mpv/video/out/x11_common.c

1747 lines
55 KiB
C
Raw Normal View History

/*
* This file is part of MPlayer.
*
* MPlayer is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* MPlayer is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with MPlayer; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*/
#include <stdio.h>
#include <stdlib.h>
#include <math.h>
#include <inttypes.h>
#include <limits.h>
#include "config.h"
#include "bstr/bstr.h"
#include "options/options.h"
#include "common/common.h"
#include "common/msg.h"
2013-12-17 00:23:09 +00:00
#include "input/input.h"
#include "input/event.h"
#include "x11_common.h"
#include "talloc.h"
#include <string.h>
#include <unistd.h>
#include <assert.h>
#include "vo.h"
#include "win_state.h"
#include "osdep/timer.h"
#include <X11/Xmd.h>
#include <X11/Xlib.h>
#include <X11/Xutil.h>
#include <X11/Xatom.h>
#include <X11/keysym.h>
#include <X11/XKBlib.h>
#if HAVE_XSS
#include <X11/extensions/scrnsaver.h>
#endif
#if HAVE_XEXT
#include <X11/extensions/dpms.h>
#endif
#if HAVE_XINERAMA
#include <X11/extensions/Xinerama.h>
#endif
#if HAVE_XF86VM
#include <X11/extensions/xf86vmode.h>
#endif
#if HAVE_XF86XK
#include <X11/XF86keysym.h>
#endif
#if HAVE_ZLIB
#include <zlib.h>
#endif
2013-12-17 00:23:09 +00:00
#include "input/input.h"
#include "input/keycodes.h"
#define vo_wm_LAYER 1
#define vo_wm_FULLSCREEN 2
#define vo_wm_STAYS_ON_TOP 4
#define vo_wm_ABOVE 8
#define vo_wm_BELOW 16
#define vo_wm_NETWM (vo_wm_FULLSCREEN | vo_wm_STAYS_ON_TOP | vo_wm_ABOVE | \
vo_wm_BELOW)
/* EWMH state actions, see
http://freedesktop.org/Standards/wm-spec/index.html#id2768769 */
#define _NET_WM_STATE_REMOVE 0 /* remove/unset property */
#define _NET_WM_STATE_ADD 1 /* add/set property */
#define _NET_WM_STATE_TOGGLE 2 /* toggle property */
#define WIN_LAYER_ONBOTTOM 2
#define WIN_LAYER_NORMAL 4
#define WIN_LAYER_ONTOP 6
#define WIN_LAYER_ABOVE_DOCK 10
#define DND_VERSION 5
// ----- Motif header: -------
#define MWM_HINTS_FUNCTIONS (1L << 0)
#define MWM_HINTS_DECORATIONS (1L << 1)
#define MWM_FUNC_RESIZE (1L << 1)
#define MWM_FUNC_MOVE (1L << 2)
#define MWM_FUNC_MINIMIZE (1L << 3)
#define MWM_FUNC_MAXIMIZE (1L << 4)
#define MWM_FUNC_CLOSE (1L << 5)
#define MWM_DECOR_ALL (1L << 0)
typedef struct
{
long flags;
long functions;
long decorations;
long input_mode;
long state;
} MotifWmHints;
static const char x11_icon[] =
#include "video/out/x11_icon.inc"
;
static struct mp_log *x11_error_output;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
static void vo_x11_update_geometry(struct vo *vo);
2013-05-16 12:02:28 +00:00
static void vo_x11_fullscreen(struct vo *vo);
static void xscreensaver_heartbeat(struct vo_x11_state *x11);
static void saver_on(struct vo_x11_state *x11);
static void saver_off(struct vo_x11_state *x11);
static void vo_x11_selectinput_witherr(struct vo *vo, Display *display,
Window w, long event_mask);
static void vo_x11_setlayer(struct vo *vo, Window vo_window, int layer);
/*
* Sends the EWMH fullscreen state event.
*
* action: could be one of _NET_WM_STATE_REMOVE -- remove state
* _NET_WM_STATE_ADD -- add state
* _NET_WM_STATE_TOGGLE -- toggle
*/
static void vo_x11_ewmh_fullscreen(struct vo_x11_state *x11, int action)
{
assert(action == _NET_WM_STATE_REMOVE || action == _NET_WM_STATE_ADD ||
action == _NET_WM_STATE_TOGGLE);
if (x11->wm_type & vo_wm_FULLSCREEN) {
XEvent xev;
/* init X event structure for _NET_WM_FULLSCREEN client message */
xev.xclient.type = ClientMessage;
xev.xclient.serial = 0;
xev.xclient.send_event = True;
xev.xclient.message_type = x11->XA_NET_WM_STATE;
2008-04-21 00:14:37 +00:00
xev.xclient.window = x11->window;
xev.xclient.format = 32;
xev.xclient.data.l[0] = action;
xev.xclient.data.l[1] = x11->XA_NET_WM_STATE_FULLSCREEN;
xev.xclient.data.l[2] = 0;
xev.xclient.data.l[3] = 0;
xev.xclient.data.l[4] = 0;
/* finally send that damn thing */
2008-04-21 00:14:37 +00:00
if (!XSendEvent(x11->display, DefaultRootWindow(x11->display), False,
SubstructureRedirectMask | SubstructureNotifyMask,
&xev))
{
MP_ERR(x11, "Couldn't send EWMH fullscreen event!\n");
}
}
}
static void vo_set_cursor_hidden(struct vo *vo, bool cursor_hidden)
{
Cursor no_ptr;
Pixmap bm_no;
XColor black, dummy;
Colormap colormap;
const char bm_no_data[] = {0, 0, 0, 0, 0, 0, 0, 0};
struct vo_x11_state *x11 = vo->x11;
Display *disp = x11->display;
Window win = x11->window;
if (cursor_hidden == x11->mouse_cursor_hidden)
return;
x11->mouse_cursor_hidden = cursor_hidden;
if (vo->opts->WinID == 0 || win == 0)
return; // do not hide if playing on the root window
if (x11->mouse_cursor_hidden) {
colormap = DefaultColormap(disp, DefaultScreen(disp));
if (!XAllocNamedColor(disp, colormap, "black", &black, &dummy))
return; // color alloc failed, give up
bm_no = XCreateBitmapFromData(disp, win, bm_no_data, 8, 8);
no_ptr = XCreatePixmapCursor(disp, bm_no, bm_no, &black, &black, 0, 0);
XDefineCursor(disp, win, no_ptr);
XFreeCursor(disp, no_ptr);
if (bm_no != None)
XFreePixmap(disp, bm_no);
XFreeColors(disp, colormap, &black.pixel, 1, 0);
} else {
XDefineCursor(x11->display, x11->window, 0);
}
}
static int x11_errorhandler(Display *display, XErrorEvent *event)
{
struct mp_log *log = x11_error_output;
char msg[60];
XGetErrorText(display, event->error_code, (char *) &msg, sizeof(msg));
mp_err(log, "X11 error: %s\n", msg);
mp_verbose(log, "Type: %x, display: %p, resourceid: %lx, serial: %lx\n",
event->type, event->display, event->resourceid, event->serial);
mp_verbose(log, "Error code: %x, request code: %x, minor code: %x\n",
event->error_code, event->request_code, event->minor_code);
return 0;
}
2008-04-20 23:50:40 +00:00
static int net_wm_support_state_test(struct vo_x11_state *x11, Atom atom)
{
#define NET_WM_STATE_TEST(x) { \
if (atom == x11->XA_NET_WM_STATE_##x) { \
MP_VERBOSE(x11, "Detected wm supports " #x " state.\n" ); \
return vo_wm_##x; \
} \
}
NET_WM_STATE_TEST(FULLSCREEN);
NET_WM_STATE_TEST(ABOVE);
NET_WM_STATE_TEST(STAYS_ON_TOP);
NET_WM_STATE_TEST(BELOW);
return 0;
}
static bool x11_get_property(struct vo_x11_state *x11, Window wnd, Atom type,
Atom **args, unsigned long *nitems)
{
int format;
unsigned long bytesafter;
return Success ==
XGetWindowProperty(x11->display, wnd, type, 0, 16384, False,
AnyPropertyType, &type, &format, nitems,
&bytesafter, (unsigned char **) args)
&& *nitems > 0 && bytesafter == 0;
}
static int vo_wm_detect(struct vo *vo)
{
2008-04-20 23:50:40 +00:00
struct vo_x11_state *x11 = vo->x11;
int i;
int wm = 0;
unsigned long nitems;
Atom *args = NULL;
Window win = x11->rootwin;
if (vo->opts->WinID >= 0)
return 0;
// -- supports layers
if (x11_get_property(x11, win, x11->XA_WIN_PROTOCOLS, &args, &nitems)) {
MP_VERBOSE(x11, "Detected wm supports layers.\n");
for (i = 0; i < nitems; i++) {
if (args[i] == x11->XA_WIN_LAYER)
wm |= vo_wm_LAYER;
}
XFree(args);
}
// --- netwm
if (x11_get_property(x11, win, x11->XA_NET_SUPPORTED, &args, &nitems)) {
MP_VERBOSE(x11, "Detected wm supports NetWM.\n");
if (vo->opts->x11_netwm) {
for (i = 0; i < nitems; i++)
wm |= net_wm_support_state_test(vo->x11, args[i]);
} else {
MP_VERBOSE(x11, "NetWM usage disabled by user.\n");
}
XFree(args);
}
if (wm == 0)
MP_VERBOSE(x11, "Unknown wm type...\n");
return wm;
}
#define XA_INIT(x) x11->XA ## x = XInternAtom(x11->display, # x, False)
2008-04-20 23:50:40 +00:00
static void init_atoms(struct vo_x11_state *x11)
{
XA_INIT(_NET_SUPPORTED);
XA_INIT(_NET_WM_STATE);
XA_INIT(_NET_WM_STATE_FULLSCREEN);
XA_INIT(_NET_WM_STATE_ABOVE);
XA_INIT(_NET_WM_STATE_STAYS_ON_TOP);
XA_INIT(_NET_WM_STATE_BELOW);
XA_INIT(_NET_WM_PID);
XA_INIT(_NET_WM_NAME);
XA_INIT(_NET_WM_ICON_NAME);
XA_INIT(_NET_WM_ICON);
XA_INIT(_NET_WM_MOVERESIZE);
XA_INIT(_WIN_PROTOCOLS);
XA_INIT(_WIN_LAYER);
XA_INIT(_WIN_HINTS);
XA_INIT(WM_PROTOCOLS);
XA_INIT(WM_DELETE_WINDOW);
XA_INIT(UTF8_STRING);
XA_INIT(TARGETS);
XA_INIT(XdndAware);
XA_INIT(XdndEnter);
XA_INIT(XdndLeave);
XA_INIT(XdndPosition);
XA_INIT(XdndStatus);
XA_INIT(XdndActionCopy);
XA_INIT(XdndTypeList);
XA_INIT(XdndDrop);
XA_INIT(XdndSelection);
XA_INIT(XdndFinished);
char buf[50];
sprintf(buf, "_NET_WM_CM_S%d", x11->screen);
x11->XA_NET_WM_CM = XInternAtom(x11->display, buf, False);
x11->XA_uri_list = XInternAtom(x11->display, "text/uri-list", False);
}
2013-05-16 12:02:28 +00:00
static void vo_x11_update_screeninfo(struct vo *vo)
{
struct mp_vo_opts *opts = vo->opts;
struct vo_x11_state *x11 = vo->x11;
bool all_screens = opts->fullscreen && opts->fsscreen_id == -2;
x11->screenrc = (struct mp_rect){.x1 = x11->ws_width, .y1 = x11->ws_height};
#if HAVE_XINERAMA
2013-10-30 20:58:55 +00:00
if (opts->screen_id >= -1 && XineramaIsActive(x11->display) && !all_screens)
{
int screen = opts->fullscreen ? opts->fsscreen_id : opts->screen_id;
XineramaScreenInfo *screens;
int num_screens;
screens = XineramaQueryScreens(x11->display, &num_screens);
if (screen >= num_screens)
screen = num_screens - 1;
if (screen == -1) {
int x = x11->win_x + x11->win_width / 2;
int y = x11->win_y + x11->win_height / 2;
for (screen = num_screens - 1; screen > 0; screen--) {
int left = screens[screen].x_org;
int right = left + screens[screen].width;
int top = screens[screen].y_org;
int bottom = top + screens[screen].height;
if (left <= x && x <= right && top <= y && y <= bottom)
break;
}
}
if (screen < 0)
screen = 0;
x11->screenrc = (struct mp_rect){
.x0 = screens[screen].x_org,
.y0 = screens[screen].y_org,
.x1 = screens[screen].x_org + screens[screen].width,
.y1 = screens[screen].y_org + screens[screen].height,
};
XFree(screens);
}
#endif
}
int vo_x11_init(struct vo *vo)
{
struct mp_vo_opts *opts = vo->opts;
char *dispName;
assert(!vo->x11);
struct vo_x11_state *x11 = talloc_ptrtype(NULL, x11);
*x11 = (struct vo_x11_state){
.log = mp_log_new(x11, vo->log, "x11"),
.olddecor = MWM_DECOR_ALL,
.oldfuncs = MWM_FUNC_MOVE | MWM_FUNC_CLOSE | MWM_FUNC_MINIMIZE |
MWM_FUNC_MAXIMIZE | MWM_FUNC_RESIZE,
.old_gravity = NorthWestGravity,
};
vo->x11 = x11;
x11_error_output = x11->log;
XSetErrorHandler(x11_errorhandler);
dispName = XDisplayName(NULL);
MP_VERBOSE(x11, "X11 opening display: %s\n", dispName);
x11->display = XOpenDisplay(dispName);
if (!x11->display) {
MP_MSG(x11, vo->probing ? MSGL_V : MSGL_ERR,
"couldn't open the X11 display (%s)!\n", dispName);
x11_error_output = NULL;
XSetErrorHandler(NULL);
talloc_free(x11);
vo->x11 = NULL;
return 0;
}
x11->screen = DefaultScreen(x11->display); // screen ID
x11->rootwin = RootWindow(x11->display, x11->screen); // root window ID
if (!opts->native_keyrepeat) {
Bool ok = False;
XkbSetDetectableAutoRepeat(x11->display, True, &ok);
x11->no_autorepeat = ok;
}
x11->xim = XOpenIM(x11->display, NULL, NULL, NULL);
if (!x11->xim)
MP_WARN(x11, "XOpenIM() failed. Unicode input will not work.\n");
2008-04-20 23:50:40 +00:00
init_atoms(vo->x11);
x11->ws_width = DisplayWidth(x11->display, x11->screen);
x11->ws_height = DisplayHeight(x11->display, x11->screen);
if (strncmp(dispName, "unix:", 5) == 0)
dispName += 4;
else if (strncmp(dispName, "localhost:", 10) == 0)
dispName += 9;
2013-10-30 20:58:55 +00:00
x11->display_is_local = dispName[0] == ':' && atoi(dispName + 1) < 10;
MP_VERBOSE(x11, "X11 running at %dx%d (\"%s\" => %s display)\n",
x11->ws_width, x11->ws_height, dispName,
x11->display_is_local ? "local" : "remote");
x11->wm_type = vo_wm_detect(vo);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo->event_fd = ConnectionNumber(x11->display);
return 1;
}
static const struct mp_keymap keymap[] = {
// special keys
{XK_Pause, MP_KEY_PAUSE}, {XK_Escape, MP_KEY_ESC},
{XK_BackSpace, MP_KEY_BS}, {XK_Tab, MP_KEY_TAB}, {XK_Return, MP_KEY_ENTER},
{XK_Menu, MP_KEY_MENU}, {XK_Print, MP_KEY_PRINT},
// cursor keys
{XK_Left, MP_KEY_LEFT}, {XK_Right, MP_KEY_RIGHT}, {XK_Up, MP_KEY_UP},
{XK_Down, MP_KEY_DOWN},
// navigation block
{XK_Insert, MP_KEY_INSERT}, {XK_Delete, MP_KEY_DELETE},
{XK_Home, MP_KEY_HOME}, {XK_End, MP_KEY_END}, {XK_Page_Up, MP_KEY_PAGE_UP},
{XK_Page_Down, MP_KEY_PAGE_DOWN},
// F-keys
{XK_F1, MP_KEY_F+1}, {XK_F2, MP_KEY_F+2}, {XK_F3, MP_KEY_F+3},
{XK_F4, MP_KEY_F+4}, {XK_F5, MP_KEY_F+5}, {XK_F6, MP_KEY_F+6},
{XK_F7, MP_KEY_F+7}, {XK_F8, MP_KEY_F+8}, {XK_F9, MP_KEY_F+9},
{XK_F10, MP_KEY_F+10}, {XK_F11, MP_KEY_F+11}, {XK_F12, MP_KEY_F+12},
// numpad independent of numlock
{XK_KP_Subtract, '-'}, {XK_KP_Add, '+'}, {XK_KP_Multiply, '*'},
{XK_KP_Divide, '/'}, {XK_KP_Enter, MP_KEY_KPENTER},
// numpad with numlock
{XK_KP_0, MP_KEY_KP0}, {XK_KP_1, MP_KEY_KP1}, {XK_KP_2, MP_KEY_KP2},
{XK_KP_3, MP_KEY_KP3}, {XK_KP_4, MP_KEY_KP4}, {XK_KP_5, MP_KEY_KP5},
{XK_KP_6, MP_KEY_KP6}, {XK_KP_7, MP_KEY_KP7}, {XK_KP_8, MP_KEY_KP8},
{XK_KP_9, MP_KEY_KP9}, {XK_KP_Decimal, MP_KEY_KPDEC},
{XK_KP_Separator, MP_KEY_KPDEC},
// numpad without numlock
{XK_KP_Insert, MP_KEY_KPINS}, {XK_KP_End, MP_KEY_KP1},
{XK_KP_Down, MP_KEY_KP2}, {XK_KP_Page_Down, MP_KEY_KP3},
{XK_KP_Left, MP_KEY_KP4}, {XK_KP_Begin, MP_KEY_KP5},
{XK_KP_Right, MP_KEY_KP6}, {XK_KP_Home, MP_KEY_KP7}, {XK_KP_Up, MP_KEY_KP8},
{XK_KP_Page_Up, MP_KEY_KP9}, {XK_KP_Delete, MP_KEY_KPDEL},
#ifdef XF86XK_AudioPause
{XF86XK_MenuKB, MP_KEY_MENU},
{XF86XK_AudioPlay, MP_KEY_PLAY}, {XF86XK_AudioPause, MP_KEY_PAUSE},
{XF86XK_AudioStop, MP_KEY_STOP}, {XF86XK_AudioPrev, MP_KEY_PREV},
{XF86XK_AudioNext, MP_KEY_NEXT}, {XF86XK_AudioMute, MP_KEY_MUTE},
{XF86XK_AudioLowerVolume, MP_KEY_VOLUME_DOWN},
{XF86XK_AudioRaiseVolume, MP_KEY_VOLUME_UP},
{XF86XK_HomePage, MP_KEY_HOMEPAGE}, {XF86XK_WWW, MP_KEY_WWW},
{XF86XK_Mail, MP_KEY_MAIL}, {XF86XK_Favorites, MP_KEY_FAVORITES},
{XF86XK_Search, MP_KEY_SEARCH}, {XF86XK_Sleep, MP_KEY_SLEEP},
#endif
{0, 0}
};
static int vo_x11_lookupkey(int key)
{
static const char *passthrough_keys = " -+*/<>`~!@#$%^&()_{}:;\"\',.?\\|=[]";
int mpkey = 0;
if ((key >= 'a' && key <= 'z') ||
(key >= 'A' && key <= 'Z') ||
(key >= '0' && key <= '9') ||
(key > 0 && key < 256 && strchr(passthrough_keys, key)))
mpkey = key;
if (!mpkey)
mpkey = lookup_keymap_table(keymap, key);
return mpkey;
}
static void vo_x11_decoration(struct vo *vo, int d)
{
struct vo_x11_state *x11 = vo->x11;
Atom mtype;
int mformat;
unsigned long mn, mb;
Atom vo_MotifHints;
MotifWmHints vo_MotifWmHints;
if (!vo->opts->WinID)
return;
vo_MotifHints = XInternAtom(x11->display, "_MOTIF_WM_HINTS", 0);
if (vo_MotifHints != None) {
if (!x11->got_motif_hints) {
MotifWmHints *mhints = NULL;
XGetWindowProperty(x11->display, x11->window,
vo_MotifHints, 0, 20, False,
vo_MotifHints, &mtype, &mformat, &mn,
&mb, (unsigned char **) &mhints);
if (mhints) {
if (mhints->flags & MWM_HINTS_DECORATIONS)
x11->olddecor = mhints->decorations;
if (mhints->flags & MWM_HINTS_FUNCTIONS)
x11->oldfuncs = mhints->functions;
XFree(mhints);
}
}
x11->got_motif_hints = true;
memset(&vo_MotifWmHints, 0, sizeof(MotifWmHints));
vo_MotifWmHints.flags =
MWM_HINTS_FUNCTIONS | MWM_HINTS_DECORATIONS;
if (d) {
vo_MotifWmHints.functions = x11->oldfuncs;
d = x11->olddecor;
}
vo_MotifWmHints.decorations = d;
XChangeProperty(x11->display, x11->window, vo_MotifHints,
vo_MotifHints, 32,
PropModeReplace,
(unsigned char *) &vo_MotifWmHints,
5);
}
}
static void vo_x11_classhint(struct vo *vo, Window window, const char *name)
{
struct mp_vo_opts *opts = vo->opts;
2008-04-20 23:50:40 +00:00
struct vo_x11_state *x11 = vo->x11;
XClassHint wmClass;
pid_t pid = getpid();
wmClass.res_name = opts->winname ? opts->winname : (char *)name;
wmClass.res_class = "mpv";
2008-04-20 23:50:40 +00:00
XSetClassHint(x11->display, window, &wmClass);
XChangeProperty(x11->display, window, x11->XA_NET_WM_PID, XA_CARDINAL,
32, PropModeReplace, (unsigned char *) &pid, 1);
}
void vo_x11_uninit(struct vo *vo)
{
2008-04-21 00:14:37 +00:00
struct vo_x11_state *x11 = vo->x11;
assert(x11);
mp_input_put_key(vo->input_ctx, MP_INPUT_RELEASE_ALL);
saver_on(x11);
2008-04-21 00:14:37 +00:00
if (x11->window != None)
vo_set_cursor_hidden(vo, false);
if (x11->f_gc != None)
2008-04-21 01:19:43 +00:00
XFreeGC(vo->x11->display, x11->f_gc);
if (x11->vo_gc != None)
XFreeGC(vo->x11->display, x11->vo_gc);
if (x11->window != None && x11->window != x11->rootwin) {
XClearWindow(x11->display, x11->window);
XUnmapWindow(x11->display, x11->window);
XSelectInput(x11->display, x11->window, StructureNotifyMask);
XDestroyWindow(x11->display, x11->window);
XEvent xev;
do {
XNextEvent(x11->display, &xev);
} while (xev.type != DestroyNotify ||
2013-10-30 20:58:55 +00:00
xev.xdestroywindow.event != x11->window);
}
if (x11->xic)
XDestroyIC(x11->xic);
if (x11->colormap != None)
XFreeColormap(vo->x11->display, x11->colormap);
MP_VERBOSE(x11, "uninit ...\n");
if (x11->xim)
XCloseIM(x11->xim);
x11_error_output = NULL;
XSetErrorHandler(NULL);
XCloseDisplay(x11->display);
talloc_free(x11);
vo->x11 = NULL;
}
static void vo_x11_dnd_init_window(struct vo *vo)
{
struct vo_x11_state *x11 = vo->x11;
Atom version = DND_VERSION;
XChangeProperty(x11->display, x11->window, x11->XAXdndAware, XA_ATOM,
32, PropModeReplace, (unsigned char *)&version, 1);
x11->dnd_property = XInternAtom(x11->display, "mpv_dnd_selection", False);
}
static void dnd_select_format(struct vo_x11_state *x11, Atom *args, int items)
{
for (int n = 0; n < items; n++) {
// There are other types; possibly not worth supporting.
if (args[n] == x11->XA_uri_list)
x11->dnd_requested_format = args[n];
}
}
static void dnd_reset(struct vo *vo)
{
struct vo_x11_state *x11 = vo->x11;
x11->dnd_src_window = 0;
x11->dnd_requested_format = 0;
}
static void vo_x11_dnd_handle_message(struct vo *vo, XClientMessageEvent *ce)
{
struct vo_x11_state *x11 = vo->x11;
if (!x11->window)
return;
if (ce->message_type == x11->XAXdndEnter) {
x11->dnd_requested_format = 0;
Window src = ce->data.l[0];
if (ce->data.l[1] & 1) {
unsigned long nitems = 0;
Atom *args = NULL;
if (x11_get_property(x11, src, x11->XAXdndTypeList, &args, &nitems)) {
dnd_select_format(x11, args, nitems);
XFree(args);
}
} else {
Atom args[3];
for (int n = 2; n <= 4; n++)
args[n - 2] = ce->data.l[n];
dnd_select_format(x11, args, 3);
}
} else if (ce->message_type == x11->XAXdndPosition) {
Window src = ce->data.l[0];
XEvent xev;
xev.xclient.type = ClientMessage;
xev.xclient.serial = 0;
xev.xclient.send_event = True;
xev.xclient.message_type = x11->XAXdndStatus;
xev.xclient.window = src;
xev.xclient.format = 32;
xev.xclient.data.l[0] = x11->window;
xev.xclient.data.l[1] = x11->dnd_requested_format ? 1 : 0;
xev.xclient.data.l[2] = 0;
xev.xclient.data.l[3] = 0;
xev.xclient.data.l[4] = x11->XAXdndActionCopy;
XSendEvent(x11->display, src, False, 0, &xev);
} else if (ce->message_type == x11->XAXdndDrop) {
x11->dnd_src_window = ce->data.l[0];
XConvertSelection(x11->display, x11->XAXdndSelection,
x11->dnd_requested_format, x11->dnd_property,
x11->window, ce->data.l[2]);
} else if (ce->message_type == x11->XAXdndLeave) {
dnd_reset(vo);
}
}
static void vo_x11_dnd_handle_selection(struct vo *vo, XSelectionEvent *se)
{
struct vo_x11_state *x11 = vo->x11;
if (!x11->window || !x11->dnd_src_window)
return;
bool success = false;
if (se->selection == x11->XAXdndSelection &&
se->property == x11->dnd_property &&
se->target == x11->dnd_requested_format)
{
Atom type;
int format;
unsigned long nitems;
unsigned long bytes_left;
unsigned char *prop;
if (XGetWindowProperty(x11->display, x11->window, x11->dnd_property,
0, 64 * 1024, False, x11->dnd_requested_format,
&type, &format, &nitems, &bytes_left, &prop)
== Success)
{
if (!bytes_left && type == x11->dnd_requested_format && format == 8)
{
// No idea if this is guaranteed to be \0-padded, so use bstr.
success = mp_event_drop_mime_data(vo->input_ctx, "text/uri-list",
(bstr){prop, nitems}) > 0;
}
XFree(prop);
}
}
XEvent xev;
xev.xclient.type = ClientMessage;
xev.xclient.serial = 0;
xev.xclient.send_event = True;
xev.xclient.message_type = x11->XAXdndFinished;
xev.xclient.window = x11->dnd_src_window;
xev.xclient.format = 32;
xev.xclient.data.l[0] = x11->window;
xev.xclient.data.l[1] = success ? 1 : 0;
xev.xclient.data.l[2] = success ? x11->XAXdndActionCopy : 0;
xev.xclient.data.l[3] = 0;
xev.xclient.data.l[4] = 0;
XSendEvent(x11->display, x11->dnd_src_window, False, 0, &xev);
dnd_reset(vo);
}
static void update_vo_size(struct vo *vo)
{
struct vo_x11_state *x11 = vo->x11;
if (x11->win_width != vo->dwidth || x11->win_height != vo->dheight) {
vo->dwidth = x11->win_width;
vo->dheight = x11->win_height;
x11->pending_vo_events |= VO_EVENT_RESIZE;
}
}
static int get_mods(unsigned int state)
{
int modifiers = 0;
if (state & ShiftMask)
modifiers |= MP_KEY_MODIFIER_SHIFT;
if (state & ControlMask)
modifiers |= MP_KEY_MODIFIER_CTRL;
if (state & Mod1Mask)
modifiers |= MP_KEY_MODIFIER_ALT;
if (state & Mod4Mask)
modifiers |= MP_KEY_MODIFIER_META;
return modifiers;
}
int vo_x11_check_events(struct vo *vo)
{
2008-04-20 23:50:40 +00:00
struct vo_x11_state *x11 = vo->x11;
Display *display = vo->x11->display;
XEvent Event;
xscreensaver_heartbeat(vo->x11);
if (vo->opts->WinID > 0)
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_update_geometry(vo);
while (XPending(display)) {
XNextEvent(display, &Event);
switch (Event.type) {
case Expose:
x11->pending_vo_events |= VO_EVENT_EXPOSE;
break;
case ConfigureNotify:
if (x11->window == None)
break;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_update_geometry(vo);
break;
case KeyPress: {
char buf[100];
KeySym keySym = 0;
int modifiers = get_mods(Event.xkey.state);
if (x11->no_autorepeat)
modifiers |= MP_KEY_STATE_DOWN;
if (x11->xic) {
Status status;
int len = Xutf8LookupString(x11->xic, &Event.xkey, buf,
sizeof(buf), &keySym, &status);
int mpkey = vo_x11_lookupkey(keySym);
if (mpkey) {
mp_input_put_key(vo->input_ctx, mpkey | modifiers);
} else if (status == XLookupChars || status == XLookupBoth) {
struct bstr t = { buf, len };
mp_input_put_key_utf8(vo->input_ctx, modifiers, t);
}
} else {
XLookupString(&Event.xkey, buf, sizeof(buf), &keySym,
&x11->compose_status);
int mpkey = vo_x11_lookupkey(keySym);
if (mpkey)
mp_input_put_key(vo->input_ctx, mpkey | modifiers);
}
break;
}
// Releasing all keys in these situations is simpler and ensures no
// keys can be get "stuck".
case FocusOut:
case KeyRelease:
{
if (x11->no_autorepeat)
mp_input_put_key(vo->input_ctx, MP_INPUT_RELEASE_ALL);
x11->win_drag_button1_down = false;
break;
}
case MotionNotify:
if (x11->win_drag_button1_down && !x11->fs &&
!mp_input_test_dragging(vo->input_ctx, Event.xmotion.x,
Event.xmotion.y))
{
mp_input_put_key(vo->input_ctx, MP_INPUT_RELEASE_ALL);
XUngrabPointer(x11->display, CurrentTime);
XEvent xev;
xev.xclient.type = ClientMessage;
xev.xclient.serial = 0;
xev.xclient.send_event = True;
xev.xclient.message_type = x11->XA_NET_WM_MOVERESIZE;
xev.xclient.window = x11->window;
xev.xclient.format = 32;
xev.xclient.data.l[0] = Event.xmotion.x_root;
xev.xclient.data.l[1] = Event.xmotion.y_root;
xev.xclient.data.l[2] = 8; // _NET_WM_MOVERESIZE_MOVE
xev.xclient.data.l[3] = 1; // button 1
xev.xclient.data.l[4] = 1; // source indication: normal
XSendEvent(x11->display, x11->rootwin, False,
SubstructureRedirectMask | SubstructureNotifyMask,
&xev);
} else {
vo_mouse_movement(vo, Event.xmotion.x, Event.xmotion.y);
}
x11->win_drag_button1_down = false;
break;
case LeaveNotify:
x11->win_drag_button1_down = false;
mp_input_put_key(vo->input_ctx, MP_KEY_MOUSE_LEAVE);
break;
case ButtonPress:
if (Event.xbutton.button == 1)
x11->win_drag_button1_down = true;
mp_input_put_key(vo->input_ctx,
(MP_MOUSE_BTN0 + Event.xbutton.button - 1) |
get_mods(Event.xbutton.state) | MP_KEY_STATE_DOWN);
break;
case ButtonRelease:
if (Event.xbutton.button == 1)
x11->win_drag_button1_down = false;
mp_input_put_key(vo->input_ctx,
(MP_MOUSE_BTN0 + Event.xbutton.button - 1) |
get_mods(Event.xbutton.state) | MP_KEY_STATE_UP);
break;
case MapNotify:
x11: fix race condition when setting aspect when leaving fullscreen vo_opengl creates a hidden X11 window to probe the OpenGL context. It must do that before creating a visible window, because VO creation and VO config are separate phases. There's a race condition involving the hidden window: when starting with --fs, and then leaving fullscreen, the unfullscreened window is sometimes set to the aspect ratio of the hidden window. I'm not sure why the window size itself uses the correct size (but corrupted by the wrong aspect), but that's perhaps because the window manager is free to ignore the size hint while honoring the aspect, or something equally messed up. It turns out this happens because x11_common.c thinks the size of the hidden window is the size of the unfullscreened window. This in turn happens because vo_x11_update_geometry() reads the size of the hidden window when called in vo_x11_fullscreen() (called from vo_x11_config_vo_window()) when mapping the fullscreen window. At that point, the window could be mapped, but not necessarily. If it's not mapped, it will get the size of the unfullscreened window... I think. One could fix this by actively waiting until the window is mapped. Try to pick a less hacky approach instead, and never read the window size until MapNotify is received. vo_x11_create_window() needs a hack, because we'd possibly set the VO's size to 0, resulting e.g. in vdpau to fail initialization. (It'll print error messages until a proper resize is received.)
2014-02-02 00:35:46 +00:00
x11->window_hidden = false;
vo_x11_clearwindow(vo, x11->window);
x11: fix race condition when setting aspect when leaving fullscreen vo_opengl creates a hidden X11 window to probe the OpenGL context. It must do that before creating a visible window, because VO creation and VO config are separate phases. There's a race condition involving the hidden window: when starting with --fs, and then leaving fullscreen, the unfullscreened window is sometimes set to the aspect ratio of the hidden window. I'm not sure why the window size itself uses the correct size (but corrupted by the wrong aspect), but that's perhaps because the window manager is free to ignore the size hint while honoring the aspect, or something equally messed up. It turns out this happens because x11_common.c thinks the size of the hidden window is the size of the unfullscreened window. This in turn happens because vo_x11_update_geometry() reads the size of the hidden window when called in vo_x11_fullscreen() (called from vo_x11_config_vo_window()) when mapping the fullscreen window. At that point, the window could be mapped, but not necessarily. If it's not mapped, it will get the size of the unfullscreened window... I think. One could fix this by actively waiting until the window is mapped. Try to pick a less hacky approach instead, and never read the window size until MapNotify is received. vo_x11_create_window() needs a hack, because we'd possibly set the VO's size to 0, resulting e.g. in vdpau to fail initialization. (It'll print error messages until a proper resize is received.)
2014-02-02 00:35:46 +00:00
vo_x11_update_geometry(vo);
x11->vo_hint.win_gravity = x11->old_gravity;
XSetWMNormalHints(display, x11->window, &x11->vo_hint);
break;
case DestroyNotify:
MP_WARN(x11, "Our window was destroyed, exiting\n");
mp_input_put_key(vo->input_ctx, MP_KEY_CLOSE_WIN);
break;
case ClientMessage:
if (Event.xclient.message_type == x11->XAWM_PROTOCOLS &&
Event.xclient.data.l[0] == x11->XAWM_DELETE_WINDOW)
mp_input_put_key(vo->input_ctx, MP_KEY_CLOSE_WIN);
vo_x11_dnd_handle_message(vo, &Event.xclient);
break;
case SelectionNotify:
vo_x11_dnd_handle_selection(vo, &Event.xselection);
break;
default:
if (Event.type == x11->ShmCompletionEvent) {
if (x11->ShmCompletionWaitCount > 0)
x11->ShmCompletionWaitCount--;
}
break;
}
}
update_vo_size(vo);
if (vo->opts->WinID >= 0 && (x11->pending_vo_events & VO_EVENT_RESIZE)) {
int x = x11->win_x, y = x11->win_y;
unsigned int w = x11->win_width, h = x11->win_height;
XMoveResizeWindow(x11->display, x11->window, x, y, w, h);
}
int ret = x11->pending_vo_events;
x11->pending_vo_events = 0;
return ret;
}
static void vo_x11_sizehint(struct vo *vo, int x, int y, int width, int height,
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
bool override_pos)
{
struct mp_vo_opts *opts = vo->opts;
2008-04-21 00:14:37 +00:00
struct vo_x11_state *x11 = vo->x11;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
bool force_pos = opts->geometry.xy_valid || // explicitly forced by user
opts->force_window_position || // resize -> reset position
opts->screen_id >= 0 || // force onto screen area
opts->WinID >= 0 || // force to fill parent
override_pos; // for fullscreen and such
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
2008-04-21 01:19:43 +00:00
x11->vo_hint.flags = 0;
if (opts->keepaspect) {
2008-04-21 01:19:43 +00:00
x11->vo_hint.flags |= PAspect;
x11->vo_hint.min_aspect.x = width;
x11->vo_hint.min_aspect.y = height;
x11->vo_hint.max_aspect.x = width;
x11->vo_hint.max_aspect.y = height;
}
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
x11->vo_hint.flags |= PSize | (force_pos ? PPosition : 0);
2008-04-21 01:19:43 +00:00
x11->vo_hint.x = x;
x11->vo_hint.y = y;
x11->vo_hint.width = width;
x11->vo_hint.height = height;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
x11->vo_hint.max_width = 0;
x11->vo_hint.max_height = 0;
libvo, libao: remove useless video and audio output drivers Some of these have only limited use, and some of these have no use at all. Remove them. They make maintainance harder and nobody needs them. It's possible that many of the removed drivers were very useful a dozen of years ago, but now it's 2012. Note that some of these could be added back, in case they were more useful than I thought. But right now, they are just a burden. Reason for removal for each module: vo_3dfx, vo_dfbmga, vo_dxr3, vo_ivtv, vo_mga, vo_s3fb, vo_tdfxfb, vo_xmga, vo_tdfx_vid: All of these are for very specific and outdated hardware. Some of them require non-standard kernel drivers or do direct HW access. vo_dga: the most crappy and ancient way to get fast output on X. vo_aa: there's vo_caca for the same purpose. vo_ggi: this never lived, and is entirely useless. vo_mpegpes: for DVB cards, I can't test this and it's crappy. vo_fbdev, vo_fbdev2: there's vo_directfb2 vo_bl: what is this even? But it's neither important, nor alive. vo_svga, vo_vesa: you want to use this? You can't be serious. vo_wii: I can't test this, and who the hell uses this? vo_xvr100: some Sun thing. vo_xover: only useful in connection with xvr100. ao_nas: still alive, but I doubt it has any meaning today. ao_sun: Sun. ao_win32: use ao_dsound or ao_portaudio instead. ao_ivtv: removed along vo_ivtv. Also get rid of anything SDL related. SDL 1.x is total crap for video output, and will be replaced with SDL 2.x soon (perhaps), so if you want to use SDL, write output drivers for SDL 2.x. Additionally, I accidentally damaged Sun support, which made me completely remove Sun/Solaris support. Nobody cares about this anyway. Some left overs from previous commits removing modules were cleaned up.
2012-07-28 18:20:17 +00:00
// Set minimum height/width to 4 to avoid off-by-one errors.
2008-04-21 01:19:43 +00:00
x11->vo_hint.flags |= PMinSize;
x11->vo_hint.min_width = x11->vo_hint.min_height = 4;
// Set the base size. A window manager might display the window
// size to the user relative to this.
// Setting these to width/height might be nice, but e.g. fluxbox can't handle it.
2008-12-06 00:16:51 +00:00
x11->vo_hint.flags |= PBaseSize;
x11->vo_hint.base_width = 0 /*width*/;
x11->vo_hint.base_height = 0 /*height*/;
2008-04-21 01:19:43 +00:00
x11->vo_hint.flags |= PWinGravity;
x11->vo_hint.win_gravity = StaticGravity;
XSetWMNormalHints(x11->display, x11->window, &x11->vo_hint);
}
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
static void vo_x11_move_resize(struct vo *vo, bool move, bool resize,
int x, int y, int w, int h)
{
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
XWindowChanges req = {.x = x, .y = y, .width = w, .height = h};
unsigned mask = (move ? CWX | CWY : 0) | (resize ? CWWidth | CWHeight : 0);
if (mask)
XConfigureWindow(vo->x11->display, vo->x11->window, mask, &req);
vo_x11_sizehint(vo, x, y, w, h, false);
}
2008-04-20 23:50:40 +00:00
static int vo_x11_get_gnome_layer(struct vo_x11_state *x11, Window win)
{
Atom type;
int format;
unsigned long nitems;
unsigned long bytesafter;
unsigned short *args = NULL;
2008-04-20 23:50:40 +00:00
if (XGetWindowProperty(x11->display, win, x11->XA_WIN_LAYER, 0, 16384,
False, AnyPropertyType, &type, &format, &nitems,
&bytesafter,
(unsigned char **) &args) == Success
&& nitems > 0 && args)
{
MP_VERBOSE(x11, "original window layer is %d.\n", *args);
return *args;
}
return WIN_LAYER_NORMAL;
}
// set a X text property that expects a UTF8_STRING type
static void vo_x11_set_property_utf8(struct vo *vo, Atom name, const char *t)
{
struct vo_x11_state *x11 = vo->x11;
XChangeProperty(x11->display, x11->window, name, x11->XAUTF8_STRING, 8,
PropModeReplace, t, strlen(t));
}
// set a X text property that expects a STRING or COMPOUND_TEXT type
static void vo_x11_set_property_string(struct vo *vo, Atom name, const char *t)
{
struct vo_x11_state *x11 = vo->x11;
XTextProperty prop = {0};
if (Xutf8TextListToTextProperty(x11->display, (char **)&t, 1,
XStdICCTextStyle, &prop) == Success)
{
XSetTextProperty(x11->display, x11->window, &prop, name);
} else {
// Strictly speaking this violates the ICCCM, but there's no way we
// can do this correctly.
vo_x11_set_property_utf8(vo, name, t);
}
if (prop.value)
XFree(prop.value);
}
static void vo_x11_update_window_title(struct vo *vo)
{
struct vo_x11_state *x11 = vo->x11;
if (!x11->window)
return;
const char *title = vo_get_window_title(vo);
vo_x11_set_property_string(vo, XA_WM_NAME, title);
vo_x11_set_property_string(vo, XA_WM_ICON_NAME, title);
vo_x11_set_property_utf8(vo, x11->XA_NET_WM_NAME, title);
vo_x11_set_property_utf8(vo, x11->XA_NET_WM_ICON_NAME, title);
}
#if HAVE_ZLIB
static bstr decompress_gz(bstr in)
{
bstr res = {0};
z_stream zstream;
uint8_t *dest;
size_t size = in.len;
int result;
zstream.zalloc = (alloc_func) 0;
zstream.zfree = (free_func) 0;
zstream.opaque = (voidpf) 0;
// 32 for gzip header, 15 for max. window bits
if (inflateInit2(&zstream, 32 + 15) != Z_OK)
goto error;
zstream.next_in = (Bytef *) in.start;
zstream.avail_in = size;
dest = NULL;
zstream.avail_out = size;
do {
size += 4000;
dest = talloc_realloc_size(NULL, dest, size);
zstream.next_out = (Bytef *) (dest + zstream.total_out);
result = inflate(&zstream, Z_NO_FLUSH);
if (result != Z_OK && result != Z_STREAM_END) {
talloc_free(dest);
dest = NULL;
inflateEnd(&zstream);
goto error;
}
zstream.avail_out += 4000;
} while (zstream.avail_out == 4000 && zstream.avail_in != 0
&& result != Z_STREAM_END);
size = zstream.total_out;
inflateEnd(&zstream);
res.start = dest;
res.len = size;
error:
return res;
}
#else
static bstr decompress_gz(bstr in)
{
return (bstr){0};
}
#endif
#define MAX_ICONS 10
static void vo_x11_set_wm_icon(struct vo_x11_state *x11)
{
int num_icons = 0;
void *icon_data[MAX_ICONS];
int icon_w[MAX_ICONS], icon_h[MAX_ICONS];
bstr uncompressed = decompress_gz((bstr){(char *)x11_icon, sizeof(x11_icon)});
bstr data = uncompressed;
while (data.len && num_icons < MAX_ICONS) {
bstr line = bstr_getline(data, &data);
if (bstr_eatstart0(&line, "icon: ")) {
int w, h;
if (bstr_sscanf(line, "%d %d", &w, &h) == 2) {
int size = w * h * 4;
icon_w[num_icons] = w;
icon_h[num_icons] = h;
icon_data[num_icons] = data.start;
num_icons++;
data = bstr_cut(data, size);
}
}
}
int icon_size = 0;
for (int n = 0; n < num_icons; n++)
icon_size += 2 + icon_w[n] * icon_h[n];
long *icon = talloc_array(NULL, long, icon_size);
long *cur = icon;
for (int n = 0; n < num_icons; n++) {
*cur++ = icon_w[n];
*cur++ = icon_h[n];
uint8_t *s = icon_data[n];
for (int i = 0; i < icon_h[n] * icon_w[n]; i++, s += 4)
*cur++ = s[0] | (s[1] << 8) | (s[2] << 16) | ((unsigned)s[3] << 24);
}
XChangeProperty(x11->display, x11->window, x11->XA_NET_WM_ICON,
XA_CARDINAL, 32, PropModeReplace,
(unsigned char *)icon, icon_size);
talloc_free(icon);
talloc_free(uncompressed.start);
}
static void find_default_visual(struct vo_x11_state *x11, XVisualInfo *vis)
{
Display *display = x11->display;
XWindowAttributes attribs;
XGetWindowAttributes(display, DefaultRootWindow(display), &attribs);
XMatchVisualInfo(display, x11->screen, attribs.depth, TrueColor, vis);
}
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
static void vo_x11_create_window(struct vo *vo, XVisualInfo *vis, int x, int y,
unsigned int w, unsigned int h)
{
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
struct vo_x11_state *x11 = vo->x11;
assert(x11->window == None);
assert(!x11->xic);
XVisualInfo vinfo_storage;
if (!vis) {
vis = &vinfo_storage;
find_default_visual(x11, vis);
}
if (x11->colormap == None) {
x11->colormap = XCreateColormap(x11->display, x11->rootwin,
vis->visual, AllocNone);
}
2013-04-15 19:18:46 +00:00
unsigned long xswamask = CWBorderPixel | CWColormap;
XSetWindowAttributes xswa = {
.border_pixel = 0,
.colormap = x11->colormap,
};
Window parent = vo->opts->WinID >= 0 ? vo->opts->WinID : x11->rootwin;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
x11->window =
XCreateWindow(x11->display, parent, x, y, w, h, 0, vis->depth,
CopyFromParent, vis->visual, xswamask, &xswa);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
XSetWMProtocols(x11->display, x11->window, &x11->XAWM_DELETE_WINDOW, 1);
x11->f_gc = XCreateGC(x11->display, x11->window, 0, 0);
x11->vo_gc = XCreateGC(x11->display, x11->window, 0, NULL);
2008-04-21 01:19:43 +00:00
XSetForeground(x11->display, x11->f_gc, 0);
if (x11->mouse_cursor_hidden) {
x11->mouse_cursor_hidden = false;
vo_set_cursor_hidden(vo, true);
}
if (x11->xim) {
x11->xic = XCreateIC(x11->xim,
XNInputStyle, XIMPreeditNone | XIMStatusNone,
XNClientWindow, x11->window,
XNFocusWindow, x11->window,
NULL);
}
vo_x11_set_wm_icon(x11);
vo_x11_update_window_title(vo);
vo_x11_dnd_init_window(vo);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
}
static void vo_x11_map_window(struct vo *vo, int x, int y, int w, int h)
{
struct vo_x11_state *x11 = vo->x11;
vo_x11_move_resize(vo, true, true, x, y, w, h);
if (!vo->opts->border)
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_decoration(vo, 0);
if (vo->opts->fullscreen && (x11->wm_type & vo_wm_FULLSCREEN)) {
Atom state = x11->XA_NET_WM_STATE_FULLSCREEN;
XChangeProperty(x11->display, x11->window, x11->XA_NET_WM_STATE, XA_ATOM,
32, PropModeAppend, (unsigned char *)&state, 1);
x11->fs = 1;
// The "saved" positions are bogus, so reset them when leaving FS again.
x11->size_changed_during_fs = true;
x11->pos_changed_during_fs = true;
}
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
// map window
vo_x11_selectinput_witherr(vo, x11->display, x11->window,
StructureNotifyMask | ExposureMask |
KeyPressMask | KeyReleaseMask |
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
ButtonPressMask | ButtonReleaseMask |
PointerMotionMask | LeaveWindowMask);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
XMapWindow(x11->display, x11->window);
}
static void vo_x11_highlevel_resize(struct vo *vo, int x, int y, int w, int h)
{
struct mp_vo_opts *opts = vo->opts;
struct vo_x11_state *x11 = vo->x11;
bool reset_pos = opts->force_window_position;
if (reset_pos) {
x11->nofs_x = x;
x11->nofs_y = y;
}
x11->nofs_width = w;
x11->nofs_height = h;
if (opts->fullscreen) {
x11->size_changed_during_fs = true;
x11->pos_changed_during_fs = reset_pos;
vo_x11_sizehint(vo, x, y, w, h, false);
} else {
vo_x11_move_resize(vo, reset_pos, true, x, y, w, h);
}
vo_x11_update_geometry(vo);
update_vo_size(vo);
}
static void wait_until_mapped(struct vo *vo)
{
struct vo_x11_state *x11 = vo->x11;
while (x11->window_hidden) {
XEvent unused;
XPeekEvent(x11->display, &unused);
vo_x11_check_events(vo);
}
}
/* Create and setup a window suitable for display
* vis: Visual to use for creating the window (NULL for default)
* x, y: position of window (might be ignored)
* width, height: size of window
* flags: flags for window creation (VOFLAG_*)
* classname: name to use for the X11 classhint
*
* If the window already exists, it just moves and resizes it.
*/
void vo_x11_config_vo_window(struct vo *vo, XVisualInfo *vis, int flags,
const char *classname)
{
struct mp_vo_opts *opts = vo->opts;
struct vo_x11_state *x11 = vo->x11;
vo_x11_update_screeninfo(vo);
struct vo_win_geometry geo;
vo_calc_window_geometry(vo, &x11->screenrc, &geo);
vo_apply_window_geometry(vo, &geo);
int x = geo.win.x0;
int y = geo.win.y0;
int width = geo.win.x1 - geo.win.x0;
int height = geo.win.y1 - geo.win.y0;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
if (opts->WinID >= 0) {
if (opts->WinID == 0)
x11->window = x11->rootwin;
XSelectInput(x11->display, opts->WinID, StructureNotifyMask);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_update_geometry(vo);
x = x11->win_x; y = x11->win_y;
width = x11->win_width; height = x11->win_height;
}
if (x11->window == None) {
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_create_window(vo, vis, x, y, width, height);
vo_x11_classhint(vo, x11->window, classname);
x11->window_hidden = true;
}
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
if (flags & VOFLAG_HIDDEN)
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
return;
bool reset_size = x11->old_dwidth != width || x11->old_dheight != height;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
x11->old_dwidth = width;
x11->old_dheight = height;
if (x11->window_hidden) {
x11->nofs_x = x;
x11->nofs_y = y;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
x11->nofs_width = width;
x11->nofs_height = height;
vo_x11_map_window(vo, x, y, width, height);
} else if (reset_size) {
vo_x11_highlevel_resize(vo, x, y, width, height);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
}
if (opts->ontop)
vo_x11_setlayer(vo, x11->window, opts->ontop);
vo_x11_fullscreen(vo);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
XSync(x11->display, False);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_update_geometry(vo);
if (x11->window_hidden) {
// The real size is only known when the window is mapped, which
// unfortunately happens asynchronous to VO initialization. At least
// vdpau needs a _some_ window size, though.
x11->win_width = width;
x11->win_height = height;
}
wait_until_mapped(vo);
update_vo_size(vo);
x11->pending_vo_events &= ~VO_EVENT_RESIZE; // implicitly done by the VO
}
static void fill_rect(struct vo *vo, GC gc, int x0, int y0, int x1, int y1)
{
2008-04-21 01:19:43 +00:00
struct vo_x11_state *x11 = vo->x11;
x0 = MPMAX(x0, 0);
y0 = MPMAX(y0, 0);
x1 = MPMIN(x1, x11->win_width);
y1 = MPMIN(y1, x11->win_height);
if (x11->window && x1 > x0 && y1 > y0)
XFillRectangle(x11->display, x11->window, gc, x0, y0, x1 - x0, y1 - y0);
}
// Clear everything outside of rc with the background color
void vo_x11_clear_background(struct vo *vo, const struct mp_rect *rc)
{
struct vo_x11_state *x11 = vo->x11;
GC gc = x11->f_gc;
int w = x11->win_width;
int h = x11->win_height;
fill_rect(vo, gc, 0, 0, w, rc->y0); // top
fill_rect(vo, gc, 0, rc->y1, w, h); // bottom
fill_rect(vo, gc, 0, rc->y0, rc->x0, rc->y1); // left
fill_rect(vo, gc, rc->x1, rc->y0, w, rc->y1); // right
XFlush(x11->display);
}
void vo_x11_clearwindow(struct vo *vo, Window vo_window)
{
2008-04-21 01:19:43 +00:00
struct vo_x11_state *x11 = vo->x11;
if (x11->f_gc == None)
return;
XFillRectangle(x11->display, vo_window, x11->f_gc, 0, 0, INT_MAX, INT_MAX);
2008-04-21 01:19:43 +00:00
XFlush(x11->display);
}
static void vo_x11_setlayer(struct vo *vo, Window vo_window, int layer)
{
2008-04-20 23:50:40 +00:00
struct vo_x11_state *x11 = vo->x11;
if (vo->opts->WinID >= 0)
return;
if (x11->wm_type & vo_wm_LAYER) {
XClientMessageEvent xev;
if (!x11->orig_layer)
x11->orig_layer = vo_x11_get_gnome_layer(x11, vo_window);
memset(&xev, 0, sizeof(xev));
xev.type = ClientMessage;
2008-04-20 23:50:40 +00:00
xev.display = x11->display;
xev.window = vo_window;
2008-04-20 23:50:40 +00:00
xev.message_type = x11->XA_WIN_LAYER;
xev.format = 32;
// if not fullscreen, stay on default layer
xev.data.l[0] = layer ? WIN_LAYER_ABOVE_DOCK : x11->orig_layer;
xev.data.l[1] = CurrentTime;
MP_VERBOSE(x11, "Layered style stay on top (layer %ld).\n",
xev.data.l[0]);
XSendEvent(x11->display, x11->rootwin, False, SubstructureNotifyMask,
(XEvent *) &xev);
} else if (x11->wm_type & vo_wm_NETWM) {
XClientMessageEvent xev;
char *state;
memset(&xev, 0, sizeof(xev));
xev.type = ClientMessage;
2008-04-20 23:50:40 +00:00
xev.message_type = x11->XA_NET_WM_STATE;
xev.display = x11->display;
xev.window = vo_window;
xev.format = 32;
xev.data.l[0] = layer;
if (x11->wm_type & vo_wm_STAYS_ON_TOP) {
2008-04-20 23:50:40 +00:00
xev.data.l[1] = x11->XA_NET_WM_STATE_STAYS_ON_TOP;
} else if (x11->wm_type & vo_wm_ABOVE) {
2008-04-20 23:50:40 +00:00
xev.data.l[1] = x11->XA_NET_WM_STATE_ABOVE;
} else if (x11->wm_type & vo_wm_FULLSCREEN) {
2008-04-20 23:50:40 +00:00
xev.data.l[1] = x11->XA_NET_WM_STATE_FULLSCREEN;
} else if (x11->wm_type & vo_wm_BELOW) {
// This is not fallback. We can safely assume that the situation
// where only NETWM_STATE_BELOW is supported doesn't exist.
2008-04-20 23:50:40 +00:00
xev.data.l[1] = x11->XA_NET_WM_STATE_BELOW;
}
XSendEvent(x11->display, x11->rootwin, False, SubstructureRedirectMask,
(XEvent *) &xev);
2008-04-20 23:50:40 +00:00
state = XGetAtomName(x11->display, xev.data.l[1]);
MP_VERBOSE(x11, "NET style stay on top (layer %d). Using state %s.\n",
layer, state);
XFree(state);
}
}
// update x11->win_x, x11->win_y, x11->win_width and x11->win_height with current values of vo->x11->window
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
static void vo_x11_update_geometry(struct vo *vo)
{
struct vo_x11_state *x11 = vo->x11;
unsigned w, h, dummy_uint;
int dummy_int;
Window dummy_win;
Window win = vo->opts->WinID > 0 ? vo->opts->WinID : x11->window;
x11: fix race condition when setting aspect when leaving fullscreen vo_opengl creates a hidden X11 window to probe the OpenGL context. It must do that before creating a visible window, because VO creation and VO config are separate phases. There's a race condition involving the hidden window: when starting with --fs, and then leaving fullscreen, the unfullscreened window is sometimes set to the aspect ratio of the hidden window. I'm not sure why the window size itself uses the correct size (but corrupted by the wrong aspect), but that's perhaps because the window manager is free to ignore the size hint while honoring the aspect, or something equally messed up. It turns out this happens because x11_common.c thinks the size of the hidden window is the size of the unfullscreened window. This in turn happens because vo_x11_update_geometry() reads the size of the hidden window when called in vo_x11_fullscreen() (called from vo_x11_config_vo_window()) when mapping the fullscreen window. At that point, the window could be mapped, but not necessarily. If it's not mapped, it will get the size of the unfullscreened window... I think. One could fix this by actively waiting until the window is mapped. Try to pick a less hacky approach instead, and never read the window size until MapNotify is received. vo_x11_create_window() needs a hack, because we'd possibly set the VO's size to 0, resulting e.g. in vdpau to fail initialization. (It'll print error messages until a proper resize is received.)
2014-02-02 00:35:46 +00:00
if (x11->window_hidden)
return;
XGetGeometry(x11->display, win, &dummy_win, &dummy_int, &dummy_int,
&w, &h, &dummy_int, &dummy_uint);
if (w <= INT_MAX && h <= INT_MAX) {
x11->win_width = w;
x11->win_height = h;
}
if (vo->opts->WinID >= 0) {
x11->win_x = 0;
x11->win_y = 0;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
} else {
XTranslateCoordinates(x11->display, x11->window, x11->rootwin, 0, 0,
&x11->win_x, &x11->win_y, &dummy_win);
}
}
2013-05-16 12:02:28 +00:00
static void vo_x11_fullscreen(struct vo *vo)
{
struct mp_vo_opts *opts = vo->opts;
struct vo_x11_state *x11 = vo->x11;
int x, y, w, h;
if (opts->fullscreen == x11->fs)
return;
if (opts->WinID >= 0) {
x11->fs = opts->fullscreen;
return;
}
if (!opts->fullscreen) {
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
// fs->win
opts->fullscreen = x11->fs = 0;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
x = x11->nofs_x;
y = x11->nofs_y;
w = x11->nofs_width;
h = x11->nofs_height;
2008-04-21 00:14:37 +00:00
vo_x11_ewmh_fullscreen(x11, _NET_WM_STATE_REMOVE); // removes fullscreen state if wm supports EWMH
if ((x11->wm_type & vo_wm_FULLSCREEN) && opts->fsscreen_id != -1) {
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
x11->size_changed_during_fs = true;
x11->pos_changed_during_fs = true;
}
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
if (x11->wm_type & vo_wm_FULLSCREEN) {
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_move_resize(vo, x11->pos_changed_during_fs,
x11->size_changed_during_fs, x, y, w, h);
}
} else {
// win->fs
opts->fullscreen = x11->fs = 1;
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_update_geometry(vo);
x11->nofs_x = x11->win_x;
x11->nofs_y = x11->win_y;
x11->nofs_width = x11->win_width;
x11->nofs_height = x11->win_height;
vo_x11_update_screeninfo(vo);
x = x11->screenrc.x0;
y = x11->screenrc.y0;
w = x11->screenrc.x1 - x;
h = x11->screenrc.y1 - y;
if ((x11->wm_type & vo_wm_FULLSCREEN) && opts->fsscreen_id != -1) {
// The EWMH fullscreen hint always works on the current screen, so
// change the current screen forcibly.
// This was observed to work under IceWM, but not Unity/Compiz and
// awesome (but --screen etc. doesn't really work on these either).
XMoveResizeWindow(x11->display, x11->window, x, y, w, h);
}
2013-10-30 20:58:55 +00:00
// sends fullscreen state to be added if wm supports EWMH
vo_x11_ewmh_fullscreen(x11, _NET_WM_STATE_ADD);
}
2013-10-30 20:58:55 +00:00
XGetWMNormalHints(x11->display, x11->window, &x11->vo_hint, &(long){0});
if (!(x11->vo_hint.flags & PWinGravity))
x11->old_gravity = NorthWestGravity;
else
x11->old_gravity = x11->vo_hint.win_gravity;
if (!(x11->wm_type & vo_wm_FULLSCREEN)) { // not needed with EWMH fs
vo_x11_decoration(vo, opts->border && !x11->fs);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
vo_x11_sizehint(vo, x, y, w, h, true);
vo_x11_setlayer(vo, x11->window, x11->fs);
2008-04-21 00:14:37 +00:00
XMoveResizeWindow(x11->display, x11->window, x, y, w, h);
}
/* some WMs lose ontop after fullscreen */
if ((!(x11->fs)) & opts->ontop)
vo_x11_setlayer(vo, x11->window, opts->ontop);
2008-04-21 00:14:37 +00:00
XMapRaised(x11->display, x11->window);
if (!(x11->wm_type & vo_wm_FULLSCREEN)) // some WMs change window pos on map
2008-04-21 00:14:37 +00:00
XMoveResizeWindow(x11->display, x11->window, x, y, w, h);
XRaiseWindow(x11->display, x11->window);
XFlush(x11->display);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
x11->size_changed_during_fs = false;
x11->pos_changed_during_fs = false;
}
2013-05-16 12:02:28 +00:00
static void vo_x11_ontop(struct vo *vo)
{
struct mp_vo_opts *opts = vo->opts;
opts->ontop = !opts->ontop;
vo_x11_setlayer(vo, vo->x11->window, opts->ontop);
}
2013-05-16 12:02:28 +00:00
static void vo_x11_border(struct vo *vo)
{
vo->opts->border = !vo->opts->border;
vo_x11_decoration(vo, vo->opts->border);
}
int vo_x11_control(struct vo *vo, int *events, int request, void *arg)
{
struct vo_x11_state *x11 = vo->x11;
switch (request) {
case VOCTRL_CHECK_EVENTS:
*events |= vo_x11_check_events(vo);
return VO_TRUE;
case VOCTRL_FULLSCREEN:
vo_x11_fullscreen(vo);
*events |= VO_EVENT_RESIZE;
return VO_TRUE;
case VOCTRL_ONTOP:
vo_x11_ontop(vo);
return VO_TRUE;
case VOCTRL_BORDER:
vo_x11_border(vo);
*events |= VO_EVENT_RESIZE;
return VO_TRUE;
return VO_TRUE;
case VOCTRL_GET_WINDOW_SIZE: {
int *s = arg;
if (!x11->window)
return VO_FALSE;
s[0] = x11->fs ? x11->nofs_width : x11->win_width;
s[1] = x11->fs ? x11->nofs_height : x11->win_height;
return VO_TRUE;
}
case VOCTRL_SET_WINDOW_SIZE: {
int *s = arg;
vo_x11_highlevel_resize(vo, x11->win_x, x11->win_y, s[0], s[1]);
return VO_TRUE;
}
case VOCTRL_SET_CURSOR_VISIBILITY:
vo_set_cursor_hidden(vo, !(*(bool *)arg));
return VO_TRUE;
case VOCTRL_KILL_SCREENSAVER:
saver_off(x11);
return VO_TRUE;
case VOCTRL_RESTORE_SCREENSAVER:
saver_on(x11);
return VO_TRUE;
case VOCTRL_UPDATE_WINDOW_TITLE:
vo_x11_update_window_title(vo);
return VO_TRUE;
}
return VO_NOTIMPL;
}
static void xscreensaver_heartbeat(struct vo_x11_state *x11)
{
double time = mp_time_sec();
if (x11->display && x11->screensaver_off &&
(time - x11->screensaver_time_last) >= 10)
{
x11->screensaver_time_last = time;
XResetScreenSaver(x11->display);
}
}
static int xss_suspend(Display *mDisplay, Bool suspend)
{
#if !HAVE_XSS
return 0;
#else
int event, error, major, minor;
if (XScreenSaverQueryExtension(mDisplay, &event, &error) != True ||
XScreenSaverQueryVersion(mDisplay, &major, &minor) != True)
return 0;
if (major < 1 || (major == 1 && minor < 1))
return 0;
XScreenSaverSuspend(mDisplay, suspend);
return 1;
#endif
}
static void saver_on(struct vo_x11_state *x11)
{
Display *mDisplay = x11->display;
if (!x11->screensaver_off)
return;
x11->screensaver_off = 0;
if (xss_suspend(mDisplay, False))
return;
#if HAVE_XEXT
if (x11->dpms_disabled) {
int nothing;
if (DPMSQueryExtension(mDisplay, &nothing, &nothing)) {
if (!DPMSEnable(mDisplay)) { // restoring power saving settings
MP_WARN(x11, "DPMS not available?\n");
} else {
// DPMS does not seem to be enabled unless we call DPMSInfo
BOOL onoff;
CARD16 state;
DPMSForceLevel(mDisplay, DPMSModeOn);
DPMSInfo(mDisplay, &state, &onoff);
if (onoff) {
MP_VERBOSE(x11, "Successfully enabled DPMS\n");
} else {
MP_WARN(x11, "Could not enable DPMS\n");
}
}
}
x11->dpms_disabled = 0;
}
#endif
}
static void saver_off(struct vo_x11_state *x11)
{
Display *mDisplay = x11->display;
int nothing;
if (x11->screensaver_off)
return;
x11->screensaver_off = 1;
if (xss_suspend(mDisplay, True))
return;
#if HAVE_XEXT
if (DPMSQueryExtension(mDisplay, &nothing, &nothing)) {
BOOL onoff;
CARD16 state;
DPMSInfo(mDisplay, &state, &onoff);
if (onoff) {
Status stat;
MP_VERBOSE(x11, "Disabling DPMS\n");
x11->dpms_disabled = 1;
stat = DPMSDisable(mDisplay); // monitor powersave off
MP_VERBOSE(x11, "DPMSDisable stat: %d\n", stat);
}
}
#endif
}
static void vo_x11_selectinput_witherr(struct vo *vo,
Display *display,
Window w,
long event_mask)
{
if (!vo->opts->enable_mouse_movements)
event_mask &= ~(PointerMotionMask | ButtonPressMask | ButtonReleaseMask);
x11_common: fix window mapping, refactor window creation/resize handling vo_opengl creates an invisible window with VOFLAG_HIDDEN in order to test whether the OpenGL context is useable. The visible window is created at a later point. This has been broken forever (in vo_gl, now called vo_opengl_old, it could be avoided by disabling auto- detection explicitly using the "yuv" sub-option). Avoiding VOFLAG_HIDDEN only mitigates the issue, and a related bug can still happen with some window managers (see below). As a hack, code was added to vo_gl to destroy the (hidden) window so that all state was destroyed. Later, more hacks were added to deal with some caveats that came with recreating the window, such as probing for the context up to 4 times. Attempt to fix the problem properly. There were two problems: first, the window was not resized to video size before mapping. This was the main cause for the placement issue, e.g. mapping the window as 320x200, and then resizing it. Second, mpv tried to force the window position with XSetWMNormalHints and PPosition with values it just read with XGetGeometry. This messes up the window manager's default placement. It seems to be a race condition, and behavior is different across various WMs too: with IceWM, the window manager's placement is usually preferred, and with Fluxbox, mpv's position is preferred. mpv's default position is centering the window on the screen, which is much nicer for video in general than typical WM default placement, so it's possible that this bug was perceived as a feature. (Users who want this have to use --geometry="50%:50%", doing this by default is probably not safe with all WMs.) Since the old code was hard to follow and full of issues, it's easier to redo it. Move general window creation stuff out of the vo_x11_config_vo_window function, and move the resize logic into it. This has been tested on IcwWM, Fluxbox, awesome, Unity/Compiz.
2013-03-02 16:08:37 +00:00
XSelectInput(display, w, NoEventMask);
// NOTE: this can raise BadAccess, which should be ignored by the X error
// handler; also see below
XSelectInput(display, w, event_mask);
// Test whether setting the event mask failed (with a BadAccess X error,
// although we don't know whether this really happened).
// This is needed for obscure situations like using --rootwin with a window
// manager active.
XWindowAttributes a;
if (XGetWindowAttributes(display, w, &a)) {
long bad = ButtonPressMask | ButtonReleaseMask | PointerMotionMask;
if ((event_mask & bad) && (a.all_event_masks & bad) &&
((a.your_event_mask & bad) != (event_mask & bad)))
{
MP_ERR(vo->x11, "X11 error: error during XSelectInput "
"call, trying without mouse events\n");
XSelectInput(display, w, event_mask & ~bad);
}
}
}
#if HAVE_XF86VM
double vo_x11_vm_get_fps(struct vo *vo)
Implement vsync-aware frame timing for VDPAU Main things added are custom frame dropping for VDPAU to work around the display FPS limit, frame timing adjustment to avoid jitter when video frame times keep falling near vsyncs, and use of VDPAU's timing feature to keep one future frame queued in advance. NVIDIA's VDPAU implementation refuses to change the displayed frame more than once per vsync. This set a limit on how much video could be sped up, and caused problems for nearly all videos on low-FPS video projectors (playing 24 FPS video on a 24 FPS projector would not work reliably as MPlayer may need to slightly speed up the video for AV sync). This commit adds a framedrop mechanism that drops some frames so that no more than one is sent for display per vsync. The code tries to select the dropped frames smartly, selecting the best one to show for each vsync. Because of the timing features needed the drop functionality currently does not work if the correct-pts option is disabled. The code also adjusts frame timing slightly to avoid jitter. If you for example play 24 FPS video content on a 72 FPS display then normally a frame would be shown for 3 vsyncs, but if the frame times happen to fall near vsyncs and change between just before and just after then there could be frames alternating between 2 and 4 vsyncs. The code changes frame timing by up to one quarter vsync interval to avoid this. The above functionality depends on having reliable vsync timing information available. The display refresh rate is not directly provided by the VDPAU API. The current code uses information from the XF86VidMode extension if available; I'm not sure how common cases where that is inaccurate are. The refresh rate can be specified manually if necessary. After the changes in this commit MPlayer now always tries to keep one frame queued for future display using VDPAU's internal timing mechanism (however no more than 50 ms to the future). This should make video playback somewhat more robust against timing inaccuracies caused by system load.
2009-11-15 02:39:22 +00:00
{
struct vo_x11_state *x11 = vo->x11;
int clock;
XF86VidModeModeLine modeline;
if (!XF86VidModeGetModeLine(x11->display, x11->screen, &clock, &modeline))
return 0;
if (modeline.privsize)
XFree(modeline.private);
Implement vsync-aware frame timing for VDPAU Main things added are custom frame dropping for VDPAU to work around the display FPS limit, frame timing adjustment to avoid jitter when video frame times keep falling near vsyncs, and use of VDPAU's timing feature to keep one future frame queued in advance. NVIDIA's VDPAU implementation refuses to change the displayed frame more than once per vsync. This set a limit on how much video could be sped up, and caused problems for nearly all videos on low-FPS video projectors (playing 24 FPS video on a 24 FPS projector would not work reliably as MPlayer may need to slightly speed up the video for AV sync). This commit adds a framedrop mechanism that drops some frames so that no more than one is sent for display per vsync. The code tries to select the dropped frames smartly, selecting the best one to show for each vsync. Because of the timing features needed the drop functionality currently does not work if the correct-pts option is disabled. The code also adjusts frame timing slightly to avoid jitter. If you for example play 24 FPS video content on a 72 FPS display then normally a frame would be shown for 3 vsyncs, but if the frame times happen to fall near vsyncs and change between just before and just after then there could be frames alternating between 2 and 4 vsyncs. The code changes frame timing by up to one quarter vsync interval to avoid this. The above functionality depends on having reliable vsync timing information available. The display refresh rate is not directly provided by the VDPAU API. The current code uses information from the XF86VidMode extension if available; I'm not sure how common cases where that is inaccurate are. The refresh rate can be specified manually if necessary. After the changes in this commit MPlayer now always tries to keep one frame queued for future display using VDPAU's internal timing mechanism (however no more than 50 ms to the future). This should make video playback somewhat more robust against timing inaccuracies caused by system load.
2009-11-15 02:39:22 +00:00
return 1e3 * clock / modeline.htotal / modeline.vtotal;
}
#else /* HAVE_XF86VM */
double vo_x11_vm_get_fps(struct vo *vo)
{
return 0;
}
#endif
bool vo_x11_screen_is_composited(struct vo *vo)
{
struct vo_x11_state *x11 = vo->x11;
return XGetSelectionOwner(x11->display, x11->XA_NET_WM_CM) != None;
}