summaryrefslogtreecommitdiffstats
path: root/osdep
Commit message (Collapse)AuthorAgeFilesLines
* win32: add more-POSIXy versions of open() and fstat()James Ross-Gowan2017-10-252-39/+296
| | | | | | | | | | | | | | Directory-opening never worked on Windows because MSVCRT's open() doesn't open directories and its fstat() doesn't recognise directory handles. These are just MSVCRT restrictions, and the Windows API itself has no problem with opening directories as file objects, so reimplement mpv's mp_open and mp_stat to use the Windows API directly. This should fix directory playback. This also populates the st_dev and st_ino fields of struct stat, so filesystem loop checking in demux_playlist.c should now work on Windows. Fixes #4711
* osx: add right MacPorts paths to our bundle $PATHAkemi2017-10-071-2/+3
| | | | also add another path for homebrew.
* osx: add some common paths to our $PATH when started from bundleAkemi2017-10-031-0/+8
| | | | | | | | | | | with the previous commit we removed the ability of loading the standard shell environment. this exact behaviour can only be re-added by either invoking the standard shell in a background process and reading it's environment vars or by manually reading all the various shell configs. both ways are kinda dirty and the former was already rejected before. for now we will just add some commonly used paths, when started from the bundle, so it can find the binaries used by mpv again. for example the youtube-dl one for our youtube-dl hook.
* osx: fix bundle on macOS High Sierra (10.13)Akemi2017-10-032-6/+21
| | | | | | | | | | | | | | | | | | | | | | Apple slightly changed the App bundle mechanism which broke wrapper scripts that invoke the actual binary. it caused the bundle to always open a new instance of mpv instead of reusing the currently running one. just removing the wrapper script would lead to several regressions, so it was replaced with a symlink to the bundle binary. detection if mpv was started from the bundle was replaced by comparing the execution name of the binary, eg the name of the symlink "mpv-bundle". additionally, because we load a standard config from the Resources folder of the bundle again, we prevent that config from being loaded if mpv wasn't started via the bundle. the psn argument has to be removed manually again. the ability of loading your standard shell environment has been removed with the wrapper. a substitution will be added with another commit. as a side effect this fixes an issues when zsh was used with common NodeJS configuration scripts. Fixes #4926 #4866
* android: posix_spawn(p) replacementsfan52017-09-224-1/+143
| | | | Signed-off-by: wm4 <wm4@nowhere>
* osdep/io: add android related bullshitJan Ekström2017-09-181-0/+27
| | | | | | | | | | | | | This fixes >2GiB files with at least API level 21. See: https://android.googlesource.com/platform/bionic/+/master/docs/32-bit-abi.md for the gritty details. * Based on libavformat's things, except we do not care about API versions or NDKs without unistd.h, which contains all sorts of things that we utilize. * Redefines lseek to always point to the 64bit version. * Redefines fseeko to always point towards an inlined static implementation that utilizes the 64bit version of lseek underneath.
* osx: fix media keys input when other Apps steal the priorityAkemi2017-09-073-0/+27
| | | | | | | | | | | other Apps do the same as mpv and tap into the global event chain. events that are handled are not being propagated down the event chain. that can lead to mpv not getting any media key events anymore when they are held back by other Apps. we will just move mpv back to the top of the event list when mpv is refocused and is not at the top of the list any more. Fixes #4834
* osx: fix build on macOS 10.11 and earlierAkemi2017-08-262-2/+8
| | | | | | | | 98f5579 added a safeguard for the deactivation of the automatic tab bar but it still needs a compile time guard. add a forward declaration for the setAllowsAutomaticWindowTabbing method of NSWindow. Fixes #4783
* win32: fix massive memory corruption (take 2)wm42017-08-211-3/+4
| | | | | | As pointed out by uau on IRC, the pointer to info is still used outside of the lock. An extremely small race condition window, but still a race condition.
* win32: fix massive memory corruptionwm42017-08-211-2/+6
| | | | | | | | | | The struct m_thread_info pointer is part of an array, that will be reallocated if another thread is created while the run_thread is just being called. In previous versions of this code, the pointer was stable (as long as the thread existed), so this was overlooked. Fixes #4770. I'm not sure why this triggers it so reliably, while it remained undetected otherwise.
* osx: remove the standard "Show Tab Bar" menu itemAkemi2017-08-181-0/+3
| | | | | it's unused and added automatically. we need to guard the deactivation since it's a 10.12+ feature.
* osx: code cleanups and cosmetic fixesAkemi2017-08-182-5/+7
| | | | silence build warnings, clean up code style and remove unused code.
* osx: refactor menu bar creationAkemi2017-08-184-81/+686
| | | | | make the menu bar look more like a standard macOS menu bar, with all the standard menu items like About, Help, Open, etc.
* osx: move menu bar creation into its own fileAkemi2017-08-186-142/+211
|
* {demux_mf,osdep/io}: disable glob usage when it is not availableJan Ekström2017-08-051-1/+1
| | | | | | | | This currently is only limited to Android. Its stdlib contains the things that mpv's POSIX check checks for, but unfortunately not glob(). This fixes Android compilation broken in 70a70b9da .
* w32_common: handle media keysJames Ross-Gowan2017-08-052-0/+23
| | | | | | | | | | | | | | | | | This was attempted before in fc9695e63b5b, but it was reverted in 1b7ce759b1f4 because it caused conflicts with other software watching the same keys (See #2041.) It seems like some PCs ship with OEM software that watches the volume keys without consuming key events and this causes them to be handled twice, once by mpv and once by the other software. In order to prevent conflicts like this, use the WM_APPCOMMAND message to handle media keys. Returning TRUE from the WM_APPCOMMAND handler should indicate to the operating system that we consumed the key event and it should not be propogated to the shell. Also, we now only listen for keys that are directly related to multimedia playback (eg. the APPCOMMAND_MEDIA_* keys.) Keys like APPCOMMAND_VOLUME_* are ignored, so they can be handled by the shell, or by other mixer software.
* subprocess-win: fix swprintf usageJames Ross-Gowan2017-07-291-1/+2
| | | | | | | The count parameter refers to the number of wide-characters to write to the buffer (including the terminating NUL character, which is always added,) not the number of bytes. See C99 7.24.2.3 or https://msdn.microsoft.com/en-us/library/ybk95axf.aspx
* subprocess-win: don't leak pipe handle on errorJames Ross-Gowan2017-07-281-1/+3
|
* subprocess-win: remove VLA usageJames Ross-Gowan2017-07-261-2/+4
| | | | This only ever waits on four handles at most anyway.
* Universal Windows Plaform (UWP) supportPedro Pombeiro2017-06-296-3/+89
| | | | | | | | libmpv only. Some things are still missing. Heavily reworked. Signed-off-by: wm4 <wm4@nowhere>
* Avoid calling close(-1)wm42017-06-292-20/+24
| | | | | | | | | | While this is perfectly OK on Unix, it causes annoying valgrind warnings, and might be otherwise confusing to others. On Windows, the runtime can actually abort the process if this is called. push.c part taken from a patch by Pedro Pombeiro.
* build: change how some OS specific source files are selectedwm42017-06-293-10/+40
| | | | | | | | | | | | | | | | | | In a bunch of cases, we emulate highly platform specific APIs on a higher level across all OSes, such as IPC, terminal, subprocess handling, and more. We have source files for each OS, and they implement all the same mpv internal API. Selecting which source file to use on an OS can be tricky, because there is partially overlapping and emulated APIs (consider Cygwin on Windows). Add a pick_first_matching_dep() function to make this slightly easier and more structured. Also add dummy backends in some cases, to deal with APIs not being available. Clarify the Windows dependency identifiers, as these are the most confusing.
* build: replace glob() check and assume it's always in POSIXwm42017-06-291-1/+1
| | | | | POSIX requires glob(), so no need to check for it. Together with the fact that we can emulate glob() on Windows, glob() is always available.
* terminal: remove unused return valuewm42017-06-273-6/+3
|
* osx: change license of OSX and cocoa files to LGPLwm42017-06-2410-71/+71
| | | | All authors of the current code have agreed.
* w32_common, w32_keyboard: change license to LGPLwm42017-06-242-14/+14
| | | | All authors agreed.
* Revert "osdep: NetBSD pthread_setname_np()"wm42017-06-221-6/+0
| | | | | | | | | | This reverts commit 2e81698d2809836d4cd7f754a78598e7bdf96c0b. Seems like this was a patch applied from someone who can't agree to LGPL relicensing (see previous commit), with the author field not properly set. This is not so important anyway, so just revert it.
* win32: pthread: avoid using TLS, simplify pthread_twm42017-06-152-48/+89
| | | | | | | | | | | | | | | | | Don't use __thread, which requires heavy runtime in some cases (such as MinGW-w64, at least under some configurations, forcing you to link to its pthread runtime DLL). The pthread_t struct was needed over a simple thread ID, because pthread_join() needed to access some sort of context from pthread_t. Further, pthread_exit() and pthread_detach() need the context of the current thread, for which we relied on TLS. Replace these uses by a global thread array. This includes all threads created by the thread wrapper. Hopefully the number of threads created by mpv is low (say, below 20), and threads are not that often created or destroyed. So just keeping them in an array with linear search lookup should be reasonable.
* terminal: change license to LGPLwm42017-06-143-32/+25
| | | | | | | | | | | | | | | | All authors of the current code have agreed. The code probably originates from a software named GySmail (as the copyright header indicates). As far as I can tell, it was written by Arpi (who has agreed), possibly with unknown co-authors. This is most likely OK, as none of the original code is around anymore anyway. I could not find a working download of GySmail, that actually contained the original getch2.c code. This also has a wild history of random people adding ifdef guards to control ioctl() vs. tcgetattr() calls. (See for example 2b1310abba4c1.) Later, the ioctl() was removed in favor of the POSIX tcgetattr(), and the ifdeffery was removed. So these people were not contacted.
* mpv.rc, mpv.exe.manifest: change license to LGPLwm42017-06-121-9/+7
| | | | | | | | | All authors have agreed. mpv.exe.manifest probably can't have a copyright header, so it doesn't. These files don't add license info to the executables. And of course the executable license is still GPL.
* win32: pthread: use the new thread naming APIJames Ross-Gowan2017-05-183-1/+27
| | | | | | | Windows, as of the Creators Update, finally has a sane API for giving a name to a thread that can be used by debuggers. This is similar to pthread_setname_np on Linux and some Unixes. Expose it in the pthread wrapper and use it for mpthread_set_name().
* osdep/compiler.h: change license to LGPLwm42017-05-081-14/+8
| | | | | | | | | | | This didn't have a header and this was implied LGPL, but this was not entirely correct. It turns out the printf attribute code was leaked in from talloc, which is "LGPL v3 or later". talloc was added to the mplayer2 code base in 27a30e310e1. This remaining code was an oversight. Since we want to reduce LGPL v3 code, replace this code with the one from ta/ta.h. This code was explicitly written newly to get rid of talloc and its license in 0933f12d28e.
* osdep/timer*: change license to LGPLwm42017-05-053-23/+21
| | | | | | | | | | | | | | All authors have agreed, with the following exceptions: e68d7f6858: wight wasn't asked (I think...), but even if he modified the patch he applied, all code added by it was removed again later. cb7768f9bb: nick could not be reached, but the declarations he added as well as the full timer.c file were deleted again in cff81fe498 and f800a42e45 (did it really take 6 years to remove unused declarations?). ffaf4af230: it looks like this person wasn't contacted, but the code added was removed again in f544bcf105.
* osx: fix deadlock on exit with libmpv on OSXwm42017-04-191-8/+11
| | | | | | | | | | | There is explicit code to handle the libmpv case, but it expects that a dispatch queue is running. This is not necessarily the case. E.g. edit the simple.c mpv example not to do any playback and to destroy the mpv handle immediately. It will freeze on exit, because nothing will release the mpv_handle. I'm not sure how this should be fixed, so disable it for now in library mode.
* Fix use of ISC licensewm42017-04-154-2/+37
| | | | | | | | | | The license text refers a "above copyright notice", so I guess it'd be good to actually provide such a notice. Add the license to some files that were missing it (since in theory, our Copyright file says that such files are LGPL by default). Remove the questionable remarks about the license in the client API.
* win32: fix mismatched free/talloc_freewm42017-04-121-1/+1
| | | | Might fix #4315.
* win32: rewrite getcwd() using GetFullPathNameWwm42017-04-111-2/+11
| | | | _wgetcwd is apparently not available in all runtimes. Well, whatever.
* win32: add UTF-8 getcwd() wrapperwm42017-04-112-0/+20
|
* osx: fix Touch Bar access on systems without Touch Bar supportAkemi2017-04-021-4/+7
| | | | | | when building with a deployment target older than an SDK with Touch Bar support we still tried to access the TouchBar, since the compile time checks are positive. to prevent this we add two runtime checks.
* osx: fix event propagation of menu bar item key shortcutsAkemi2017-03-271-4/+3
| | | | | | a small logic error in 063ca8f broke the filtering of the menu bar item key shortcuts. they where called several times per event because the propagation wasn't stopped properly.
* osx: fix key input in certain circumstancesAkemi2017-03-265-25/+15
| | | | | | | | | | | | | for a reason i can just assume some key events can vanish from the event chain and mpv seems unresponsive. after quite some testing i could confirm that the events are present at the first entry point of the event chain, the sendEvent method of the Application, and that they vanish at a point afterwards. now we use that entry point to grab keyDown and keyUp events. we also stop propagating those key events to prevent the no key input' error sound. if we ever need the key events somewhere down the event chain we need to start propagating them again. though this is not necessary currently.
* osx: initial Touch Bar supportAkemi2017-03-267-0/+433
|
* osx: add key mappings for previous and next Media KeysAkemi2017-03-091-3/+5
| | | | Fixes #4204
* osx: drop support for OS X 10.7 and earlierAkemi2017-02-272-31/+0
|
* cocoa: fix dragging out of focus windowAkemi2017-02-211-0/+1
| | | | | | | | | | fffab30 introduced a small regression where the cursor couldn't be unhidden after refocusing. the problem is that no mouseUp event was reported in our events_view. work around this with a separate event monitor. this also fixes another regression when the window is being dragged from the title bar. #4174
* osx: improve bundle handlingAkemi2017-02-161-50/+5
| | | | | | | | | | | | | | | | | we have two problems here. first when mpv is started from the bundle it uses its own environment variables and possibly can't find for example the youtube-dl binary for our youtube-dl hook. second we couldn't reliable determine when mpv was started from the bundle, which led to the pseudo-gui usage even when the binary was invoked from a shell. to prevent this we will wrap the bundle binary with a shell script, which will only be called when we start mpv from the bundle. this way we can get the same environment variables, like $PATH, for our bundle and additional we can set the pseudo-gui only when started through this script. it is also possible to detect the bundle usage properly and accurately through the usage of another environment var. Fixes #2061
* cocoa: gracefully quit from the Dock's context menuAkemi2017-02-161-4/+10
| | | | | | | | | | | | quitting mpv from the Dock's context menu leaves the shell in a bad state where you can't see your input any more and other weirdnesses. in a big refactor (afdc9c4) the method to handle this case was actually copied over but the the code to register the event itself was removed or forgotten by accident, leaving some lines of dead code. i re-added the event and slightly adjusted the function. the function was slightly changed so the head of it is consistent with our other events and it is associated with this commit and functionality in the future.
* atomic: remove __atomic builtin usagewm42017-02-131-30/+7
| | | | | | | | Using these was a temporary solution while some compilers implemented the underlying atomic mechanisms, but not the C11 language parts (or that's what I guess). Not really useful for us anymore. Also, there is the slight risk of having subtly incorrect semantics by using potentially changing compiler internals and such.
* vo_opengl: angle: rewrite with custom swap chainJames Ross-Gowan2017-02-071-0/+8
| | | | | | | | | | | | | This replaces the old backend that exclusively used EGL windowing with one that can also use ANGLE's ability to render to directly to a texture. The advantage of this is that it allows mpv to create the swap chain itself and this allows mpv to use a flip-mode swap chain on a HWND (which avoids problems with DirectComposition) and to use a longer swap chain that has six backbuffers by default (which reportedly fixes problems with rendering 24fps video on 24Hz monitors.) Also, "screenshot window" should now work on DXGI 1.2 and up (Windows 8 and up.)
* win32: add COM-specific SAFE_RELEASE to windows_utils.hJames Ross-Gowan2017-01-301-0/+4
| | | | | | | | | | | | | | | See: https://msdn.microsoft.com/en-us/library/windows/desktop/dd743946.aspx Microsoft example code often uses a SAFE_RELEASE macro like the one in the above link. This makes it easier to avoid errors when releasing COM interfaces. It also reduces noise in COM-heavy code. ao_wasapi.h also had a macro called SAFE_RELEASE, though unlike the version above, its SAFE_RELEASE macro accepted a second parameter which allowed it to destroy arbitrary objects other than just COM interfaces. This renames ao_wasapi's SAFE_RELEASE to SAFE_DESTROY, which should more accurately reflect what it does and prevent confusion with the Microsoft version.
* atomic: drop __sync builtinswm42017-01-271-19/+0
| | | | | | | | | | The correctness of the stdatomic.h emulation via the __sync builtins is questionable, and we've been relying on exact stdatomic semantics for a while, so just get rid of it. Compilers which support __sync but not stdatomic.h will use to the slow mutex fallback. Not sure about the __atomic builtins. It doesn't seem to harm either, so leave it for now.
* Update copyright yearwm42017-01-011-1/+1
| | | | What kind of bullshit forces you to do this every year anyway.
* cocoa: cosmetic fixesAkemi2016-12-162-6/+11
|
* cocoa: fullscreen refactoringAkemi2016-12-151-0/+1
| | | | | | | | | | this replaces the old fullscreen with the native macOS fullscreen. additional the --fs-black-out-screens was removed since the new API doesn't support it in a way the old one did. it can possibly be re-added if done manually. Fixes #2857 #3272 #1352 #2062 #3864
* cocoa: support append file to paylist on dropAkemi2016-12-141-1/+4
| | | | | | | | | When dropping a file on mpv, either on the window or the App bundle icon, while holding the shift key the dropped files will be appended to the playlist. Fixes #2166
* osx: update the menu and remove conflicting itemAkemi2016-12-121-7/+5
| | | | | | | | | | | Remove 'Quit mpv & remember playback position' from the menu because it conflicts with the global logout shortcut. add separator between 'Hide' and 'Quit' for consistency with other Apps. also rename the 'Movie' menu to 'Video'. it's a bit more generic. Fixes #3865
* win32: fix some Clang warningsJames Ross-Gowan2016-11-171-3/+3
| | | | | | | | - win32-console-wrapper.c was inconsistently using the explicit Unicode versions of some Windows API functions and structures. - vo.c should use llabs for int64_t, since long is 32-bit on Windows. - vo_direct3d.c had a potential use of an uninitialized variable if it took the first goto error_exit.
* build: fix compilation with mingw-w64/ClangJames Ross-Gowan2016-11-171-1/+1
|