt, void *val, double f);
// Set the option value in dst to the contents of src.
// (If the option is dynamic, the old value in *dst has to be freed.)
// Return values:
// M_OPT_UNKNOWN: src is in an unknown format
// M_OPT_INVALID: src is incorrectly formatted
// >= 0: success
// other error code: some other error, essentially M_OPT_INVALID refined
int (*set)(const m_option_t *opt, void *dst, struct mpv_node *src);
// Copy the option value in src to dst. Use ta_parent for any dynamic
// memory allocations. It's explicitly allowed to have mpv_node reference
// static strings (and even mpv_node_list.keys), though.
int (*get)(const m_option_t *opt, void *ta_parent, struct mpv_node *dst,
void *src);
// Return whether the values are the same. (There are no "unordered"
// results; for example, two floats with the value NaN compare equal. Other
// ambiguous floats, such as +0 and -0 compare equal. Some option types may
// incorrectly report unequal for values that are equal, such as sets (if
// the element order is different, which incorrectly matters), but values
// duplicated with m_option_copy() always return as equal. Empty strings
// and NULL strings are equal. Ambiguous unicode representations compare
// unequal.)
// If not set, values are always considered equal (=> not really optional).
bool (*equal)(const m_option_t *opt, void *a, void *b);
// Optional: list of suffixes, terminated with a {0} entry. An empty list
// behaves like the list being NULL.
const struct m_option_action *actions;
};
// Option description
struct m_option {
// Option name.
options: introduce bool option type, use it for --fullscreen The option code is very old and was added to MPlayer in the early 2000s, when C99 was still new. MPlayer did not use the "bool" type anywhere,l and the logical option equivalent to bool, the "flag" option type, used int, with the convention that only the values 0 and 1 are allowed. mpv may have hammered many, many additional tentacles to the option code, but some of the basics never changed, and m_option_type_flag still uses int. This seems a bit weird, since mpv uses bool for booleans. So finally introduce an m_option_type_bool. To avoid duplicating too much code, change the flag code to bool, and "reimplement" m_option_type_flag on top of m_option_type_bool. As a "demonstration", change the --fullscreen option to this new type. Ideally, all options would be changed too bool, and m_option_type_flag would be removed. But that is a lot of monotonous thankless work, so I'm not doing it, and making it a painful years long transition. At the same time, I'm introducing a new concept for option declarations. Instead of OPT_BOOL(), which define the full m_option struct contents, there's OPTF_BOOL(), which only takes the option field name itself. The name is provided via a normal struct field initializer. Other fields (such as flags) can be provided via designated initializers. The advantage of this is that we don't need tons of nested vararg macros. We also don't need to deal with 0-sized varargs being a pain (and in fact they are not a thing in standard C99 and probably C11). There is no need to provide a mandatory flags argument either, which is the reason why so many OPT_ macros are used with a "0" argument. (The flag argument seems to confuse other developers; they either don't immediately recognize what it is, and sometimes it's supposed to be the option's default value.) Not having to mess with the flag argument in such option macros is also a reason for the removal of M_OPT_RANGE etc., for the better or worse. The only place that special-cased the _flag option type was in command.c; change it to use something effectively very similar that automatically includes the new _bool option type. Everything else should be transparent to the change. The fullscreen option change should be transparent too, as C99 bool is basically an integer type that is clamped to 0/1 (except in Swift, Swift sucks).
2020-03-14 01:07:35 +00:00
// Option declarations can use this as positional field.
const char *name;
// Option type.
const m_option_type_t *type;
// See \ref OptionFlags.
unsigned int flags;
int offset;
// Most numeric types restrict the range to [min, max] if min<max (this
// implies that if min/max are not set, the full range is used). In all
// cases, the actual range is clamped to the type's native range.
// Float types use [DBL_MIN, DBL_MAX], though by setting min or max to
// -/+INFINITY, the range can be extended to INFINITY. (This part is buggy
// for "float".)
// Preferably use M_RANGE() to set these fields.
double min, max;
// Type dependent data (for all kinds of extended settings).
void *priv;
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
// Initialize variable to given default before parsing options
const void *defval;
// Print a warning when this option is used (for options with no direct
// replacement.)
const char *deprecation_message;
options: Make validation and help possible for all option types Today, validation is only possible for string type options. But there's no particular reason why it needs to be restricted in this way, and there are potential uses, to allow other options to be validated without forcing the option to have to reimplement parsing from scratch. The first part, simply making the validation function an explicit field instead of overloading priv is simple enough. But if we only do that, then the validation function still needs to deal with the raw pre-parsed string. Instead, we want to allow the value to be parsed before it is validated. That in turn leads to us having validator functions that should be type aware. Unfortunately, that means we need to keep the explicit macro like OPT_STRING_VALIDATE() as a way to enforce the correct typing of the function. Otherwise, we'd have to have the validator take a void * and hope the implementation can cast it correctly. For help, we don't have this problem, as help doesn't look at the value. Then, we turn validators that are really help generators into explicit help functions and where a validator is help + validation, we split them into two parts. I have, however, left functions that need to query information for both help and validation as single functions to avoid code duplication. In this change, I have not added an other OPT_FOO_VALIDATE() macros as they are not needed, but I will add some in a separate change to illustrate the pattern.
2021-02-21 00:41:44 +00:00
// Optional function that validates a param value for this option.
m_opt_generic_validate_fn validate;
// Optional function that displays help. Will replace type-specific help.
int (*help)(struct mp_log *log, const m_option_t *opt, struct bstr name);
};
char *format_file_size(int64_t size);
// The option is forbidden in config files.
#define M_OPT_NOCFG (1 << 2)
// The option should be set during command line pre-parsing
#define M_OPT_PRE_PARSE (1 << 4)
// The option expects a file name (or a list of file names)
#define M_OPT_FILE (1 << 5)
// Do not add as property.
#define M_OPT_NOPROP (1 << 6)
// Enable special semantics for some options when parsing the string "help".
#define M_OPT_HAVE_HELP (1 << 7)
// The following are also part of the M_OPT_* flags, and are used to update
// certain groups of options.
2018-05-12 15:29:27 +00:00
#define UPDATE_OPT_FIRST (1 << 8)
#define UPDATE_TERM (1 << 8) // terminal options
sub: make filter_sdh a "proper" filter, allow runtime changes Until now, filter_sdh was simply a function that was called by sd_ass directly (if enabled). I want to add another filter, so it's time to turn this into a somewhat more general subtitle filtering infrastructure. I pondered whether to reuse the audio/video filtering stuff - but better not. Also, since subtitles are horrible and tend to refuse proper abstraction, it's still messed into sd_ass, instead of working on the dec_sub.c level. Actually mpv used to have subtitle "filters" and even made subtitle converters part of it, but it was fairly horrible, so don't do that again. In addition, make runtime changes possible. Since this was supposed to be a quick hack, I just decided to put all subtitle filter options into a separate option group (=> simpler change notification), to manually push the change through the playloop (like it was sort of before for OSD options), and to recreate the sub filter chain completely in every change. Should be good enough. One strangeness is that due to prefetching and such, most subtitle packets (or those some time ahead) are actually done filtering when we change, so the user still needs to manually seek to actually refresh everything. And since subtitle data is usually cached in ASS_Track (for other terrible but user-friendly reasons), we also must clear the subtitle data, but of course only on seek, since otherwise all subtitles would just disappear. What a fucking mess, but such is life. We could trigger a "refresh seek" to make this more automatic, but I don't feel like it currently. This is slightly inefficient (lots of allocations and copying), but I decided that it doesn't matter. Could matter slightly for crazy ASS subtitles that render with thousands of events. Not very well tested. Still seems to work, but I didn't have many test cases.
2020-02-16 00:02:17 +00:00
#define UPDATE_SUB_FILT (1 << 9) // subtitle filter options
#define UPDATE_OSD (1 << 10) // related to OSD rendering
#define UPDATE_BUILTIN_SCRIPTS (1 << 11) // osc/ytdl/stats
#define UPDATE_IMGPAR (1 << 12) // video image params overrides
#define UPDATE_INPUT (1 << 13) // mostly --input-* options
#define UPDATE_AUDIO (1 << 14) // --audio-channels etc.
#define UPDATE_PRIORITY (1 << 15) // --priority (Windows-only)
#define UPDATE_SCREENSAVER (1 << 16) // --stop-screensaver
#define UPDATE_VOL (1 << 17) // softvol related options
#define UPDATE_LAVFI_COMPLEX (1 << 18) // --lavfi-complex
#define UPDATE_HWDEC (1 << 20) // --hwdec
#define UPDATE_DVB_PROG (1 << 21) // some --dvbin-...
#define UPDATE_SUB_HARD (1 << 22) // subtitle opts. that need full reinit
#define UPDATE_SUB_EXTS (1 << 23) // update internal list of sub exts
#define UPDATE_VIDEO (1 << 24) // force redraw if needed
#define UPDATE_OPT_LAST (1 << 24)
// All bits between _FIRST and _LAST (inclusive)
#define UPDATE_OPTS_MASK \
(((UPDATE_OPT_LAST << 1) - 1) & ~(unsigned)(UPDATE_OPT_FIRST - 1))
// type_float/type_double: string "default" is parsed as NaN (and reverse)
#define M_OPT_DEFAULT_NAN (1 << 25)
// type time: string "no" maps to MP_NOPTS_VALUE (if unset, NOPTS is rejected)
#define M_OPT_ALLOW_NO (1 << 26)
// type channels: disallow "auto" (still accept ""), limit list to at most 1 item.
#define M_OPT_CHANNELS_LIMITED (1 << 27)
// Like M_OPT_TYPE_OPTIONAL_PARAM.
#define M_OPT_OPTIONAL_PARAM (1 << 30)
// These are kept for compatibility with older code.
#define CONF_NOCFG M_OPT_NOCFG
#define CONF_PRE_PARSE M_OPT_PRE_PARSE
// These flags are used to describe special parser capabilities or behavior.
// The parameter is optional and by default no parameter is preferred. If
// ambiguous syntax is used ("--opt value"), the command line parser will
// assume that the argument takes no parameter. In config files, these
// options can be used without "=" and value.
#define M_OPT_TYPE_OPTIONAL_PARAM (1 << 0)
options: get rid of ambiguous option parsing Options parsing used to be ambiguous, as in the splitting into option and values pairs was ambiguous. Example: -option -something It wasn't clear whether -option actually takes an argument or not. The string "-something" could either be a separate option, or an argument to "-option". The code had to call the option specific parser function to resolve this. This made everything complicated and didn't even have a real use. There was only one case where this was actually used: string lists (m_option_type_string_list) and options based on it. That is because this option type actually turns a single option into a proxy for several real arguments, e.g. "vf*" can handle "-vf-add" and "-vf-clr". Options suffixed with "-clr" are the only options of this group which take no arguments. This is ambiguous only with the "old syntax" (as shown above). The "new" option syntax always puts option name and value into same argument. (E.g. "--option=--something" or "--option" "--something".) Simplify the code by making it statically known whether an option takes a parameter or not with the flag M_OPT_TYPE_OLD_SYNTAX_NO_PARAM. If it's set, the option parser assumes the option takes no argument. The only real ambiguity left, string list options that end on "-clr", are special cased in the parser. Remove some duplication of the logic in the command line parser by moving all argument splitting logic into split_opt(). (It's arguable whether that can be considered code duplication, but now the code is a bit simpler anyway. This might be subjective.) Remove the "ambiguous" parameter from all option parsing related code. Make m_config unaware of the pre-parsing concept. Make most CONF_NOCFG options also CONF_GLOBAL (except those explicitly usable as per-file options.)
2012-08-05 21:34:28 +00:00
// Behaves fundamentally like a choice or a superset of it (all allowed string
// values are from a fixed set, although other types of values like numbers
// might be allowed too). E.g. m_option_type_choice and m_option_type_flag.
#define M_OPT_TYPE_CHOICE (1 << 1)
// When m_option.min/max are set, they denote a value range.
#define M_OPT_TYPE_USES_RANGE (1 << 2)
///////////////////////////// Parser flags /////////////////////////////////
// OptionParserReturn
//
// On success parsers return a number >= 0.
//
// To indicate that MPlayer should exit without playing anything,
// parsers return M_OPT_EXIT.
//
// On error one of the following (negative) error codes is returned:
// For use by higher level APIs when the option name is invalid.
#define M_OPT_UNKNOWN -1
// Returned when a parameter is needed but wasn't provided.
#define M_OPT_MISSING_PARAM -2
// Returned when the given parameter couldn't be parsed.
#define M_OPT_INVALID -3
// Returned if the value is "out of range". The exact meaning may
// vary from type to type.
#define M_OPT_OUT_OF_RANGE -4
// The option doesn't take a parameter.
#define M_OPT_DISALLOW_PARAM -5
// Returned when MPlayer should exit. Used by various help stuff.
#define M_OPT_EXIT -6
char *m_option_strerror(int code);
options: Make validation and help possible for all option types Today, validation is only possible for string type options. But there's no particular reason why it needs to be restricted in this way, and there are potential uses, to allow other options to be validated without forcing the option to have to reimplement parsing from scratch. The first part, simply making the validation function an explicit field instead of overloading priv is simple enough. But if we only do that, then the validation function still needs to deal with the raw pre-parsed string. Instead, we want to allow the value to be parsed before it is validated. That in turn leads to us having validator functions that should be type aware. Unfortunately, that means we need to keep the explicit macro like OPT_STRING_VALIDATE() as a way to enforce the correct typing of the function. Otherwise, we'd have to have the validator take a void * and hope the implementation can cast it correctly. For help, we don't have this problem, as help doesn't look at the value. Then, we turn validators that are really help generators into explicit help functions and where a validator is help + validation, we split them into two parts. I have, however, left functions that need to query information for both help and validation as single functions to avoid code duplication. In this change, I have not added an other OPT_FOO_VALIDATE() macros as they are not needed, but I will add some in a separate change to illustrate the pattern.
2021-02-21 00:41:44 +00:00
// Base function to parse options. Includes calling help and validation
// callbacks. Only when this functionality is for some reason required to not
// happen should the parse function pointer be utilized by itself.
//
// See \ref m_option_type::parse.
int m_option_parse(struct mp_log *log, const m_option_t *opt,
struct bstr name, struct bstr param, void *dst);
// Helper to print options, see \ref m_option_type::print.
static inline char *m_option_print(const m_option_t *opt, const void *val_ptr)
{
if (opt->type->print)
return opt->type->print(opt, val_ptr);
else
return NULL;
}
static inline char *m_option_pretty_print(const m_option_t *opt,
const void *val_ptr)
{
if (opt->type->pretty_print)
return opt->type->pretty_print(opt, val_ptr);
else
return m_option_print(opt, val_ptr);
}
// Helper around \ref m_option_type::copy.
static inline void m_option_copy(const m_option_t *opt, void *dst,
const void *src)
{
if (opt->type->copy)
opt->type->copy(opt, dst, src);
}
// Helper around \ref m_option_type::free.
static inline void m_option_free(const m_option_t *opt, void *dst)
{
if (opt->type->free)
opt->type->free(dst);
}
// see m_option_type.set
static inline int m_option_set_node(const m_option_t *opt, void *dst,
struct mpv_node *src)
{
if (opt->type->set)
return opt->type->set(opt, dst, src);
return M_OPT_UNKNOWN;
}
// Call m_option_parse for strings, m_option_set_node otherwise.
int m_option_set_node_or_string(struct mp_log *log, const m_option_t *opt,
const char *name, void *dst, struct mpv_node *src);
// see m_option_type.get
static inline int m_option_get_node(const m_option_t *opt, void *ta_parent,
struct mpv_node *dst, void *src)
{
if (opt->type->get)
return opt->type->get(opt, ta_parent, dst, src);
return M_OPT_UNKNOWN;
}
static inline bool m_option_equal(const m_option_t *opt, void *a, void *b)
{
// Handle trivial equivalence.
// If not implemented, assume this type has no actual values => always equal.
if (a == b || !opt->type->equal)
return true;
return opt->type->equal(opt, a, b);
}
int m_option_required_params(const m_option_t *opt);
extern const char m_option_path_separator;
// Cause a compilation warning if typeof(expr) != type.
// Should be used with pointer types only.
#define MP_EXPECT_TYPE(type, expr) (0 ? (type)0 : (expr))
// This behaves like offsetof(type, member), but will cause a compilation
// warning if typeof(member) != expected_member_type.
// It uses some trickery to make it compile as expression.
#define MP_CHECKED_OFFSETOF(type, member, expected_member_type) \
(offsetof(type, member) + (0 && MP_EXPECT_TYPE(expected_member_type*, \
&((type*)0)->member)))
#define OPT_TYPED_FIELD(type_, c_type, field) \
.type = &type_, \
.offset = MP_CHECKED_OFFSETOF(OPT_BASE_STRUCT, field, c_type)
#define OPTION_LIST_SEPARATOR ','
#define OPTDEF_STR(s) .defval = (void *)&(char * const){s}
#define OPTDEF_INT(i) .defval = (void *)&(const int){i}
#define OPTDEF_INT64(i) .defval = (void *)&(const int64_t){i}
#define OPTDEF_FLOAT(f) .defval = (void *)&(const float){f}
#define OPTDEF_DOUBLE(d) .defval = (void *)&(const double){d}
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
#define M_RANGE(a, b) .min = (double) (a), .max = (double) (b)
#define OPT_BOOL(field) \
OPT_TYPED_FIELD(m_option_type_bool, bool, field)
#define OPT_INT(field) \
OPT_TYPED_FIELD(m_option_type_int, int, field)
options: introduce bool option type, use it for --fullscreen The option code is very old and was added to MPlayer in the early 2000s, when C99 was still new. MPlayer did not use the "bool" type anywhere,l and the logical option equivalent to bool, the "flag" option type, used int, with the convention that only the values 0 and 1 are allowed. mpv may have hammered many, many additional tentacles to the option code, but some of the basics never changed, and m_option_type_flag still uses int. This seems a bit weird, since mpv uses bool for booleans. So finally introduce an m_option_type_bool. To avoid duplicating too much code, change the flag code to bool, and "reimplement" m_option_type_flag on top of m_option_type_bool. As a "demonstration", change the --fullscreen option to this new type. Ideally, all options would be changed too bool, and m_option_type_flag would be removed. But that is a lot of monotonous thankless work, so I'm not doing it, and making it a painful years long transition. At the same time, I'm introducing a new concept for option declarations. Instead of OPT_BOOL(), which define the full m_option struct contents, there's OPTF_BOOL(), which only takes the option field name itself. The name is provided via a normal struct field initializer. Other fields (such as flags) can be provided via designated initializers. The advantage of this is that we don't need tons of nested vararg macros. We also don't need to deal with 0-sized varargs being a pain (and in fact they are not a thing in standard C99 and probably C11). There is no need to provide a mandatory flags argument either, which is the reason why so many OPT_ macros are used with a "0" argument. (The flag argument seems to confuse other developers; they either don't immediately recognize what it is, and sometimes it's supposed to be the option's default value.) Not having to mess with the flag argument in such option macros is also a reason for the removal of M_OPT_RANGE etc., for the better or worse. The only place that special-cased the _flag option type was in command.c; change it to use something effectively very similar that automatically includes the new _bool option type. Everything else should be transparent to the change. The fullscreen option change should be transparent too, as C99 bool is basically an integer type that is clamped to 0/1 (except in Swift, Swift sucks).
2020-03-14 01:07:35 +00:00
#define OPT_INT64(field) \
OPT_TYPED_FIELD(m_option_type_int64, int64_t, field)
#define OPT_FLOAT(field) \
OPT_TYPED_FIELD(m_option_type_float, float, field)
#define OPT_DOUBLE(field) \
OPT_TYPED_FIELD(m_option_type_double, double, field)
#define OPT_STRING(field) \
OPT_TYPED_FIELD(m_option_type_string, char*, field)
#define OPT_STRINGLIST(field) \
OPT_TYPED_FIELD(m_option_type_string_list, char**, field)
#define OPT_KEYVALUELIST(field) \
OPT_TYPED_FIELD(m_option_type_keyvalue_list, char**, field)
#define OPT_PATHLIST(field) \
OPT_TYPED_FIELD(m_option_type_string_list, char**, field), \
.priv = (void *)&m_option_path_separator
#define OPT_TIME(field) \
OPT_TYPED_FIELD(m_option_type_time, double, field)
#define OPT_REL_TIME(field) \
OPT_TYPED_FIELD(m_option_type_rel_time, struct m_rel_time, field)
#define OPT_COLOR(field) \
OPT_TYPED_FIELD(m_option_type_color, struct m_color, field)
#define OPT_BYTE_SIZE(field) \
OPT_TYPED_FIELD(m_option_type_byte_size, int64_t, field)
// (Approximation of x<=SIZE_MAX/2 for m_option.max, which is double.)
#define M_MAX_MEM_BYTES MPMIN((1ULL << 62), (size_t)-1 / 2)
#define OPT_GEOMETRY(field) \
OPT_TYPED_FIELD(m_option_type_geometry, struct m_geometry, field)
#define OPT_SIZE_BOX(field) \
OPT_TYPED_FIELD(m_option_type_size_box, struct m_geometry, field)
#define OPT_RECT(field) \
OPT_TYPED_FIELD(m_option_type_rect, struct m_geometry, field)
#define OPT_TRACKCHOICE(field) \
OPT_CHOICE(field, {"no", -2}, {"auto", -1}), \
M_RANGE(0, 8190)
#define OPT_MSGLEVELS(field) \
OPT_TYPED_FIELD(m_option_type_msglevels, char **, field)
#define OPT_ASPECT(field) \
OPT_TYPED_FIELD(m_option_type_aspect, double, field)
#define OPT_IMAGEFORMAT(field) \
OPT_TYPED_FIELD(m_option_type_imgfmt, int, field)
#define OPT_AUDIOFORMAT(field) \
OPT_TYPED_FIELD(m_option_type_afmt, int, field)
#define OPT_CHANNELS(field) \
OPT_TYPED_FIELD(m_option_type_channels, struct m_channels, field)
2015-03-04 16:20:47 +00:00
#define OPT_INT_VALIDATE(field, validate_fn) \
OPT_TYPED_FIELD(m_option_type_int, int, field), \
.validate = (m_opt_generic_validate_fn) \
MP_EXPECT_TYPE(m_opt_int_validate_fn, validate_fn)
#define OPT_STRING_VALIDATE(field, validate_fn) \
OPT_TYPED_FIELD(m_option_type_string, char*, field), \
options: Make validation and help possible for all option types Today, validation is only possible for string type options. But there's no particular reason why it needs to be restricted in this way, and there are potential uses, to allow other options to be validated without forcing the option to have to reimplement parsing from scratch. The first part, simply making the validation function an explicit field instead of overloading priv is simple enough. But if we only do that, then the validation function still needs to deal with the raw pre-parsed string. Instead, we want to allow the value to be parsed before it is validated. That in turn leads to us having validator functions that should be type aware. Unfortunately, that means we need to keep the explicit macro like OPT_STRING_VALIDATE() as a way to enforce the correct typing of the function. Otherwise, we'd have to have the validator take a void * and hope the implementation can cast it correctly. For help, we don't have this problem, as help doesn't look at the value. Then, we turn validators that are really help generators into explicit help functions and where a validator is help + validation, we split them into two parts. I have, however, left functions that need to query information for both help and validation as single functions to avoid code duplication. In this change, I have not added an other OPT_FOO_VALIDATE() macros as they are not needed, but I will add some in a separate change to illustrate the pattern.
2021-02-21 00:41:44 +00:00
.validate = (m_opt_generic_validate_fn) \
MP_EXPECT_TYPE(m_opt_string_validate_fn, validate_fn)
#define M_CHOICES(...) \
.priv = (void *)&(const struct m_opt_choice_alternatives[]){ __VA_ARGS__, {0}}
// Variant which takes a pointer to struct m_opt_choice_alternatives directly
#define OPT_CHOICE_C(field, choices) \
OPT_TYPED_FIELD(m_option_type_choice, int, field), \
.priv = (void *)MP_EXPECT_TYPE(const struct m_opt_choice_alternatives*, choices)
// Variant where you pass a struct m_opt_choice_alternatives initializer
#define OPT_CHOICE(field, ...) \
OPT_TYPED_FIELD(m_option_type_choice, int, field), \
M_CHOICES(__VA_ARGS__)
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
#define OPT_FLAGS(field, ...) \
OPT_TYPED_FIELD(m_option_type_flags, int, field), \
M_CHOICES(__VA_ARGS__)
#define OPT_SETTINGSLIST(field, objlist) \
OPT_TYPED_FIELD(m_option_type_obj_settings_list, m_obj_settings_t*, field), \
.priv = (void*)MP_EXPECT_TYPE(const struct m_obj_list*, objlist)
#define OPT_FOURCC(field) \
OPT_TYPED_FIELD(m_option_type_fourcc, int, field)
#define OPT_CYCLEDIR(field) \
OPT_TYPED_FIELD(m_option_type_cycle_dir, double, field)
// subconf must have the type struct m_sub_options.
// All sub-options are prefixed with "name-" and are added to the current
// (containing) option list.
// If name is "", add the sub-options directly instead.
// "field" refers to the field, that must be a pointer to a field described by
// the subconf struct.
#define OPT_SUBSTRUCT(field, subconf) \
.offset = offsetof(OPT_BASE_STRUCT, field), \
.type = &m_option_type_subconfig, .priv = (void*)&subconf
// Non-fields
#define OPT_ALIAS(newname) \
.type = &m_option_type_alias, .priv = newname, .offset = -1
// If "--optname" was removed, but "--newname" has the same semantics.
// It will be redirected, and a warning will be printed on first use.
#define OPT_REPLACED_MSG(newname, msg) \
.type = &m_option_type_alias, .priv = newname, \
.deprecation_message = (msg), .offset = -1
// Same, with a generic deprecation message.
#define OPT_REPLACED(newname) OPT_REPLACED_MSG(newname, "")
// Alias, resolved on the CLI/config file/profile parser level only.
#define OPT_CLI_ALIAS(newname) \
.type = &m_option_type_cli_alias, .priv = newname, \
.flags = M_OPT_NOPROP, .offset = -1
// "--optname" doesn't exist, but inform the user about a replacement with msg.
#define OPT_REMOVED(msg) \
.type = &m_option_type_removed, .priv = msg, \
.deprecation_message = "", .flags = M_OPT_NOPROP, .offset = -1
#define OPT_PRINT(fn) \
.flags = M_OPT_NOCFG | M_OPT_PRE_PARSE | M_OPT_NOPROP, \
.type = &m_option_type_print_fn, \
.priv = MP_EXPECT_TYPE(m_opt_print_fn, fn), \
.offset = -1
#endif /* MPLAYER_M_OPTION_H */