mpv/stream/stream.h

281 lines
8.7 KiB
C
Raw Normal View History

/*
* This file is part of mpv.
*
stream: change license to LGPL All relevant authors have agreed. There are two exceptions, patches by authors who could not be reached. This commit tries to remove their copyright. a0f08fbe: messes with the seeking code corner cases. The EOF flag logic was changed at some point, and always had a flaky history (see e.g. 347cf972 50274ca3 70411f27 5999efb9 0d5e6084 ff08d0c3 2e2f77e3 de5566f0 9554a844, all which happened after that patch, MPlayer ones without that patch). I claim that all of the copyright the patch might have added is gone. Except the message in stream_seek(), which this commit removes. The other code removed/changed in stream_seek() is probably not from that patch, but it doesn't hurt to be sure, and also makes it more readable. (It might change the behavior so that sometimes the eof flag is set after a seek, but it doesn't matter here.) 2aa6acd9: it looks like the seek_forward() modified by this patch was later moved to stream.c and renamed to stream_skip_read() in a790f2133. (Looking closer at it, it was actually modified again a bunch of times, fixing the logic.) I rewrote it in this commit. The code ended up rather similar, which probably could lead to doubts whether this was done properly, but I guess the reader of this will just have to believe me. I knew what stream_skip_read() was supposed to do (which was reinforced when I tried to replace it on the caller side), without reading the pre-existing code in detail. I had to "relearn" the logic how buf_pos and bug_len work - it was actually easy to see from stream_read_char() how to skip the data, essentially by generalizing its logic from 1 byte to N bytes. From the old code I only "used" the fact that it's obviously a while(len>0) look, that has to call stream_fill_buffer repeatedly to make progress. At first I actually didn't use stream_fill_buffer_by(), but the variant without _by, but readded it when I checked why the old code used it (see cd7ec016e7). This has to be good enough. In the end, it's hard to argue that this could be implemented in a way not using such a loop. Other than this, I could add the usual remarks about how this code was not modularized in the past, and how stream.c contained DVD code, and how this was later modularized, moving the copyright to other files, and so on. Also, if someone wrote a stream module, and was not asked about LGPL relicensing, we don't consider the entry in stream_list[] copyrightable.
2017-06-19 14:07:42 +00:00
* 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
stream: change license to LGPL All relevant authors have agreed. There are two exceptions, patches by authors who could not be reached. This commit tries to remove their copyright. a0f08fbe: messes with the seeking code corner cases. The EOF flag logic was changed at some point, and always had a flaky history (see e.g. 347cf972 50274ca3 70411f27 5999efb9 0d5e6084 ff08d0c3 2e2f77e3 de5566f0 9554a844, all which happened after that patch, MPlayer ones without that patch). I claim that all of the copyright the patch might have added is gone. Except the message in stream_seek(), which this commit removes. The other code removed/changed in stream_seek() is probably not from that patch, but it doesn't hurt to be sure, and also makes it more readable. (It might change the behavior so that sometimes the eof flag is set after a seek, but it doesn't matter here.) 2aa6acd9: it looks like the seek_forward() modified by this patch was later moved to stream.c and renamed to stream_skip_read() in a790f2133. (Looking closer at it, it was actually modified again a bunch of times, fixing the logic.) I rewrote it in this commit. The code ended up rather similar, which probably could lead to doubts whether this was done properly, but I guess the reader of this will just have to believe me. I knew what stream_skip_read() was supposed to do (which was reinforced when I tried to replace it on the caller side), without reading the pre-existing code in detail. I had to "relearn" the logic how buf_pos and bug_len work - it was actually easy to see from stream_read_char() how to skip the data, essentially by generalizing its logic from 1 byte to N bytes. From the old code I only "used" the fact that it's obviously a while(len>0) look, that has to call stream_fill_buffer repeatedly to make progress. At first I actually didn't use stream_fill_buffer_by(), but the variant without _by, but readded it when I checked why the old code used it (see cd7ec016e7). This has to be good enough. In the end, it's hard to argue that this could be implemented in a way not using such a loop. Other than this, I could add the usual remarks about how this code was not modularized in the past, and how stream.c contained DVD code, and how this was later modularized, moving the copyright to other files, and so on. Also, if someone wrote a stream module, and was not asked about LGPL relicensing, we don't consider the entry in stream_list[] copyrightable.
2017-06-19 14:07:42 +00:00
* GNU Lesser General Public License for more details.
*
stream: change license to LGPL All relevant authors have agreed. There are two exceptions, patches by authors who could not be reached. This commit tries to remove their copyright. a0f08fbe: messes with the seeking code corner cases. The EOF flag logic was changed at some point, and always had a flaky history (see e.g. 347cf972 50274ca3 70411f27 5999efb9 0d5e6084 ff08d0c3 2e2f77e3 de5566f0 9554a844, all which happened after that patch, MPlayer ones without that patch). I claim that all of the copyright the patch might have added is gone. Except the message in stream_seek(), which this commit removes. The other code removed/changed in stream_seek() is probably not from that patch, but it doesn't hurt to be sure, and also makes it more readable. (It might change the behavior so that sometimes the eof flag is set after a seek, but it doesn't matter here.) 2aa6acd9: it looks like the seek_forward() modified by this patch was later moved to stream.c and renamed to stream_skip_read() in a790f2133. (Looking closer at it, it was actually modified again a bunch of times, fixing the logic.) I rewrote it in this commit. The code ended up rather similar, which probably could lead to doubts whether this was done properly, but I guess the reader of this will just have to believe me. I knew what stream_skip_read() was supposed to do (which was reinforced when I tried to replace it on the caller side), without reading the pre-existing code in detail. I had to "relearn" the logic how buf_pos and bug_len work - it was actually easy to see from stream_read_char() how to skip the data, essentially by generalizing its logic from 1 byte to N bytes. From the old code I only "used" the fact that it's obviously a while(len>0) look, that has to call stream_fill_buffer repeatedly to make progress. At first I actually didn't use stream_fill_buffer_by(), but the variant without _by, but readded it when I checked why the old code used it (see cd7ec016e7). This has to be good enough. In the end, it's hard to argue that this could be implemented in a way not using such a loop. Other than this, I could add the usual remarks about how this code was not modularized in the past, and how stream.c contained DVD code, and how this was later modularized, moving the copyright to other files, and so on. Also, if someone wrote a stream module, and was not asked about LGPL relicensing, we don't consider the entry in stream_list[] copyrightable.
2017-06-19 14:07:42 +00:00
* 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_STREAM_H
#define MPLAYER_STREAM_H
#include "common/msg.h"
#include <stdbool.h>
#include <stdio.h>
#include <string.h>
#include <inttypes.h>
#include <sys/types.h>
#include <fcntl.h>
#include "misc/bstr.h"
#define STREAM_BUFFER_SIZE 2048
#define STREAM_MAX_SECTOR_SIZE (8 * 1024)
// Max buffer for initial probe.
#define STREAM_MAX_BUFFER_SIZE (2 * 1024 * 1024)
2013-06-06 18:39:53 +00:00
// stream->mode
#define STREAM_READ 0
#define STREAM_WRITE 1
2013-06-06 18:39:53 +00:00
// flags for stream_open_ext (this includes STREAM_READ and STREAM_WRITE)
#define STREAM_SAFE_ONLY 4
#define STREAM_NETWORK_ONLY 8
#define STREAM_UNSAFE -3
#define STREAM_NO_MATCH -2
#define STREAM_UNSUPPORTED -1
#define STREAM_ERROR 0
#define STREAM_OK 1
enum stream_ctrl {
STREAM_CTRL_GET_SIZE = 1,
// Cache
STREAM_CTRL_GET_CACHE_INFO,
STREAM_CTRL_SET_CACHE_SIZE,
STREAM_CTRL_SET_READAHEAD,
// stream_memory.c
STREAM_CTRL_SET_CONTENTS,
// stream_rar.c
STREAM_CTRL_GET_BASE_FILENAME,
// Certain network protocols
STREAM_CTRL_RECONNECT,
STREAM_CTRL_AVSEEK,
STREAM_CTRL_HAS_AVSEEK,
STREAM_CTRL_GET_METADATA,
// TV
STREAM_CTRL_TV_SET_SCAN,
STREAM_CTRL_SET_TV_FREQ,
STREAM_CTRL_GET_TV_FREQ,
STREAM_CTRL_SET_TV_COLORS,
STREAM_CTRL_GET_TV_COLORS,
STREAM_CTRL_TV_SET_NORM,
STREAM_CTRL_TV_STEP_NORM,
STREAM_CTRL_TV_SET_CHAN,
STREAM_CTRL_TV_GET_CHAN,
STREAM_CTRL_TV_STEP_CHAN,
STREAM_CTRL_TV_LAST_CHAN,
STREAM_CTRL_DVB_SET_CHANNEL,
STREAM_CTRL_DVB_SET_CHANNEL_NAME,
STREAM_CTRL_DVB_GET_CHANNEL_NAME,
STREAM_CTRL_DVB_STEP_CHANNEL,
// Optical discs
STREAM_CTRL_GET_TIME_LENGTH,
STREAM_CTRL_GET_DVD_INFO,
STREAM_CTRL_GET_DISC_NAME,
STREAM_CTRL_GET_NUM_CHAPTERS,
STREAM_CTRL_GET_CURRENT_TIME,
STREAM_CTRL_GET_CHAPTER_TIME,
STREAM_CTRL_SEEK_TO_TIME,
STREAM_CTRL_GET_ASPECT_RATIO,
STREAM_CTRL_GET_NUM_ANGLES,
STREAM_CTRL_GET_ANGLE,
STREAM_CTRL_SET_ANGLE,
STREAM_CTRL_GET_NUM_TITLES,
STREAM_CTRL_GET_TITLE_LENGTH, // double* (in: title number, out: len)
STREAM_CTRL_GET_LANG,
STREAM_CTRL_GET_CURRENT_TITLE,
STREAM_CTRL_SET_CURRENT_TITLE,
};
// for STREAM_CTRL_GET_CACHE_INFO
struct stream_cache_info {
int64_t size;
int64_t fill;
bool idle;
int64_t speed;
};
struct stream_lang_req {
int type; // STREAM_AUDIO, STREAM_SUB
int id;
char name[50];
};
struct stream_dvd_info_req {
unsigned int palette[16];
int num_subs;
};
// for STREAM_CTRL_SET_TV_COLORS
#define TV_COLOR_BRIGHTNESS 1
#define TV_COLOR_HUE 2
#define TV_COLOR_SATURATION 3
#define TV_COLOR_CONTRAST 4
// for STREAM_CTRL_AVSEEK
struct stream_avseek {
int stream_index;
int64_t timestamp;
int flags;
};
struct stream;
typedef struct stream_info_st {
const char *name;
2013-06-06 18:39:53 +00:00
// opts is set from ->opts
int (*open)(struct stream *st);
const char *const *protocols;
bool can_write; // correctly checks for READ/WRITE modes
bool is_safe; // opening is no security issue, even with remote provided URLs
bool is_network; // used to restrict remote playlist entries to remote URLs
} stream_info_t;
typedef struct stream {
const struct stream_info_st *info;
// Read
int (*fill_buffer)(struct stream *s, char *buffer, int max_len);
// Write
int (*write_buffer)(struct stream *s, char *buffer, int len);
// Seek
int (*seek)(struct stream *s, int64_t pos);
// Control
// Will be later used to let streams like dvd and cdda report
// their structure (ie tracks, chapters, etc)
int (*control)(struct stream *s, int cmd, void *arg);
// Close
void (*close)(struct stream *s);
int sector_size; // sector size (seek will be aligned on this size if non 0)
int read_chunk; // maximum amount of data to read at once to limit latency
unsigned int buf_pos, buf_len;
int64_t pos;
int eof;
int mode; //STREAM_READ or STREAM_WRITE
void *priv; // used for DVD, TV, RTSP etc
char *url; // filename/url (possibly including protocol prefix)
char *path; // filename (url without protocol prefix)
char *mime_type; // when HTTP streaming is used
char *demuxer; // request demuxer to be used
char *lavf_type; // name of expected demuxer type for lavf
bool streaming : 1; // known to be a network stream if true
bool seekable : 1; // presence of general byte seeking support
bool fast_skip : 1; // consider stream fast enough to fw-seek by skipping
bool is_network : 1; // original stream_info_t.is_network flag
bool allow_caching : 1; // stream cache makes sense
bool caching : 1; // is a cache, or accesses a cache
bool is_local_file : 1; // from the filesystem
bool is_directory : 1; // directory on the filesystem
bool access_references : 1; // open other streams
struct mp_log *log;
struct mpv_global *global;
stream: redo playback abort handling This mechanism originates from MPlayer's way of dealing with blocking network, but it's still useful. On opening and closing, mpv waits for network synchronously, and also some obscure commands and use-cases can lead to such blocking. In these situations, the stream is asynchronously forced to stop by "interrupting" it. The old design interrupting I/O was a bit broken: polling with a callback, instead of actively interrupting it. Change the direction of this. There is no callback anymore, and the player calls mp_cancel_trigger() to force the stream to return. libavformat (via stream_lavf.c) has the old broken design, and fixing it would require fixing libavformat, which won't happen so quickly. So we have to keep that part. But everything above the stream layer is prepared for a better design, and more sophisticated methods than mp_cancel_test() could be easily introduced. There's still one problem: commands are still run in the central playback loop, which we assume can block on I/O in the worst case. That's not a problem yet, because we simply mark some commands as being able to stop playback of the current file ("quit" etc.), so input.c could abort playback as soon as such a command is queued. But there are also commands abort playback only conditionally, and the logic for that is in the playback core and thus "unreachable". For example, "playlist_next" aborts playback only if there's a next file. We don't want it to always abort playback. As a quite ugly hack, abort playback only if at least 2 abort commands are queued - this pretty much happens only if the core is frozen and doesn't react to input.
2014-09-13 12:23:08 +00:00
struct mp_cancel *cancel; // cancellation notification
struct stream *underlying; // e.g. cache wrapper
// Includes additional padding in case sizes get rounded up by sector size.
unsigned char buffer[];
} stream_t;
int stream_fill_buffer(stream_t *s);
struct mp_cache_opts;
bool stream_wants_cache(stream_t *stream, struct mp_cache_opts *opts);
int stream_enable_cache_defaults(stream_t **stream);
cache: make the stream cache a proper stream that wraps other streams Before this commit, the cache was franken-hacked on top of the stream API. You had to use special functions (like cache_stream_fill_buffer() instead of stream_fill_buffer()), which would access the stream in a cached manner. The whole idea about the previous design was that the cache runs in a thread or in a forked process, while the cache awa functions made sure the stream instance looked consistent to the user. If you used the normal functions instead of the special ones while the cache was running, you were out of luck. Make it a bit more reasonable by turning the cache into a stream on its own. This makes it behave exactly like a normal stream. The stream callbacks call into the original (uncached) stream to do work. No special cache functions or redirections are needed. The only different thing about cache streams is that they are created by special functions, instead of being part of the auto_open_streams[] array. To make things simpler, remove the threading implementation, which was messed into the code. The threading code could perhaps be kept, but I don't really want to have to worry about this special case. A proper threaded implementation will be added later. Remove the cache enabling code from stream_radio.c. Since enabling the cache involves replacing the old stream with a new one, the code as-is can't be kept. It would be easily possible to enable the cache by requesting a cache size (which is also much simpler). But nobody uses stream_radio.c and I can't even test this thing, and the cache is probably not really important for it either.
2013-05-24 16:49:09 +00:00
// Internal
int stream_cache_init(stream_t *cache, stream_t *stream,
struct mp_cache_opts *opts);
int stream_file_cache_init(stream_t *cache, stream_t *stream,
struct mp_cache_opts *opts);
cache: make the stream cache a proper stream that wraps other streams Before this commit, the cache was franken-hacked on top of the stream API. You had to use special functions (like cache_stream_fill_buffer() instead of stream_fill_buffer()), which would access the stream in a cached manner. The whole idea about the previous design was that the cache runs in a thread or in a forked process, while the cache awa functions made sure the stream instance looked consistent to the user. If you used the normal functions instead of the special ones while the cache was running, you were out of luck. Make it a bit more reasonable by turning the cache into a stream on its own. This makes it behave exactly like a normal stream. The stream callbacks call into the original (uncached) stream to do work. No special cache functions or redirections are needed. The only different thing about cache streams is that they are created by special functions, instead of being part of the auto_open_streams[] array. To make things simpler, remove the threading implementation, which was messed into the code. The threading code could perhaps be kept, but I don't really want to have to worry about this special case. A proper threaded implementation will be added later. Remove the cache enabling code from stream_radio.c. Since enabling the cache involves replacing the old stream with a new one, the code as-is can't be kept. It would be easily possible to enable the cache by requesting a cache size (which is also much simpler). But nobody uses stream_radio.c and I can't even test this thing, and the cache is probably not really important for it either.
2013-05-24 16:49:09 +00:00
int stream_write_buffer(stream_t *s, unsigned char *buf, int len);
inline static int stream_read_char(stream_t *s)
{
return (s->buf_pos < s->buf_len) ? s->buffer[s->buf_pos++] :
cache: make the stream cache a proper stream that wraps other streams Before this commit, the cache was franken-hacked on top of the stream API. You had to use special functions (like cache_stream_fill_buffer() instead of stream_fill_buffer()), which would access the stream in a cached manner. The whole idea about the previous design was that the cache runs in a thread or in a forked process, while the cache awa functions made sure the stream instance looked consistent to the user. If you used the normal functions instead of the special ones while the cache was running, you were out of luck. Make it a bit more reasonable by turning the cache into a stream on its own. This makes it behave exactly like a normal stream. The stream callbacks call into the original (uncached) stream to do work. No special cache functions or redirections are needed. The only different thing about cache streams is that they are created by special functions, instead of being part of the auto_open_streams[] array. To make things simpler, remove the threading implementation, which was messed into the code. The threading code could perhaps be kept, but I don't really want to have to worry about this special case. A proper threaded implementation will be added later. Remove the cache enabling code from stream_radio.c. Since enabling the cache involves replacing the old stream with a new one, the code as-is can't be kept. It would be easily possible to enable the cache by requesting a cache size (which is also much simpler). But nobody uses stream_radio.c and I can't even test this thing, and the cache is probably not really important for it either.
2013-05-24 16:49:09 +00:00
(stream_fill_buffer(s) ? s->buffer[s->buf_pos++] : -256);
}
unsigned char *stream_read_line(stream_t *s, unsigned char *mem, int max,
int utf16);
int stream_skip_bom(struct stream *s);
inline static int stream_eof(stream_t *s)
{
return s->eof;
}
inline static int64_t stream_tell(stream_t *s)
{
return s->pos + s->buf_pos - s->buf_len;
}
bool stream_skip(stream_t *s, int64_t len);
bool stream_seek(stream_t *s, int64_t pos);
int stream_read(stream_t *s, char *mem, int total);
int stream_read_partial(stream_t *s, char *buf, int buf_size);
struct bstr stream_peek(stream_t *s, int len);
void stream_drop_buffers(stream_t *s);
int64_t stream_get_size(stream_t *s);
struct mpv_global;
struct bstr stream_read_complete(struct stream *s, void *talloc_ctx,
int max_size);
struct bstr stream_read_file(const char *filename, void *talloc_ctx,
struct mpv_global *global, int max_size);
int stream_control(stream_t *s, int cmd, void *arg);
void free_stream(stream_t *s);
stream: redo playback abort handling This mechanism originates from MPlayer's way of dealing with blocking network, but it's still useful. On opening and closing, mpv waits for network synchronously, and also some obscure commands and use-cases can lead to such blocking. In these situations, the stream is asynchronously forced to stop by "interrupting" it. The old design interrupting I/O was a bit broken: polling with a callback, instead of actively interrupting it. Change the direction of this. There is no callback anymore, and the player calls mp_cancel_trigger() to force the stream to return. libavformat (via stream_lavf.c) has the old broken design, and fixing it would require fixing libavformat, which won't happen so quickly. So we have to keep that part. But everything above the stream layer is prepared for a better design, and more sophisticated methods than mp_cancel_test() could be easily introduced. There's still one problem: commands are still run in the central playback loop, which we assume can block on I/O in the worst case. That's not a problem yet, because we simply mark some commands as being able to stop playback of the current file ("quit" etc.), so input.c could abort playback as soon as such a command is queued. But there are also commands abort playback only conditionally, and the logic for that is in the playback core and thus "unreachable". For example, "playlist_next" aborts playback only if there's a next file. We don't want it to always abort playback. As a quite ugly hack, abort playback only if at least 2 abort commands are queued - this pretty much happens only if the core is frozen and doesn't react to input.
2014-09-13 12:23:08 +00:00
struct stream *stream_create(const char *url, int flags,
struct mp_cancel *c, struct mpv_global *global);
struct stream *stream_open(const char *filename, struct mpv_global *global);
stream_t *open_output_stream(const char *filename, struct mpv_global *global);
2013-06-20 22:47:58 +00:00
stream_t *open_memory_stream(void *data, int len);
void mp_url_unescape_inplace(char *buf);
char *mp_url_escape(void *talloc_ctx, const char *s, const char *ok);
stream: redo playback abort handling This mechanism originates from MPlayer's way of dealing with blocking network, but it's still useful. On opening and closing, mpv waits for network synchronously, and also some obscure commands and use-cases can lead to such blocking. In these situations, the stream is asynchronously forced to stop by "interrupting" it. The old design interrupting I/O was a bit broken: polling with a callback, instead of actively interrupting it. Change the direction of this. There is no callback anymore, and the player calls mp_cancel_trigger() to force the stream to return. libavformat (via stream_lavf.c) has the old broken design, and fixing it would require fixing libavformat, which won't happen so quickly. So we have to keep that part. But everything above the stream layer is prepared for a better design, and more sophisticated methods than mp_cancel_test() could be easily introduced. There's still one problem: commands are still run in the central playback loop, which we assume can block on I/O in the worst case. That's not a problem yet, because we simply mark some commands as being able to stop playback of the current file ("quit" etc.), so input.c could abort playback as soon as such a command is queued. But there are also commands abort playback only conditionally, and the logic for that is in the playback core and thus "unreachable". For example, "playlist_next" aborts playback only if there's a next file. We don't want it to always abort playback. As a quite ugly hack, abort playback only if at least 2 abort commands are queued - this pretty much happens only if the core is frozen and doesn't react to input.
2014-09-13 12:23:08 +00:00
struct mp_cancel *mp_cancel_new(void *talloc_ctx);
void mp_cancel_trigger(struct mp_cancel *c);
bool mp_cancel_test(struct mp_cancel *c);
bool mp_cancel_wait(struct mp_cancel *c, double timeout);
stream: redo playback abort handling This mechanism originates from MPlayer's way of dealing with blocking network, but it's still useful. On opening and closing, mpv waits for network synchronously, and also some obscure commands and use-cases can lead to such blocking. In these situations, the stream is asynchronously forced to stop by "interrupting" it. The old design interrupting I/O was a bit broken: polling with a callback, instead of actively interrupting it. Change the direction of this. There is no callback anymore, and the player calls mp_cancel_trigger() to force the stream to return. libavformat (via stream_lavf.c) has the old broken design, and fixing it would require fixing libavformat, which won't happen so quickly. So we have to keep that part. But everything above the stream layer is prepared for a better design, and more sophisticated methods than mp_cancel_test() could be easily introduced. There's still one problem: commands are still run in the central playback loop, which we assume can block on I/O in the worst case. That's not a problem yet, because we simply mark some commands as being able to stop playback of the current file ("quit" etc.), so input.c could abort playback as soon as such a command is queued. But there are also commands abort playback only conditionally, and the logic for that is in the playback core and thus "unreachable". For example, "playlist_next" aborts playback only if there's a next file. We don't want it to always abort playback. As a quite ugly hack, abort playback only if at least 2 abort commands are queued - this pretty much happens only if the core is frozen and doesn't react to input.
2014-09-13 12:23:08 +00:00
void mp_cancel_reset(struct mp_cancel *c);
void *mp_cancel_get_event(struct mp_cancel *c); // win32 HANDLE
lua: add an utility function for starting processes Because 1) Lua is terrible, and 2) popen() is terrible. Unfortunately, since Unix is also terrible, this turned out more complicated than I hoped. As a consequence and to avoid that this code has to be maintained forever, add a disclaimer that any function in Lua's utils module can disappear any time. The complexity seems a bit ridiculous, especially for a feature so far removed from actual video playback, so if it turns out that we don't really need this function, it will be dropped again. The motivation for this commit is the same as with 8e4fa5fc. Note that there is an "#ifndef __GLIBC__". The GNU people are very special people and thought it'd be convenient to actually declare "environ", even though the POSIX people, which are also very special people, state that no header declares this and that the user has to declare this manually. Since the GNU people overtook the Unix world with their very clever "embrace, extend, extinguish" strategy, but not 100%, and trying to build without _GNU_SOURCE is hopeless; but since there might be Unix environments which support _GNU_SOURCE features partially, this means that in practice "environ" will be randomly declared or not declared by system headers. Also, gcc was written by very clever people too, and prints a warning if an external variable is declared twice (I didn't check, but I suppose redeclaring is legal C, and not even the gcc people are clever enough to only warn against a definitely not legal C construct, although sometimes they do this), ...and since we at mpv hate compiler warnings, we seek to silence them all. Adding a configure test just for a warning seems too radical, so we special-case this against __GLIBC__, which is hopefully not defined on other libcs, especially not libcs which don't implement all aspects of _GNU_SOURCE, and redefine "environ" on systems even if the headers define it already (because they support _GNU_SOURCE - as I mentioned before, the clever GNU people wrote software THAT portable that other libcs just gave up and implemented parts of _GNU_SOURCE, although probably not all), which means that compiling mpv will print a warning about "environ" being redefined, but at least this won't happen on my system, so all is fine. However, should someone complain about this warning, I will force whoever complained about this warning to read this ENTIRE commit message, and if possible, will also force them to eat a printed-out copy of the GNU Manifesto, and if that is not enough, maybe this person could even be forced to convince the very clever POSIX people of not doing crap like this: having the user to manually declare somewhat central symbols - but I doubt it's possible, because the POSIX people are too far gone and only care about maintaining compatibility with old versions of AIX and HP-UX. Oh, also, this code contains some subtle and obvious issues, but writing about this is not fun.
2014-10-18 23:42:28 +00:00
int mp_cancel_get_fd(struct mp_cancel *c);
stream: redo playback abort handling This mechanism originates from MPlayer's way of dealing with blocking network, but it's still useful. On opening and closing, mpv waits for network synchronously, and also some obscure commands and use-cases can lead to such blocking. In these situations, the stream is asynchronously forced to stop by "interrupting" it. The old design interrupting I/O was a bit broken: polling with a callback, instead of actively interrupting it. Change the direction of this. There is no callback anymore, and the player calls mp_cancel_trigger() to force the stream to return. libavformat (via stream_lavf.c) has the old broken design, and fixing it would require fixing libavformat, which won't happen so quickly. So we have to keep that part. But everything above the stream layer is prepared for a better design, and more sophisticated methods than mp_cancel_test() could be easily introduced. There's still one problem: commands are still run in the central playback loop, which we assume can block on I/O in the worst case. That's not a problem yet, because we simply mark some commands as being able to stop playback of the current file ("quit" etc.), so input.c could abort playback as soon as such a command is queued. But there are also commands abort playback only conditionally, and the logic for that is in the playback core and thus "unreachable". For example, "playlist_next" aborts playback only if there's a next file. We don't want it to always abort playback. As a quite ugly hack, abort playback only if at least 2 abort commands are queued - this pretty much happens only if the core is frozen and doesn't react to input.
2014-09-13 12:23:08 +00:00
// stream_file.c
char *mp_file_url_to_filename(void *talloc_ctx, bstr url);
char *mp_file_get_path(void *talloc_ctx, bstr url);
// stream_lavf.c
struct AVDictionary;
void mp_setup_av_network_options(struct AVDictionary **dict,
struct mpv_global *global,
struct mp_log *log);
2014-06-30 10:49:01 +00:00
void stream_print_proto_list(struct mp_log *log);
char **stream_get_proto_list(void);
bool stream_has_proto(const char *proto);
2014-06-30 10:49:01 +00:00
#endif /* MPLAYER_STREAM_H */