FaSTTUBe/GST-Tensordecoder-ov_ep
46
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
GST-Tensordecoder-ov_ep/subprojects/gstreamer/gst/gstutils.c
Sebastian Dröge 8b0ae3ba87 clock: Fix calculation for number of bits needed to store a 64 bit value 
It was using log2(n) but what actually is needed is log2(n) + 1. Also add a
fast-path that uses __builtin_clzll() on gcc/clang.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/7406>
2024-08-29 10:55:35 +00:00

4857 lines
140 KiB
C
Raw Blame History

/* GStreamer
* Copyright (C) 1999,2000 Erik Walthinsen <omega@cse.ogi.edu>
* 2000 Wim Taymans <wtay@chello.be>
* 2002 Thomas Vander Stichele <thomas@apestaart.org>
* 2004 Wim Taymans <wim@fluendo.com>
* 2015 Jan Schmidt <jan@centricular.com>
*
* gstutils.c: Utility functions
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Library General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library 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
* Library General Public License for more details.
*
* You should have received a copy of the GNU Library General Public
* License along with this library; if not, write to the
* Free Software Foundation, Inc., 51 Franklin St, Fifth Floor,
* Boston, MA 02110-1301, USA.
*/
/**
* SECTION:gstutils
* @title: GstUtils
* @short_description: Various utility functions
*
*/
/* FIXME 2.0: suppress warnings for deprecated API such as GValueArray
* with newer GLib versions (>= 2.31.0) */
#define GLIB_DISABLE_DEPRECATION_WARNINGS
#include "gst_private.h"
#include <stdio.h>
#include <string.h>
#include "gstghostpad.h"
#include "gstutils.h"
#include "gsterror.h"
#include "gstinfo.h"
#include "gstparse.h"
#include "gstvalue.h"
#include "gstquark.h"
#include <glib/gi18n-lib.h>
#include "glib-compat-private.h"
#include <math.h>
static void
gst_util_dump_mem_offset (const guchar * mem, guint size, guint offset)
{
guint i, j;
GString *string = g_string_sized_new (50);
GString *chars = g_string_sized_new (18);
i = j = 0;
while (i < size) {
if (g_ascii_isprint (mem[i]))
g_string_append_c (chars, mem[i]);
else
g_string_append_c (chars, '.');
g_string_append_printf (string, "%02x ", mem[i]);
j++;
i++;
if (j == 16 || i == size) {
g_print ("%08x (%p): %-48.48s %-16.16s\n", i - j + offset, mem + i - j,
string->str, chars->str);
g_string_set_size (string, 0);
g_string_set_size (chars, 0);
j = 0;
}
}
g_string_free (string, TRUE);
g_string_free (chars, TRUE);
}
/**
* gst_util_dump_mem:
* @mem: (array length=size): a pointer to the memory to dump
* @size: the size of the memory block to dump
*
* Dumps the memory block into a hex representation. Useful for debugging.
*/
void
gst_util_dump_mem (const guchar * mem, guint size)
{
gst_util_dump_mem_offset (mem, size, 0);
}
/**
* gst_util_dump_buffer:
* @buf: a #GstBuffer whose memory to dump
*
* Dumps the buffer memory into a hex representation. Useful for debugging.
*
* Since: 1.14
*/
void
gst_util_dump_buffer (GstBuffer * buf)
{
GstMapInfo map;
GstMemory *mem;
guint n_memory;
guint i;
guint offset;
n_memory = gst_buffer_n_memory (buf);
if (n_memory == 1) {
if (gst_buffer_map (buf, &map, GST_MAP_READ)) {
gst_util_dump_mem (map.data, map.size);
gst_buffer_unmap (buf, &map);
}
} else if (n_memory > 1) {
/* gst_buffer_map() will merge multiple memory segments into one contiguous
* area so we need to use gst_memory_map() in order not to affect the
* contents of buf */
offset = 0;
for (i = 0; i < n_memory; ++i) {
g_print ("[Memory #%u]\n", i);
mem = gst_buffer_get_memory (buf, i);
if (gst_memory_map (mem, &map, GST_MAP_READ)) {
gst_util_dump_mem_offset (map.data, map.size, offset);
offset += map.size;
gst_memory_unmap (mem, &map);
}
gst_memory_unref (mem);
}
} else {
g_print ("[Empty]\n");
}
}
/**
* gst_util_set_value_from_string:
* @value: (out caller-allocates): the value to set
* @value_str: the string to get the value from
*
* Converts the string to the type of the value and
* sets the value with it.
*
* Note that this function is dangerous as it does not return any indication
* if the conversion worked or not.
*/
void
gst_util_set_value_from_string (GValue * value, const gchar * value_str)
{
gboolean res;
g_return_if_fail (value != NULL);
g_return_if_fail (value_str != NULL);
GST_CAT_DEBUG (GST_CAT_PARAMS, "parsing '%s' to type %s", value_str,
g_type_name (G_VALUE_TYPE (value)));
res = gst_value_deserialize (value, value_str);
if (!res && G_VALUE_TYPE (value) == G_TYPE_BOOLEAN) {
/* backwards compat, all booleans that fail to parse are false */
g_value_set_boolean (value, FALSE);
res = TRUE;
}
g_return_if_fail (res);
}
/**
* gst_util_set_object_arg:
* @object: the object to set the argument of
* @name: the name of the argument to set
* @value: the string value to set
*
* Converts the string value to the type of the objects argument and
* sets the argument with it.
*
* Note that this function silently returns if @object has no property named
* @name or when @value cannot be converted to the type of the property.
*/
void
gst_util_set_object_arg (GObject * object, const gchar * name,
const gchar * value)
{
GParamSpec *pspec;
GType value_type;
GValue v = { 0, };
g_return_if_fail (G_IS_OBJECT (object));
g_return_if_fail (name != NULL);
g_return_if_fail (value != NULL);
pspec = g_object_class_find_property (G_OBJECT_GET_CLASS (object), name);
if (!pspec)
return;
value_type = pspec->value_type;
GST_CAT_DEBUG_OBJECT (GST_CAT_PARAMS, object,
"pspec->flags is %d, pspec->value_type is %s",
pspec->flags, g_type_name (value_type));
if (!(pspec->flags & G_PARAM_WRITABLE))
return;
g_value_init (&v, value_type);
/* special case for element <-> xml (de)serialisation */
if (value_type == GST_TYPE_STRUCTURE && strcmp (value, "NULL") == 0) {
g_value_set_boxed (&v, NULL);
goto done;
}
if (!gst_value_deserialize_with_pspec (&v, value, pspec))
return;
done:
g_object_set_property (object, pspec->name, &v);
g_value_unset (&v);
}
/**
* gst_util_set_object_array:
* @object: the object to set the array to
* @name: the name of the property to set
* @array: a #GValueArray containing the values
*
* Transfer a #GValueArray to %GST_TYPE_ARRAY and set this value on the
* specified property name. This allow language bindings to set GST_TYPE_ARRAY
* properties which are otherwise not an accessible type.
*
* Since: 1.12
*/
gboolean
gst_util_set_object_array (GObject * object, const gchar * name,
const GValueArray * array)
{
GValue v1 = G_VALUE_INIT, v2 = G_VALUE_INIT;
gboolean ret = FALSE;
g_value_init (&v1, G_TYPE_VALUE_ARRAY);
g_value_init (&v2, GST_TYPE_ARRAY);
g_value_set_static_boxed (&v1, array);
if (g_value_transform (&v1, &v2)) {
g_object_set_property (object, name, &v2);
ret = TRUE;
}
g_value_unset (&v1);
g_value_unset (&v2);
return ret;
}
/**
* gst_util_get_object_array:
* @object: the object to set the array to
* @name: the name of the property to set
* @array: (out): a return #GValueArray
*
* Get a property of type %GST_TYPE_ARRAY and transform it into a
* #GValueArray. This allow language bindings to get GST_TYPE_ARRAY
* properties which are otherwise not an accessible type.
*
* Since: 1.12
*/
gboolean
gst_util_get_object_array (GObject * object, const gchar * name,
GValueArray ** array)
{
GValue v1 = G_VALUE_INIT, v2 = G_VALUE_INIT;
gboolean ret = FALSE;
g_value_init (&v1, G_TYPE_VALUE_ARRAY);
g_value_init (&v2, GST_TYPE_ARRAY);
g_object_get_property (object, name, &v2);
if (g_value_transform (&v2, &v1)) {
*array = g_value_get_boxed (&v1);
ret = TRUE;
}
g_value_unset (&v2);
return ret;
}
/* work around error C2520: conversion from unsigned __int64 to double
* not implemented, use signed __int64
*
* These are implemented as functions because on some platforms a 64bit int to
* double conversion is not defined/implemented.
*/
/**
* gst_util_guint64_to_gdouble:
* @value: The #guint64 value to convert to double
*
* Returns: @value casted to #gdouble
*/
gdouble
gst_util_guint64_to_gdouble (guint64 value)
{
if (value & G_GINT64_CONSTANT (0x8000000000000000))
return (gdouble) ((gint64) value) + (gdouble) 18446744073709551616.;
else
return (gdouble) ((gint64) value);
}
/**
* gst_util_gdouble_to_guint64:
* @value: The #gdouble value to convert guint64 double
*
* Returns: @value casted to #guint64
*/
guint64
gst_util_gdouble_to_guint64 (gdouble value)
{
if (value < (gdouble) 9223372036854775808.) /* 1 << 63 */
return ((guint64) ((gint64) value));
value -= (gdouble) 18446744073709551616.;
return ((guint64) ((gint64) value));
}
#ifndef HAVE_UINT128_T
/* convenience struct for getting high and low uint32 parts of
* a guint64 */
typedef union
{
guint64 ll;
struct
{
#if G_BYTE_ORDER == G_BIG_ENDIAN
guint32 high, low;
#else
guint32 low, high;
#endif
} l;
} GstUInt64;
#if defined (__x86_64__) && defined (__GNUC__)
static inline void
gst_util_uint64_mul_uint64 (GstUInt64 * c1, GstUInt64 * c0, guint64 arg1,
guint64 arg2)
{
__asm__ __volatile__ ("mulq %3":"=a" (c0->ll), "=d" (c1->ll)
:"a" (arg1), "g" (arg2)
);
}
#else /* defined (__x86_64__) */
/* multiply two 64-bit unsigned ints into a 128-bit unsigned int. the high
* and low 64 bits of the product are placed in c1 and c0 respectively.
* this operation cannot overflow. */
static inline void
gst_util_uint64_mul_uint64 (GstUInt64 * c1, GstUInt64 * c0, guint64 arg1,
guint64 arg2)
{
GstUInt64 a1, b0;
GstUInt64 v, n;
/* prepare input */
v.ll = arg1;
n.ll = arg2;
/* do 128 bits multiply
* nh nl
* * vh vl
* ----------
* a0 = vl * nl
* a1 = vl * nh
* b0 = vh * nl
* b1 = + vh * nh
* -------------------
* c1h c1l c0h c0l
*
* "a0" is optimized away, result is stored directly in c0. "b1" is
* optimized away, result is stored directly in c1.
*/
c0->ll = (guint64) v.l.low * n.l.low;
a1.ll = (guint64) v.l.low * n.l.high;
b0.ll = (guint64) v.l.high * n.l.low;
/* add the high word of a0 to the low words of a1 and b0 using c1 as
* scratch space to capture the carry. the low word of the result becomes
* the final high word of c0 */
c1->ll = (guint64) c0->l.high + a1.l.low + b0.l.low;
c0->l.high = c1->l.low;
/* add the carry from the result above (found in the high word of c1) and
* the high words of a1 and b0 to b1, the result is c1. */
c1->ll = (guint64) v.l.high * n.l.high + c1->l.high + a1.l.high + b0.l.high;
}
#endif /* defined (__x86_64__) */
#if defined (__x86_64__) && defined (__GNUC__)
static inline guint64
gst_util_div128_64 (GstUInt64 c1, GstUInt64 c0, guint64 denom)
{
guint64 res;
__asm__ __volatile__ ("divq %3":"=a" (res)
:"d" (c1.ll), "a" (c0.ll), "g" (denom)
);
return res;
}
#else
/* count leading zeros */
static inline guint
gst_util_clz (guint32 val)
{
guint s;
s = val | (val >> 1);
s |= (s >> 2);
s |= (s >> 4);
s |= (s >> 8);
s = ~(s | (s >> 16));
s = s - ((s >> 1) & 0x55555555);
s = (s & 0x33333333) + ((s >> 2) & 0x33333333);
s = (s + (s >> 4)) & 0x0f0f0f0f;
s += (s >> 8);
s = (s + (s >> 16)) & 0x3f;
return s;
}
/* based on Hacker's Delight p152 */
static inline guint64
gst_util_div128_64 (GstUInt64 c1, GstUInt64 c0, guint64 denom)
{
GstUInt64 q1, q0, rhat;
GstUInt64 v, cmp1, cmp2;
guint s;
v.ll = denom;
/* count number of leading zeroes, we know they must be in the high
* part of denom since denom > G_MAXUINT32. */
s = gst_util_clz (v.l.high);
if (s > 0) {
/* normalize divisor and dividend */
v.ll <<= s;
c1.ll = (c1.ll << s) | (c0.l.high >> (32 - s));
c0.ll <<= s;
}
q1.ll = c1.ll / v.l.high;
rhat.ll = c1.ll - q1.ll * v.l.high;
cmp1.l.high = rhat.l.low;
cmp1.l.low = c0.l.high;
cmp2.ll = q1.ll * v.l.low;
while (q1.l.high || cmp2.ll > cmp1.ll) {
q1.ll--;
rhat.ll += v.l.high;
if (rhat.l.high)
break;
cmp1.l.high = rhat.l.low;
cmp2.ll -= v.l.low;
}
c1.l.high = c1.l.low;
c1.l.low = c0.l.high;
c1.ll -= q1.ll * v.ll;
q0.ll = c1.ll / v.l.high;
rhat.ll = c1.ll - q0.ll * v.l.high;
cmp1.l.high = rhat.l.low;
cmp1.l.low = c0.l.low;
cmp2.ll = q0.ll * v.l.low;
while (q0.l.high || cmp2.ll > cmp1.ll) {
q0.ll--;
rhat.ll += v.l.high;
if (rhat.l.high)
break;
cmp1.l.high = rhat.l.low;
cmp2.ll -= v.l.low;
}
q0.l.high += q1.l.low;
return q0.ll;
}
#endif /* defined (__GNUC__) */
/* This always gives the correct result because:
* a) val <= G_MAXUINT64-1
* b) (c0,c1) <= G_MAXUINT64 * (G_MAXUINT64-1)
* or
* (c0,c1) == G_MAXUINT64 * G_MAXUINT64 and denom < G_MAXUINT64
* (note: num==denom case is handled by short path)
* This means that (c0,c1) either has enough space for val
* or that the overall result will overflow anyway.
*/
/* add correction with carry */
#define CORRECT(c0,c1,val) \
if (val) { \
if (G_MAXUINT64 - c0.ll < val) { \
if (G_UNLIKELY (c1.ll == G_MAXUINT64)) \
/* overflow */ \
return G_MAXUINT64; \
c1.ll++; \
} \
c0.ll += val; \
}
static guint64
gst_util_uint64_scale_uint64_unchecked (guint64 val, guint64 num,
guint64 denom, guint64 correct)
{
GstUInt64 c1, c0;
/* compute 128-bit numerator product */
gst_util_uint64_mul_uint64 (&c1, &c0, val, num);
/* perform rounding correction */
CORRECT (c0, c1, correct);
/* high word as big as or bigger than denom --> overflow */
if (G_UNLIKELY (c1.ll >= denom))
return G_MAXUINT64;
/* compute quotient, fits in 64 bits */
return gst_util_div128_64 (c1, c0, denom);
}
#else
#define GST_MAXUINT128 ((__uint128_t) -1)
static guint64
gst_util_uint64_scale_uint64_unchecked (guint64 val, guint64 num,
guint64 denom, guint64 correct)
{
__uint128_t tmp;
/* Calculate val * num */
tmp = ((__uint128_t) val) * ((__uint128_t) num);
/* overflow checks */
if (G_UNLIKELY (GST_MAXUINT128 - correct < tmp))
return G_MAXUINT64;
/* perform rounding correction */
tmp += correct;
/* Divide by denom */
tmp /= denom;
/* if larger than G_MAXUINT64 --> overflow */
if (G_UNLIKELY (tmp > G_MAXUINT64))
return G_MAXUINT64;
/* compute quotient, fits in 64 bits */
return (guint64) tmp;
}
#endif
#if !defined (__x86_64__) && !defined (HAVE_UINT128_T)
static inline void
gst_util_uint64_mul_uint32 (GstUInt64 * c1, GstUInt64 * c0, guint64 arg1,
guint32 arg2)
{
GstUInt64 a;
a.ll = arg1;
c0->ll = (guint64) a.l.low * arg2;
c1->ll = (guint64) a.l.high * arg2 + c0->l.high;
c0->l.high = 0;
}
/* divide a 96-bit unsigned int by a 32-bit unsigned int when we know the
* quotient fits into 64 bits. the high 64 bits and low 32 bits of the
* numerator are expected in c1 and c0 respectively. */
static inline guint64
gst_util_div96_32 (guint64 c1, guint64 c0, guint32 denom)
{
c0 += (c1 % denom) << 32;
return ((c1 / denom) << 32) + (c0 / denom);
}
static inline guint64
gst_util_uint64_scale_uint32_unchecked (guint64 val, guint32 num,
guint32 denom, guint32 correct)
{
GstUInt64 c1, c0;
/* compute 96-bit numerator product */
gst_util_uint64_mul_uint32 (&c1, &c0, val, num);
/* condition numerator based on rounding mode */
CORRECT (c0, c1, correct);
/* high 32 bits as big as or bigger than denom --> overflow */
if (G_UNLIKELY (c1.l.high >= denom))
return G_MAXUINT64;
/* compute quotient, fits in 64 bits */
return gst_util_div96_32 (c1.ll, c0.ll, denom);
}
#endif
/* the guts of the gst_util_uint64_scale() variants */
static guint64
_gst_util_uint64_scale (guint64 val, guint64 num, guint64 denom,
guint64 correct)
{
g_return_val_if_fail (denom != 0, G_MAXUINT64);
if (G_UNLIKELY (num == 0))
return 0;
if (G_UNLIKELY (num == denom))
return val;
/* on 64bits we always use a full 128bits multiply/division */
#if !defined (__x86_64__) && !defined (HAVE_UINT128_T)
/* denom is low --> try to use 96 bit muldiv */
if (G_LIKELY (denom <= G_MAXUINT32)) {
/* num is low --> use 96 bit muldiv */
if (G_LIKELY (num <= G_MAXUINT32))
return gst_util_uint64_scale_uint32_unchecked (val, (guint32) num,
(guint32) denom, correct);
/* num is high but val is low --> swap and use 96-bit muldiv */
if (G_LIKELY (val <= G_MAXUINT32))
return gst_util_uint64_scale_uint32_unchecked (num, (guint32) val,
(guint32) denom, correct);
}
#endif /* !defined (__x86_64__) && !defined (HAVE_UINT128_T) */
/* val is high and num is high --> use 128-bit muldiv */
return gst_util_uint64_scale_uint64_unchecked (val, num, denom, correct);
}
/**
* gst_util_uint64_scale:
* @val: the number to scale
* @num: the numerator of the scale ratio
* @denom: the denominator of the scale ratio
*
* Scale @val by the rational number @num / @denom, avoiding overflows and
* underflows and without loss of precision.
*
* This function can potentially be very slow if val and num are both
* greater than G_MAXUINT32.
*
* Returns: @val * @num / @denom. In the case of an overflow, this
* function returns G_MAXUINT64. If the result is not exactly
* representable as an integer it is truncated. See also
* gst_util_uint64_scale_round(), gst_util_uint64_scale_ceil(),
* gst_util_uint64_scale_int(), gst_util_uint64_scale_int_round(),
* gst_util_uint64_scale_int_ceil().
*/
guint64
gst_util_uint64_scale (guint64 val, guint64 num, guint64 denom)
{
return _gst_util_uint64_scale (val, num, denom, 0);
}
/**
* gst_util_uint64_scale_round:
* @val: the number to scale
* @num: the numerator of the scale ratio
* @denom: the denominator of the scale ratio
*
* Scale @val by the rational number @num / @denom, avoiding overflows and
* underflows and without loss of precision.
*
* This function can potentially be very slow if val and num are both
* greater than G_MAXUINT32.
*
* Returns: @val * @num / @denom. In the case of an overflow, this
* function returns G_MAXUINT64. If the result is not exactly
* representable as an integer, it is rounded to the nearest integer
* (half-way cases are rounded up). See also gst_util_uint64_scale(),
* gst_util_uint64_scale_ceil(), gst_util_uint64_scale_int(),
* gst_util_uint64_scale_int_round(), gst_util_uint64_scale_int_ceil().
*/
guint64
gst_util_uint64_scale_round (guint64 val, guint64 num, guint64 denom)
{
return _gst_util_uint64_scale (val, num, denom, denom >> 1);
}
/**
* gst_util_uint64_scale_ceil:
* @val: the number to scale
* @num: the numerator of the scale ratio
* @denom: the denominator of the scale ratio
*
* Scale @val by the rational number @num / @denom, avoiding overflows and
* underflows and without loss of precision.
*
* This function can potentially be very slow if val and num are both
* greater than G_MAXUINT32.
*
* Returns: @val * @num / @denom. In the case of an overflow, this
* function returns G_MAXUINT64. If the result is not exactly
* representable as an integer, it is rounded up. See also
* gst_util_uint64_scale(), gst_util_uint64_scale_round(),
* gst_util_uint64_scale_int(), gst_util_uint64_scale_int_round(),
* gst_util_uint64_scale_int_ceil().
*/
guint64
gst_util_uint64_scale_ceil (guint64 val, guint64 num, guint64 denom)
{
return _gst_util_uint64_scale (val, num, denom, denom - 1);
}
/* the guts of the gst_util_uint64_scale_int() variants */
static guint64
_gst_util_uint64_scale_int (guint64 val, gint num, gint denom, gint correct)
{
g_return_val_if_fail (denom > 0, G_MAXUINT64);
g_return_val_if_fail (num >= 0, G_MAXUINT64);
if (G_UNLIKELY (num == 0))
return 0;
if (G_UNLIKELY (num == denom))
return val;
if (val <= G_MAXUINT32) {
/* simple case. num and denom are not negative so casts are OK. when
* not truncating, the additions to the numerator cannot overflow
* because val*num <= G_MAXUINT32 * G_MAXINT32 < G_MAXUINT64 -
* G_MAXINT32, so there's room to add another gint32. */
val *= (guint64) num;
/* add rounding correction */
val += correct;
return val / (guint64) denom;
}
#if !defined (__x86_64__) && !defined (HAVE_UINT128_T)
/* num and denom are not negative so casts are OK */
return gst_util_uint64_scale_uint32_unchecked (val, (guint32) num,
(guint32) denom, (guint32) correct);
#else
/* always use full 128bits scale */
return gst_util_uint64_scale_uint64_unchecked (val, num, denom, correct);
#endif
}
/**
* gst_util_uint64_scale_int:
* @val: guint64 (such as a #GstClockTime) to scale.
* @num: numerator of the scale factor.
* @denom: denominator of the scale factor.
*
* Scale @val by the rational number @num / @denom, avoiding overflows and
* underflows and without loss of precision. @num must be non-negative and
* @denom must be positive.
*
* Returns: @val * @num / @denom. In the case of an overflow, this
* function returns G_MAXUINT64. If the result is not exactly
* representable as an integer, it is truncated. See also
* gst_util_uint64_scale_int_round(), gst_util_uint64_scale_int_ceil(),
* gst_util_uint64_scale(), gst_util_uint64_scale_round(),
* gst_util_uint64_scale_ceil().
*/
guint64
gst_util_uint64_scale_int (guint64 val, gint num, gint denom)
{
return _gst_util_uint64_scale_int (val, num, denom, 0);
}
/**
* gst_util_uint64_scale_int_round:
* @val: guint64 (such as a #GstClockTime) to scale.
* @num: numerator of the scale factor.
* @denom: denominator of the scale factor.
*
* Scale @val by the rational number @num / @denom, avoiding overflows and
* underflows and without loss of precision. @num must be non-negative and
* @denom must be positive.
*
* Returns: @val * @num / @denom. In the case of an overflow, this
* function returns G_MAXUINT64. If the result is not exactly
* representable as an integer, it is rounded to the nearest integer
* (half-way cases are rounded up). See also gst_util_uint64_scale_int(),
* gst_util_uint64_scale_int_ceil(), gst_util_uint64_scale(),
* gst_util_uint64_scale_round(), gst_util_uint64_scale_ceil().
*/
guint64
gst_util_uint64_scale_int_round (guint64 val, gint num, gint denom)
{
/* we can use a shift to divide by 2 because denom is required to be
* positive. */
return _gst_util_uint64_scale_int (val, num, denom, denom >> 1);
}
/**
* gst_util_uint64_scale_int_ceil:
* @val: guint64 (such as a #GstClockTime) to scale.
* @num: numerator of the scale factor.
* @denom: denominator of the scale factor.
*
* Scale @val by the rational number @num / @denom, avoiding overflows and
* underflows and without loss of precision. @num must be non-negative and
* @denom must be positive.
*
* Returns: @val * @num / @denom. In the case of an overflow, this
* function returns G_MAXUINT64. If the result is not exactly
* representable as an integer, it is rounded up. See also
* gst_util_uint64_scale_int(), gst_util_uint64_scale_int_round(),
* gst_util_uint64_scale(), gst_util_uint64_scale_round(),
* gst_util_uint64_scale_ceil().
*/
guint64
gst_util_uint64_scale_int_ceil (guint64 val, gint num, gint denom)
{
return _gst_util_uint64_scale_int (val, num, denom, denom - 1);
}
/**
* gst_util_seqnum_next:
*
* Return a constantly incrementing sequence number.
*
* This function is used internally to GStreamer to be able to determine which
* events and messages are "the same". For example, elements may set the seqnum
* on a segment-done message to be the same as that of the last seek event, to
* indicate that event and the message correspond to the same segment.
*
* This function never returns %GST_SEQNUM_INVALID (which is 0).
*
* Returns: A constantly incrementing 32-bit unsigned integer, which might
* overflow at some point. Use gst_util_seqnum_compare() to make sure
* you handle wraparound correctly.
*/
guint32
gst_util_seqnum_next (void)
{
static gint counter = 1;
gint ret = g_atomic_int_add (&counter, 1);
/* Make sure we don't return 0 */
if (G_UNLIKELY (ret == GST_SEQNUM_INVALID))
ret = g_atomic_int_add (&counter, 1);
return ret;
}
/**
* gst_util_seqnum_compare:
* @s1: A sequence number.
* @s2: Another sequence number.
*
* Compare two sequence numbers, handling wraparound.
*
* The current implementation just returns (gint32)(@s1 - @s2).
*
* Returns: A negative number if @s1 is before @s2, 0 if they are equal, or a
* positive number if @s1 is after @s2.
*/
gint32
gst_util_seqnum_compare (guint32 s1, guint32 s2)
{
return (gint32) (s1 - s2);
}
/* -----------------------------------------------------
*
* The following code will be moved out of the main
* gstreamer library someday.
*/
#include "gstpad.h"
/**
* gst_element_create_all_pads:
* @element: (transfer none): a #GstElement to create pads for
*
* Creates a pad for each pad template that is always available.
* This function is only useful during object initialization of
* subclasses of #GstElement.
*/
void
gst_element_create_all_pads (GstElement * element)
{
GList *padlist;
/* FIXME: lock element */
padlist =
gst_element_class_get_pad_template_list (GST_ELEMENT_CLASS
(G_OBJECT_GET_CLASS (element)));
while (padlist) {
GstPadTemplate *padtempl = (GstPadTemplate *) padlist->data;
if (padtempl->presence == GST_PAD_ALWAYS) {
GstPad *pad;
pad = gst_pad_new_from_template (padtempl, padtempl->name_template);
gst_element_add_pad (element, pad);
}
padlist = padlist->next;
}
}
/**
* gst_element_get_compatible_pad_template:
* @element: (transfer none): a #GstElement to get a compatible pad template for
* @compattempl: (transfer none): the #GstPadTemplate to find a compatible
* template for
*
* Retrieves a pad template from @element that is compatible with @compattempl.
* Pads from compatible templates can be linked together.
*
* Returns: (transfer none) (nullable): a compatible #GstPadTemplate,
* or %NULL if none was found. No unreferencing is necessary.
*/
GstPadTemplate *
gst_element_get_compatible_pad_template (GstElement * element,
GstPadTemplate * compattempl)
{
GstPadTemplate *newtempl = NULL;
GList *padlist;
GstElementClass *class;
gboolean compatible;
g_return_val_if_fail (element != NULL, NULL);
g_return_val_if_fail (GST_IS_ELEMENT (element), NULL);
g_return_val_if_fail (compattempl != NULL, NULL);
class = GST_ELEMENT_GET_CLASS (element);
padlist = gst_element_class_get_pad_template_list (class);
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS,
"Looking for a suitable pad template in %s out of %d templates...",
GST_ELEMENT_NAME (element), g_list_length (padlist));
while (padlist) {
GstPadTemplate *padtempl = (GstPadTemplate *) padlist->data;
/* Ignore name
* Ignore presence
* Check direction (must be opposite)
* Check caps
*/
GST_CAT_LOG (GST_CAT_CAPS,
"checking pad template %s", padtempl->name_template);
if (padtempl->direction != compattempl->direction) {
GST_CAT_DEBUG (GST_CAT_CAPS,
"compatible direction: found %s pad template \"%s\"",
padtempl->direction == GST_PAD_SRC ? "src" : "sink",
padtempl->name_template);
GST_CAT_DEBUG (GST_CAT_CAPS,
"intersecting %" GST_PTR_FORMAT, GST_PAD_TEMPLATE_CAPS (compattempl));
GST_CAT_DEBUG (GST_CAT_CAPS,
"..and %" GST_PTR_FORMAT, GST_PAD_TEMPLATE_CAPS (padtempl));
compatible = gst_caps_can_intersect (GST_PAD_TEMPLATE_CAPS (compattempl),
GST_PAD_TEMPLATE_CAPS (padtempl));
GST_CAT_DEBUG (GST_CAT_CAPS, "caps are %scompatible",
(compatible ? "" : "not "));
if (compatible) {
newtempl = padtempl;
break;
}
}
padlist = g_list_next (padlist);
}
if (newtempl)
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS,
"Returning new pad template %p", newtempl);
else
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "No compatible pad template found");
return newtempl;
}
/**
* gst_element_get_pad_from_template:
* @element: (transfer none): a #GstElement.
* @templ: (transfer none): a #GstPadTemplate belonging to @element.
*
* Gets a pad from @element described by @templ. If the presence of @templ is
* #GST_PAD_REQUEST, requests a new pad. Can return %NULL for #GST_PAD_SOMETIMES
* templates.
*
* Returns: (transfer full) (nullable): the #GstPad, or %NULL if one
* could not be found or created.
*/
static GstPad *
gst_element_get_pad_from_template (GstElement * element, GstPadTemplate * templ)
{
GstPad *ret = NULL;
GstPadPresence presence;
/* If this function is ever exported, we need check the validity of `element'
* and `templ', and to make sure the template actually belongs to the
* element. */
presence = GST_PAD_TEMPLATE_PRESENCE (templ);
switch (presence) {
case GST_PAD_ALWAYS:
case GST_PAD_SOMETIMES:
ret = gst_element_get_static_pad (element, templ->name_template);
if (!ret && presence == GST_PAD_ALWAYS)
g_warning
("Element %s has an ALWAYS template %s, but no pad of the same name",
GST_OBJECT_NAME (element), templ->name_template);
break;
case GST_PAD_REQUEST:
ret = gst_element_request_pad (element, templ, NULL, NULL);
break;
}
return ret;
}
/*
* gst_element_request_compatible_pad:
* @element: a #GstElement.
* @templ: the #GstPadTemplate to which the new pad should be able to link.
*
* Requests a pad from @element. The returned pad should be unlinked and
* compatible with @templ. Might return an existing pad, or request a new one.
*
* Returns: (nullable): a #GstPad, or %NULL if one could not be found
* or created.
*/
static GstPad *
gst_element_request_compatible_pad (GstElement * element,
GstPadTemplate * templ)
{
GstPadTemplate *templ_new;
GstPad *pad = NULL;
g_return_val_if_fail (GST_IS_ELEMENT (element), NULL);
g_return_val_if_fail (GST_IS_PAD_TEMPLATE (templ), NULL);
/* FIXME: should really loop through the templates, testing each for
* compatibility and pad availability. */
templ_new = gst_element_get_compatible_pad_template (element, templ);
if (templ_new)
pad = gst_element_get_pad_from_template (element, templ_new);
/* This can happen for non-request pads. */
if (pad && GST_PAD_PEER (pad)) {
gst_object_unref (pad);
pad = NULL;
}
return pad;
}
/*
* Checks if the source pad and the sink pad can be linked.
* Both @srcpad and @sinkpad must be unlinked and have a parent.
*/
static gboolean
gst_pad_check_link (GstPad * srcpad, GstPad * sinkpad)
{
/* generic checks */
g_return_val_if_fail (GST_IS_PAD (srcpad), FALSE);
g_return_val_if_fail (GST_IS_PAD (sinkpad), FALSE);
GST_CAT_INFO (GST_CAT_PADS, "trying to link %s:%s and %s:%s",
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (sinkpad));
if (GST_PAD_PEER (srcpad) != NULL) {
GST_CAT_INFO (GST_CAT_PADS, "Source pad %s:%s has a peer, failed",
GST_DEBUG_PAD_NAME (srcpad));
return FALSE;
}
if (GST_PAD_PEER (sinkpad) != NULL) {
GST_CAT_INFO (GST_CAT_PADS, "Sink pad %s:%s has a peer, failed",
GST_DEBUG_PAD_NAME (sinkpad));
return FALSE;
}
if (!GST_PAD_IS_SRC (srcpad)) {
GST_CAT_INFO (GST_CAT_PADS, "Src pad %s:%s is not source pad, failed",
GST_DEBUG_PAD_NAME (srcpad));
return FALSE;
}
if (!GST_PAD_IS_SINK (sinkpad)) {
GST_CAT_INFO (GST_CAT_PADS, "Sink pad %s:%s is not sink pad, failed",
GST_DEBUG_PAD_NAME (sinkpad));
return FALSE;
}
if (GST_PAD_PARENT (srcpad) == NULL) {
GST_CAT_INFO (GST_CAT_PADS, "Src pad %s:%s has no parent, failed",
GST_DEBUG_PAD_NAME (srcpad));
return FALSE;
}
if (GST_PAD_PARENT (sinkpad) == NULL) {
GST_CAT_INFO (GST_CAT_PADS, "Sink pad %s:%s has no parent, failed",
GST_DEBUG_PAD_NAME (srcpad));
return FALSE;
}
return TRUE;
}
/**
* gst_element_get_compatible_pad:
* @element: (transfer none): a #GstElement in which the pad should be found.
* @pad: (transfer none): the #GstPad to find a compatible one for.
* @caps: (nullable): the #GstCaps to use as a filter.
*
* Looks for an unlinked pad to which the given pad can link. It is not
* guaranteed that linking the pads will work, though it should work in most
* cases.
*
* This function will first attempt to find a compatible unlinked ALWAYS pad,
* and if none can be found, it will request a compatible REQUEST pad by looking
* at the templates of @element.
*
* Returns: (transfer full) (nullable): the #GstPad to which a link
* can be made, or %NULL if one cannot be found. gst_object_unref()
* after usage.
*/
GstPad *
gst_element_get_compatible_pad (GstElement * element, GstPad * pad,
GstCaps * caps)
{
GstIterator *pads;
GstPadTemplate *templ;
GstCaps *templcaps;
GstPad *foundpad = NULL;
gboolean done;
GValue padptr = { 0, };
g_return_val_if_fail (GST_IS_ELEMENT (element), NULL);
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
GST_CAT_DEBUG_OBJECT (GST_CAT_ELEMENT_PADS, element,
"finding pad compatible with %s:%s", GST_DEBUG_PAD_NAME (pad));
g_return_val_if_fail (GST_PAD_PEER (pad) == NULL, NULL);
done = FALSE;
/* try to get an existing unlinked pad */
if (GST_PAD_IS_SRC (pad)) {
pads = gst_element_iterate_sink_pads (element);
} else if (GST_PAD_IS_SINK (pad)) {
pads = gst_element_iterate_src_pads (element);
} else {
pads = gst_element_iterate_pads (element);
}
while (!done) {
switch (gst_iterator_next (pads, &padptr)) {
case GST_ITERATOR_OK:
{
GstPad *peer;
GstPad *current;
GstPad *srcpad;
GstPad *sinkpad;
current = g_value_get_object (&padptr);
GST_CAT_LOG_OBJECT (GST_CAT_ELEMENT_PADS, element,
"examining pad %s:%s", GST_DEBUG_PAD_NAME (current));
if (GST_PAD_IS_SRC (current)) {
srcpad = current;
sinkpad = pad;
} else {
srcpad = pad;
sinkpad = current;
}
peer = gst_pad_get_peer (current);
if (peer == NULL && gst_pad_check_link (srcpad, sinkpad)) {
GstCaps *temp, *intersection;
gboolean compatible;
/* Now check if the two pads' caps are compatible */
temp = gst_pad_query_caps (pad, NULL);
if (caps) {
intersection = gst_caps_intersect (temp, caps);
gst_caps_unref (temp);
} else {
intersection = temp;
}
temp = gst_pad_query_caps (current, NULL);
compatible = gst_caps_can_intersect (temp, intersection);
gst_caps_unref (temp);
gst_caps_unref (intersection);
if (compatible) {
GST_CAT_DEBUG_OBJECT (GST_CAT_ELEMENT_PADS, element,
"found existing unlinked compatible pad %s:%s",
GST_DEBUG_PAD_NAME (current));
gst_iterator_free (pads);
current = gst_object_ref (current);
g_value_unset (&padptr);
return current;
} else {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "incompatible pads");
}
} else {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS,
"already linked or cannot be linked (peer = %" GST_PTR_FORMAT ")",
peer);
}
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "unreffing pads");
g_value_reset (&padptr);
if (peer)
gst_object_unref (peer);
break;
}
case GST_ITERATOR_DONE:
done = TRUE;
break;
case GST_ITERATOR_RESYNC:
gst_iterator_resync (pads);
break;
case GST_ITERATOR_ERROR:
g_assert_not_reached ();
break;
}
}
g_value_unset (&padptr);
gst_iterator_free (pads);
GST_CAT_DEBUG_OBJECT (GST_CAT_ELEMENT_PADS, element,
"Could not find a compatible unlinked always pad to link to %s:%s, now checking request pads",
GST_DEBUG_PAD_NAME (pad));
/* try to create a new one */
/* requesting is a little crazy, we need a template. Let's create one */
templcaps = gst_pad_query_caps (pad, NULL);
if (caps) {
GstCaps *inter = gst_caps_intersect (templcaps, caps);
gst_caps_unref (templcaps);
templcaps = inter;
}
templ = gst_pad_template_new ((gchar *) GST_PAD_NAME (pad),
GST_PAD_DIRECTION (pad), GST_PAD_ALWAYS, templcaps);
gst_caps_unref (templcaps);
foundpad = gst_element_request_compatible_pad (element, templ);
gst_object_unref (templ);
if (foundpad) {
GST_CAT_DEBUG_OBJECT (GST_CAT_ELEMENT_PADS, element,
"found existing request pad %s:%s", GST_DEBUG_PAD_NAME (foundpad));
return foundpad;
}
GST_CAT_INFO_OBJECT (GST_CAT_ELEMENT_PADS, element,
"Could not find a compatible pad to link to %s:%s",
GST_DEBUG_PAD_NAME (pad));
return NULL;
}
/**
* gst_element_state_get_name:
* @state: a #GstState to get the name of.
*
* Gets a string representing the given state.
*
* Returns: (transfer none): a string with the name of the state.
*/
const gchar *
gst_element_state_get_name (GstState state)
{
switch (state) {
case GST_STATE_VOID_PENDING:
return "VOID_PENDING";
case GST_STATE_NULL:
return "NULL";
case GST_STATE_READY:
return "READY";
case GST_STATE_PLAYING:
return "PLAYING";
case GST_STATE_PAUSED:
return "PAUSED";
default:
/* This is a memory leak */
return g_strdup_printf ("UNKNOWN!(%d)", state);
}
}
/**
* gst_element_state_change_return_get_name:
* @state_ret: a #GstStateChangeReturn to get the name of.
*
* Gets a string representing the given state change result.
*
* Returns: (transfer none): a string with the name of the state
* result.
*/
const gchar *
gst_element_state_change_return_get_name (GstStateChangeReturn state_ret)
{
switch (state_ret) {
case GST_STATE_CHANGE_FAILURE:
return "FAILURE";
case GST_STATE_CHANGE_SUCCESS:
return "SUCCESS";
case GST_STATE_CHANGE_ASYNC:
return "ASYNC";
case GST_STATE_CHANGE_NO_PREROLL:
return "NO PREROLL";
default:
/* This is a memory leak */
return g_strdup_printf ("UNKNOWN!(%d)", state_ret);
}
}
/**
* gst_state_change_get_name:
* @transition: a #GstStateChange to get the name of.
*
* Gets a string representing the given state transition.
*
* Returns: (transfer none): a string with the name of the state
* result.
*
* Since: 1.14
*/
const gchar *
gst_state_change_get_name (GstStateChange transition)
{
switch (transition) {
case GST_STATE_CHANGE_NULL_TO_READY:
return "NULL->READY";
case GST_STATE_CHANGE_READY_TO_PAUSED:
return "READY->PAUSED";
case GST_STATE_CHANGE_PAUSED_TO_PLAYING:
return "PAUSED->PLAYING";
case GST_STATE_CHANGE_PLAYING_TO_PAUSED:
return "PLAYING->PAUSED";
case GST_STATE_CHANGE_PAUSED_TO_READY:
return "PAUSED->READY";
case GST_STATE_CHANGE_READY_TO_NULL:
return "READY->NULL";
case GST_STATE_CHANGE_NULL_TO_NULL:
return "NULL->NULL";
case GST_STATE_CHANGE_READY_TO_READY:
return "READY->READY";
case GST_STATE_CHANGE_PAUSED_TO_PAUSED:
return "PAUSED->PAUSED";
case GST_STATE_CHANGE_PLAYING_TO_PLAYING:
return "PLAYING->PLAYING";
}
return "Unknown state return";
}
static gboolean
gst_element_factory_can_accept_all_caps_in_direction (GstElementFactory *
factory, const GstCaps * caps, GstPadDirection direction)
{
GList *templates;
g_return_val_if_fail (factory != NULL, FALSE);
g_return_val_if_fail (caps != NULL, FALSE);
templates = factory->staticpadtemplates;
while (templates) {
GstStaticPadTemplate *template = (GstStaticPadTemplate *) templates->data;
if (template->direction == direction) {
GstCaps *templcaps = gst_static_caps_get (&template->static_caps);
if (gst_caps_is_always_compatible (caps, templcaps)) {
gst_caps_unref (templcaps);
return TRUE;
}
gst_caps_unref (templcaps);
}
templates = g_list_next (templates);
}
return FALSE;
}
static gboolean
gst_element_factory_can_accept_any_caps_in_direction (GstElementFactory *
factory, const GstCaps * caps, GstPadDirection direction)
{
GList *templates;
g_return_val_if_fail (factory != NULL, FALSE);
g_return_val_if_fail (caps != NULL, FALSE);
templates = factory->staticpadtemplates;
while (templates) {
GstStaticPadTemplate *template = (GstStaticPadTemplate *) templates->data;
if (template->direction == direction) {
GstCaps *templcaps = gst_static_caps_get (&template->static_caps);
if (gst_caps_can_intersect (caps, templcaps)) {
gst_caps_unref (templcaps);
return TRUE;
}
gst_caps_unref (templcaps);
}
templates = g_list_next (templates);
}
return FALSE;
}
/**
* gst_element_factory_can_sink_all_caps:
* @factory: factory to query
* @caps: the caps to check
*
* Checks if the factory can sink all possible capabilities.
*
* Returns: %TRUE if the caps are fully compatible.
*/
gboolean
gst_element_factory_can_sink_all_caps (GstElementFactory * factory,
const GstCaps * caps)
{
return gst_element_factory_can_accept_all_caps_in_direction (factory, caps,
GST_PAD_SINK);
}
/**
* gst_element_factory_can_src_all_caps:
* @factory: factory to query
* @caps: the caps to check
*
* Checks if the factory can src all possible capabilities.
*
* Returns: %TRUE if the caps are fully compatible.
*/
gboolean
gst_element_factory_can_src_all_caps (GstElementFactory * factory,
const GstCaps * caps)
{
return gst_element_factory_can_accept_all_caps_in_direction (factory, caps,
GST_PAD_SRC);
}
/**
* gst_element_factory_can_sink_any_caps:
* @factory: factory to query
* @caps: the caps to check
*
* Checks if the factory can sink any possible capability.
*
* Returns: %TRUE if the caps have a common subset.
*/
gboolean
gst_element_factory_can_sink_any_caps (GstElementFactory * factory,
const GstCaps * caps)
{
return gst_element_factory_can_accept_any_caps_in_direction (factory, caps,
GST_PAD_SINK);
}
/**
* gst_element_factory_can_src_any_caps:
* @factory: factory to query
* @caps: the caps to check
*
* Checks if the factory can src any possible capability.
*
* Returns: %TRUE if the caps have a common subset.
*/
gboolean
gst_element_factory_can_src_any_caps (GstElementFactory * factory,
const GstCaps * caps)
{
return gst_element_factory_can_accept_any_caps_in_direction (factory, caps,
GST_PAD_SRC);
}
/* if return val is true, *direct_child is a caller-owned ref on the direct
* child of ancestor that is part of object's ancestry */
static gboolean
object_has_ancestor (GstObject * object, GstObject * ancestor,
GstObject ** direct_child)
{
GstObject *child, *parent;
if (direct_child)
*direct_child = NULL;
child = gst_object_ref (object);
parent = gst_object_get_parent (object);
while (parent) {
if (ancestor == parent) {
if (direct_child)
*direct_child = child;
else
gst_object_unref (child);
gst_object_unref (parent);
return TRUE;
}
gst_object_unref (child);
child = parent;
parent = gst_object_get_parent (parent);
}
gst_object_unref (child);
return FALSE;
}
/* caller owns return */
static GstObject *
find_common_root (GstObject * o1, GstObject * o2)
{
GstObject *top = o1;
GstObject *kid1, *kid2;
GstObject *root = NULL;
while (GST_OBJECT_PARENT (top))
top = GST_OBJECT_PARENT (top);
/* the itsy-bitsy spider... */
if (!object_has_ancestor (o2, top, &kid2))
return NULL;
root = gst_object_ref (top);
while (TRUE) {
if (!object_has_ancestor (o1, kid2, &kid1)) {
gst_object_unref (kid2);
return root;
}
gst_object_unref (root);
root = kid2;
if (!object_has_ancestor (o2, kid1, &kid2)) {
gst_object_unref (kid1);
return root;
}
gst_object_unref (root);
root = kid1;
}
}
/* caller does not own return */
static GstPad *
ghost_up (GstElement * e, GstPad * pad)
{
static gint ghost_pad_index = 0;
GstPad *gpad;
gchar *name;
GstState current;
GstState next;
GstObject *parent = GST_OBJECT_PARENT (e);
name = g_strdup_printf ("ghost%d", ghost_pad_index++);
gpad = gst_ghost_pad_new (name, pad);
g_free (name);
GST_STATE_LOCK (parent);
gst_element_get_state (GST_ELEMENT (parent), &current, &next, 0);
if (current > GST_STATE_READY || next >= GST_STATE_PAUSED)
gst_pad_set_active (gpad, TRUE);
if (!gst_element_add_pad ((GstElement *) parent, gpad)) {
g_warning ("Pad named %s already exists in element %s\n",
GST_OBJECT_NAME (gpad), GST_OBJECT_NAME (parent));
GST_STATE_UNLOCK (parent);
return NULL;
}
GST_STATE_UNLOCK (parent);
return gpad;
}
static void
remove_pad (gpointer ppad, gpointer unused)
{
GstPad *pad = ppad;
if (!gst_element_remove_pad ((GstElement *) GST_OBJECT_PARENT (pad), pad))
g_warning ("Couldn't remove pad %s from element %s",
GST_OBJECT_NAME (pad), GST_OBJECT_NAME (GST_OBJECT_PARENT (pad)));
}
static gboolean
prepare_link_maybe_ghosting (GstPad ** src, GstPad ** sink,
GSList ** pads_created)
{
GstObject *root;
GstObject *e1, *e2;
GSList *pads_created_local = NULL;
g_assert (pads_created);
e1 = GST_OBJECT_PARENT (*src);
e2 = GST_OBJECT_PARENT (*sink);
if (G_UNLIKELY (e1 == NULL)) {
GST_CAT_WARNING (GST_CAT_ELEMENT_PADS,
"Trying to ghost a pad that doesn't have a parent: %" GST_PTR_FORMAT,
*src);
return FALSE;
}
if (G_UNLIKELY (e2 == NULL)) {
GST_CAT_WARNING (GST_CAT_ELEMENT_PADS,
"Trying to ghost a pad that doesn't have a parent: %" GST_PTR_FORMAT,
*sink);
return FALSE;
}
if (GST_OBJECT_PARENT (e1) == GST_OBJECT_PARENT (e2)) {
GST_CAT_INFO (GST_CAT_PADS, "%s and %s in same bin, no need for ghost pads",
GST_OBJECT_NAME (e1), GST_OBJECT_NAME (e2));
return TRUE;
}
GST_CAT_INFO (GST_CAT_PADS, "%s and %s not in same bin, making ghost pads",
GST_OBJECT_NAME (e1), GST_OBJECT_NAME (e2));
/* we need to setup some ghost pads */
root = find_common_root (e1, e2);
if (!root) {
if (GST_OBJECT_PARENT (e1) == NULL)
g_warning ("Trying to link elements %s and %s that don't share a common "
"ancestor: %s hasn't been added to a bin or pipeline, but %s is in %s",
GST_ELEMENT_NAME (e1), GST_ELEMENT_NAME (e2),
GST_ELEMENT_NAME (e1), GST_ELEMENT_NAME (e2),
GST_ELEMENT_NAME (GST_OBJECT_PARENT (e2)));
else if (GST_OBJECT_PARENT (e2) == NULL)
g_warning ("Trying to link elements %s and %s that don't share a common "
"ancestor: %s hasn't been added to a bin or pipeline, and %s is in %s",
GST_ELEMENT_NAME (e1), GST_ELEMENT_NAME (e2),
GST_ELEMENT_NAME (e2), GST_ELEMENT_NAME (e1),
GST_ELEMENT_NAME (GST_OBJECT_PARENT (e1)));
else
g_warning ("Trying to link elements %s and %s that don't share a common "
"ancestor: %s is in %s, and %s is in %s",
GST_ELEMENT_NAME (e1), GST_ELEMENT_NAME (e2),
GST_ELEMENT_NAME (e1), GST_ELEMENT_NAME (GST_OBJECT_PARENT (e1)),
GST_ELEMENT_NAME (e2), GST_ELEMENT_NAME (GST_OBJECT_PARENT (e2)));
return FALSE;
}
while (GST_OBJECT_PARENT (e1) != root) {
*src = ghost_up ((GstElement *) e1, *src);
if (!*src)
goto cleanup_fail;
e1 = GST_OBJECT_PARENT (*src);
pads_created_local = g_slist_prepend (pads_created_local, *src);
}
while (GST_OBJECT_PARENT (e2) != root) {
*sink = ghost_up ((GstElement *) e2, *sink);
if (!*sink)
goto cleanup_fail;
e2 = GST_OBJECT_PARENT (*sink);
pads_created_local = g_slist_prepend (pads_created_local, *sink);
}
gst_object_unref (root);
*pads_created = g_slist_concat (*pads_created, pads_created_local);
return TRUE;
cleanup_fail:
gst_object_unref (root);
g_slist_foreach (pads_created_local, remove_pad, NULL);
g_slist_free (pads_created_local);
return FALSE;
}
static gboolean
pad_link_maybe_ghosting (GstPad * src, GstPad * sink, GstPadLinkCheck flags)
{
GSList *pads_created = NULL;
gboolean ret;
if (!prepare_link_maybe_ghosting (&src, &sink, &pads_created)) {
ret = FALSE;
} else {
ret = (gst_pad_link_full (src, sink, flags) == GST_PAD_LINK_OK);
}
if (!ret) {
g_slist_foreach (pads_created, remove_pad, NULL);
}
g_slist_free (pads_created);
return ret;
}
/**
* gst_pad_link_maybe_ghosting_full:
* @src: a #GstPad
* @sink: a #GstPad
* @flags: some #GstPadLinkCheck flags
*
* Links @src to @sink, creating any #GstGhostPad's in between as necessary.
*
* This is a convenience function to save having to create and add intermediate
* #GstGhostPad's as required for linking across #GstBin boundaries.
*
* If @src or @sink pads don't have parent elements or do not share a common
* ancestor, the link will fail.
*
* Calling gst_pad_link_maybe_ghosting_full() with
* @flags == %GST_PAD_LINK_CHECK_DEFAULT is the recommended way of linking
* pads with safety checks applied.
*
* Returns: whether the link succeeded.
*
* Since: 1.10
*/
gboolean
gst_pad_link_maybe_ghosting_full (GstPad * src, GstPad * sink,
GstPadLinkCheck flags)
{
g_return_val_if_fail (GST_IS_PAD (src), FALSE);
g_return_val_if_fail (GST_IS_PAD (sink), FALSE);
return pad_link_maybe_ghosting (src, sink, flags);
}
/**
* gst_pad_link_maybe_ghosting:
* @src: a #GstPad
* @sink: a #GstPad
*
* Links @src to @sink, creating any #GstGhostPad's in between as necessary.
*
* This is a convenience function to save having to create and add intermediate
* #GstGhostPad's as required for linking across #GstBin boundaries.
*
* If @src or @sink pads don't have parent elements or do not share a common
* ancestor, the link will fail.
*
* Returns: whether the link succeeded.
*
* Since: 1.10
*/
gboolean
gst_pad_link_maybe_ghosting (GstPad * src, GstPad * sink)
{
g_return_val_if_fail (GST_IS_PAD (src), FALSE);
g_return_val_if_fail (GST_IS_PAD (sink), FALSE);
return gst_pad_link_maybe_ghosting_full (src, sink,
GST_PAD_LINK_CHECK_DEFAULT);
}
static void
release_and_unref_pad (GstElement * element, GstPad * pad, gboolean requestpad)
{
if (pad) {
if (requestpad)
gst_element_release_request_pad (element, pad);
gst_object_unref (pad);
}
}
/**
* gst_element_link_pads_full:
* @src: a #GstElement containing the source pad.
* @srcpadname: (nullable): the name of the #GstPad in source element
* or %NULL for any pad.
* @dest: (transfer none): the #GstElement containing the destination pad.
* @destpadname: (nullable): the name of the #GstPad in destination element,
* or %NULL for any pad.
* @flags: the #GstPadLinkCheck to be performed when linking pads.
*
* Links the two named pads of the source and destination elements.
* Side effect is that if one of the pads has no parent, it becomes a
* child of the parent of the other element. If they have different
* parents, the link fails.
*
* Calling gst_element_link_pads_full() with @flags == %GST_PAD_LINK_CHECK_DEFAULT
* is the same as calling gst_element_link_pads() and the recommended way of
* linking pads with safety checks applied.
*
* This is a convenience function for gst_pad_link_full().
*
* Returns: %TRUE if the pads could be linked, %FALSE otherwise.
*/
gboolean
gst_element_link_pads_full (GstElement * src, const gchar * srcpadname,
GstElement * dest, const gchar * destpadname, GstPadLinkCheck flags)
{
const GList *srcpads, *destpads, *srctempls, *desttempls, *l;
GstPad *srcpad, *destpad;
GstPadTemplate *srctempl, *desttempl;
GstElementClass *srcclass, *destclass;
gboolean srcrequest, destrequest;
/* checks */
g_return_val_if_fail (GST_IS_ELEMENT (src), FALSE);
g_return_val_if_fail (GST_IS_ELEMENT (dest), FALSE);
GST_CAT_INFO (GST_CAT_ELEMENT_PADS,
"trying to link element %s:%s to element %s:%s", GST_ELEMENT_NAME (src),
srcpadname ? srcpadname : "(any)", GST_ELEMENT_NAME (dest),
destpadname ? destpadname : "(any)");
srcrequest = FALSE;
destrequest = FALSE;
/* get a src pad */
if (srcpadname) {
/* name specified, look it up */
if (!(srcpad = gst_element_get_static_pad (src, srcpadname))) {
if ((srcpad = gst_element_request_pad_simple (src, srcpadname)))
srcrequest = TRUE;
}
if (!srcpad) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "no pad %s:%s",
GST_ELEMENT_NAME (src), srcpadname);
return FALSE;
} else {
if (!(GST_PAD_DIRECTION (srcpad) == GST_PAD_SRC)) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "pad %s:%s is no src pad",
GST_DEBUG_PAD_NAME (srcpad));
release_and_unref_pad (src, srcpad, srcrequest);
return FALSE;
}
if (GST_PAD_PEER (srcpad) != NULL) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS,
"pad %s:%s is already linked to %s:%s", GST_DEBUG_PAD_NAME (srcpad),
GST_DEBUG_PAD_NAME (GST_PAD_PEER (srcpad)));
/* already linked request pads look like static pads, so the request pad
* was never requested a second time above, so no need to release it */
gst_object_unref (srcpad);
return FALSE;
}
}
srcpads = NULL;
} else {
/* no name given, get the first available pad */
GST_OBJECT_LOCK (src);
srcpads = GST_ELEMENT_PADS (src);
srcpad = srcpads ? GST_PAD_CAST (srcpads->data) : NULL;
if (srcpad)
gst_object_ref (srcpad);
GST_OBJECT_UNLOCK (src);
}
/* get a destination pad */
if (destpadname) {
/* name specified, look it up */
if (!(destpad = gst_element_get_static_pad (dest, destpadname))) {
if ((destpad = gst_element_request_pad_simple (dest, destpadname)))
destrequest = TRUE;
}
if (!destpad) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "no pad %s:%s",
GST_ELEMENT_NAME (dest), destpadname);
release_and_unref_pad (src, srcpad, srcrequest);
return FALSE;
} else {
if (!(GST_PAD_DIRECTION (destpad) == GST_PAD_SINK)) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "pad %s:%s is no sink pad",
GST_DEBUG_PAD_NAME (destpad));
release_and_unref_pad (src, srcpad, srcrequest);
release_and_unref_pad (dest, destpad, destrequest);
return FALSE;
}
if (GST_PAD_PEER (destpad) != NULL) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS,
"pad %s:%s is already linked to %s:%s",
GST_DEBUG_PAD_NAME (destpad),
GST_DEBUG_PAD_NAME (GST_PAD_PEER (destpad)));
release_and_unref_pad (src, srcpad, srcrequest);
/* already linked request pads look like static pads, so the request pad
* was never requested a second time above, so no need to release it */
gst_object_unref (destpad);
return FALSE;
}
}
destpads = NULL;
} else {
/* no name given, get the first available pad */
GST_OBJECT_LOCK (dest);
destpads = GST_ELEMENT_PADS (dest);
destpad = destpads ? GST_PAD_CAST (destpads->data) : NULL;
if (destpad)
gst_object_ref (destpad);
GST_OBJECT_UNLOCK (dest);
}
if (srcpadname && destpadname) {
gboolean result;
/* two explicitly specified pads */
result = pad_link_maybe_ghosting (srcpad, destpad, flags);
if (result) {
gst_object_unref (srcpad);
gst_object_unref (destpad);
} else {
release_and_unref_pad (src, srcpad, srcrequest);
release_and_unref_pad (dest, destpad, destrequest);
}
return result;
}
if (srcpad) {
/* loop through the allowed pads in the source, trying to find a
* compatible destination pad */
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS,
"looping through allowed src and dest pads");
do {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "trying src pad %s:%s",
GST_DEBUG_PAD_NAME (srcpad));
if ((GST_PAD_DIRECTION (srcpad) == GST_PAD_SRC) &&
(GST_PAD_PEER (srcpad) == NULL)) {
gboolean temprequest = FALSE;
GstPad *temp;
if (destpadname) {
temp = destpad;
gst_object_ref (temp);
} else {
temp = gst_element_get_compatible_pad (dest, srcpad, NULL);
if (temp && GST_PAD_PAD_TEMPLATE (temp)
&& GST_PAD_TEMPLATE_PRESENCE (GST_PAD_PAD_TEMPLATE (temp)) ==
GST_PAD_REQUEST) {
temprequest = TRUE;
}
}
if (temp && pad_link_maybe_ghosting (srcpad, temp, flags)) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "linked pad %s:%s to pad %s:%s",
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (temp));
if (destpad)
gst_object_unref (destpad);
gst_object_unref (srcpad);
gst_object_unref (temp);
return TRUE;
}
if (temp) {
if (temprequest)
gst_element_release_request_pad (dest, temp);
gst_object_unref (temp);
}
}
/* find a better way for this mess */
if (srcpads) {
srcpads = g_list_next (srcpads);
if (srcpads) {
gst_object_unref (srcpad);
srcpad = GST_PAD_CAST (srcpads->data);
gst_object_ref (srcpad);
}
}
} while (srcpads);
}
if (srcpadname) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "no link possible from %s:%s to %s",
GST_DEBUG_PAD_NAME (srcpad), GST_ELEMENT_NAME (dest));
/* no need to release any request pad as both src- and destpadname must be
* set to end up here, but this case has already been taken care of above */
if (destpad)
gst_object_unref (destpad);
destpad = NULL;
}
if (srcpad) {
release_and_unref_pad (src, srcpad, srcrequest);
srcpad = NULL;
}
if (destpad) {
/* loop through the existing pads in the destination */
do {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "trying dest pad %s:%s",
GST_DEBUG_PAD_NAME (destpad));
if ((GST_PAD_DIRECTION (destpad) == GST_PAD_SINK) &&
(GST_PAD_PEER (destpad) == NULL)) {
GstPad *temp = gst_element_get_compatible_pad (src, destpad, NULL);
gboolean temprequest = FALSE;
if (temp && GST_PAD_PAD_TEMPLATE (temp)
&& GST_PAD_TEMPLATE_PRESENCE (GST_PAD_PAD_TEMPLATE (temp)) ==
GST_PAD_REQUEST) {
temprequest = TRUE;
}
if (temp && pad_link_maybe_ghosting (temp, destpad, flags)) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "linked pad %s:%s to pad %s:%s",
GST_DEBUG_PAD_NAME (temp), GST_DEBUG_PAD_NAME (destpad));
gst_object_unref (temp);
gst_object_unref (destpad);
return TRUE;
}
release_and_unref_pad (src, temp, temprequest);
}
if (destpads) {
destpads = g_list_next (destpads);
if (destpads) {
gst_object_unref (destpad);
destpad = GST_PAD_CAST (destpads->data);
gst_object_ref (destpad);
}
}
} while (destpads);
}
if (destpadname) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "no link possible from %s to %s:%s",
GST_ELEMENT_NAME (src), GST_DEBUG_PAD_NAME (destpad));
release_and_unref_pad (dest, destpad, destrequest);
return FALSE;
} else {
/* no need to release any request pad as the case of unset destpatname and
* destpad being a request pad has already been taken care of when looking
* though the destination pads above */
if (destpad) {
gst_object_unref (destpad);
}
destpad = NULL;
}
srcclass = GST_ELEMENT_GET_CLASS (src);
destclass = GST_ELEMENT_GET_CLASS (dest);
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS,
"we might have request pads on both sides, checking...");
srctempls = gst_element_class_get_pad_template_list (srcclass);
desttempls = gst_element_class_get_pad_template_list (destclass);
if (srctempls && desttempls) {
while (srctempls) {
srctempl = (GstPadTemplate *) srctempls->data;
if (srctempl->presence == GST_PAD_REQUEST) {
for (l = desttempls; l; l = l->next) {
desttempl = (GstPadTemplate *) l->data;
if (desttempl->presence == GST_PAD_REQUEST &&
desttempl->direction != srctempl->direction) {
GstCaps *srccaps, *destcaps;
srccaps = gst_pad_template_get_caps (srctempl);
destcaps = gst_pad_template_get_caps (desttempl);
if (gst_caps_is_always_compatible (srccaps, destcaps)) {
srcpad =
gst_element_request_pad (src, srctempl,
srctempl->name_template, NULL);
destpad =
gst_element_request_pad (dest, desttempl,
desttempl->name_template, NULL);
if (srcpad && destpad
&& pad_link_maybe_ghosting (srcpad, destpad, flags)) {
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS,
"linked pad %s:%s to pad %s:%s",
GST_DEBUG_PAD_NAME (srcpad), GST_DEBUG_PAD_NAME (destpad));
gst_object_unref (srcpad);
gst_object_unref (destpad);
gst_caps_unref (srccaps);
gst_caps_unref (destcaps);
return TRUE;
}
/* it failed, so we release the request pads */
if (srcpad) {
gst_element_release_request_pad (src, srcpad);
gst_object_unref (srcpad);
}
if (destpad) {
gst_element_release_request_pad (dest, destpad);
gst_object_unref (destpad);
}
}
gst_caps_unref (srccaps);
gst_caps_unref (destcaps);
}
}
}
srctempls = srctempls->next;
}
}
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "no link possible from %s to %s",
GST_ELEMENT_NAME (src), GST_ELEMENT_NAME (dest));
return FALSE;
}
/**
* gst_element_link_pads:
* @src: a #GstElement containing the source pad.
* @srcpadname: (nullable): the name of the #GstPad in source element
* or %NULL for any pad.
* @dest: (transfer none): the #GstElement containing the destination pad.
* @destpadname: (nullable): the name of the #GstPad in destination element,
* or %NULL for any pad.
*
* Links the two named pads of the source and destination elements.
* Side effect is that if one of the pads has no parent, it becomes a
* child of the parent of the other element. If they have different
* parents, the link fails.
*
* Returns: %TRUE if the pads could be linked, %FALSE otherwise.
*/
gboolean
gst_element_link_pads (GstElement * src, const gchar * srcpadname,
GstElement * dest, const gchar * destpadname)
{
return gst_element_link_pads_full (src, srcpadname, dest, destpadname,
GST_PAD_LINK_CHECK_DEFAULT);
}
/**
* gst_element_link_pads_filtered:
* @src: a #GstElement containing the source pad.
* @srcpadname: (nullable): the name of the #GstPad in source element
* or %NULL for any pad.
* @dest: (transfer none): the #GstElement containing the destination pad.
* @destpadname: (nullable): the name of the #GstPad in destination element
* or %NULL for any pad.
* @filter: (transfer none) (nullable): the #GstCaps to filter the link,
* or %NULL for no filter.
*
* Links the two named pads of the source and destination elements. Side effect
* is that if one of the pads has no parent, it becomes a child of the parent of
* the other element. If they have different parents, the link fails. If @caps
* is not %NULL, makes sure that the caps of the link is a subset of @caps.
*
* Returns: %TRUE if the pads could be linked, %FALSE otherwise.
*/
gboolean
gst_element_link_pads_filtered (GstElement * src, const gchar * srcpadname,
GstElement * dest, const gchar * destpadname, GstCaps * filter)
{
/* checks */
g_return_val_if_fail (GST_IS_ELEMENT (src), FALSE);
g_return_val_if_fail (GST_IS_ELEMENT (dest), FALSE);
g_return_val_if_fail (filter == NULL || GST_IS_CAPS (filter), FALSE);
if (filter) {
GstElement *capsfilter;
GstObject *parent;
GstState state, pending;
gboolean lr1, lr2;
capsfilter = gst_element_factory_make ("capsfilter", NULL);
if (!capsfilter) {
GST_CAT_ERROR (GST_CAT_ELEMENT_PADS, "Could not make a capsfilter");
return FALSE;
}
parent = gst_object_get_parent (GST_OBJECT (src));
g_return_val_if_fail (GST_IS_BIN (parent), FALSE);
gst_element_get_state (GST_ELEMENT_CAST (parent), &state, &pending, 0);
if (!gst_bin_add (GST_BIN (parent), capsfilter)) {
GST_CAT_ERROR (GST_CAT_ELEMENT_PADS, "Could not add capsfilter");
gst_object_unref (parent);
return FALSE;
}
if (pending != GST_STATE_VOID_PENDING)
state = pending;
gst_element_set_state (capsfilter, state);
gst_object_unref (parent);
g_object_set (capsfilter, "caps", filter, NULL);
lr1 = gst_element_link_pads (src, srcpadname, capsfilter, "sink");
lr2 = gst_element_link_pads (capsfilter, "src", dest, destpadname);
if (lr1 && lr2) {
return TRUE;
} else {
if (!lr1) {
GST_INFO ("Could not link pads: %s:%s - capsfilter:sink",
GST_ELEMENT_NAME (src), srcpadname);
} else {
GST_INFO ("Could not link pads: capsfilter:src - %s:%s",
GST_ELEMENT_NAME (dest), destpadname);
}
gst_element_set_state (capsfilter, GST_STATE_NULL);
/* this will unlink and unref as appropriate */
gst_bin_remove (GST_BIN (GST_OBJECT_PARENT (capsfilter)), capsfilter);
return FALSE;
}
} else {
if (gst_element_link_pads (src, srcpadname, dest, destpadname)) {
return TRUE;
} else {
GST_INFO ("Could not link pads: %s:%s - %s:%s", GST_ELEMENT_NAME (src),
srcpadname, GST_ELEMENT_NAME (dest), destpadname);
return FALSE;
}
}
}
/**
* gst_element_link:
* @src: (transfer none): a #GstElement containing the source pad.
* @dest: (transfer none): the #GstElement containing the destination pad.
*
* Links @src to @dest. The link must be from source to
* destination; the other direction will not be tried. The function looks for
* existing pads that aren't linked yet. It will request new pads if necessary.
* Such pads need to be released manually when unlinking.
* If multiple links are possible, only one is established.
*
* Make sure you have added your elements to a bin or pipeline with
* gst_bin_add() before trying to link them.
*
* Returns: %TRUE if the elements could be linked, %FALSE otherwise.
*/
gboolean
gst_element_link (GstElement * src, GstElement * dest)
{
return gst_element_link_pads (src, NULL, dest, NULL);
}
/**
* gst_element_link_many:
* @element_1: (transfer none): the first #GstElement in the link chain.
* @element_2: (transfer none): the second #GstElement in the link chain.
* @...: the %NULL-terminated list of elements to link in order.
*
* Chain together a series of elements. Uses gst_element_link().
* Make sure you have added your elements to a bin or pipeline with
* gst_bin_add() before trying to link them.
*
* Returns: %TRUE on success, %FALSE otherwise.
*/
gboolean
gst_element_link_many (GstElement * element_1, GstElement * element_2, ...)
{
gboolean res = TRUE;
va_list args;
g_return_val_if_fail (GST_IS_ELEMENT (element_1), FALSE);
g_return_val_if_fail (GST_IS_ELEMENT (element_2), FALSE);
va_start (args, element_2);
while (element_2) {
if (!gst_element_link (element_1, element_2)) {
res = FALSE;
break;
}
element_1 = element_2;
element_2 = va_arg (args, GstElement *);
}
va_end (args);
return res;
}
/**
* gst_element_link_filtered:
* @src: a #GstElement containing the source pad.
* @dest: (transfer none): the #GstElement containing the destination pad.
* @filter: (transfer none) (nullable): the #GstCaps to filter the link,
* or %NULL for no filter.
*
* Links @src to @dest using the given caps as filtercaps.
* The link must be from source to
* destination; the other direction will not be tried. The function looks for
* existing pads that aren't linked yet. It will request new pads if necessary.
* If multiple links are possible, only one is established.
*
* Make sure you have added your elements to a bin or pipeline with
* gst_bin_add() before trying to link them.
*
* Returns: %TRUE if the pads could be linked, %FALSE otherwise.
*/
gboolean
gst_element_link_filtered (GstElement * src, GstElement * dest,
GstCaps * filter)
{
return gst_element_link_pads_filtered (src, NULL, dest, NULL, filter);
}
/**
* gst_element_unlink_pads:
* @src: a (transfer none): #GstElement containing the source pad.
* @srcpadname: the name of the #GstPad in source element.
* @dest: (transfer none): a #GstElement containing the destination pad.
* @destpadname: the name of the #GstPad in destination element.
*
* Unlinks the two named pads of the source and destination elements.
*
* This is a convenience function for gst_pad_unlink().
*/
void
gst_element_unlink_pads (GstElement * src, const gchar * srcpadname,
GstElement * dest, const gchar * destpadname)
{
GstPad *srcpad, *destpad;
gboolean srcrequest, destrequest;
srcrequest = destrequest = FALSE;
g_return_if_fail (src != NULL);
g_return_if_fail (GST_IS_ELEMENT (src));
g_return_if_fail (srcpadname != NULL);
g_return_if_fail (dest != NULL);
g_return_if_fail (GST_IS_ELEMENT (dest));
g_return_if_fail (destpadname != NULL);
/* obtain the pads requested */
if (!(srcpad = gst_element_get_static_pad (src, srcpadname)))
if ((srcpad = gst_element_request_pad_simple (src, srcpadname)))
srcrequest = TRUE;
if (srcpad == NULL) {
GST_CAT_WARNING_OBJECT (GST_CAT_ELEMENT_PADS, src,
"source element has no pad \"%s\"", srcpadname);
return;
}
if (!(destpad = gst_element_get_static_pad (dest, destpadname)))
if ((destpad = gst_element_request_pad_simple (dest, destpadname)))
destrequest = TRUE;
if (destpad == NULL) {
GST_CAT_WARNING_OBJECT (GST_CAT_ELEMENT_PADS, dest,
"destination element has no pad \"%s\"", destpadname);
goto free_src;
}
/* we're satisfied they can be unlinked, let's do it */
gst_pad_unlink (srcpad, destpad);
if (destrequest)
gst_element_release_request_pad (dest, destpad);
gst_object_unref (destpad);
free_src:
if (srcrequest)
gst_element_release_request_pad (src, srcpad);
gst_object_unref (srcpad);
}
/**
* gst_element_unlink_many:
* @element_1: (transfer none): the first #GstElement in the link chain.
* @element_2: (transfer none): the second #GstElement in the link chain.
* @...: the %NULL-terminated list of elements to unlink in order.
*
* Unlinks a series of elements. Uses gst_element_unlink().
*/
void
gst_element_unlink_many (GstElement * element_1, GstElement * element_2, ...)
{
va_list args;
g_return_if_fail (element_1 != NULL && element_2 != NULL);
g_return_if_fail (GST_IS_ELEMENT (element_1) && GST_IS_ELEMENT (element_2));
va_start (args, element_2);
while (element_2) {
gst_element_unlink (element_1, element_2);
element_1 = element_2;
element_2 = va_arg (args, GstElement *);
}
va_end (args);
}
/**
* gst_element_unlink:
* @src: (transfer none): the source #GstElement to unlink.
* @dest: (transfer none): the sink #GstElement to unlink.
*
* Unlinks all source pads of the source element with all sink pads
* of the sink element to which they are linked.
*
* If the link has been made using gst_element_link(), it could have created an
* requestpad, which has to be released using gst_element_release_request_pad().
*/
void
gst_element_unlink (GstElement * src, GstElement * dest)
{
GstIterator *pads;
gboolean done = FALSE;
GValue data = { 0, };
g_return_if_fail (GST_IS_ELEMENT (src));
g_return_if_fail (GST_IS_ELEMENT (dest));
GST_CAT_DEBUG (GST_CAT_ELEMENT_PADS, "unlinking \"%s\" and \"%s\"",
GST_ELEMENT_NAME (src), GST_ELEMENT_NAME (dest));
pads = gst_element_iterate_pads (src);
while (!done) {
switch (gst_iterator_next (pads, &data)) {
case GST_ITERATOR_OK:
{
GstPad *pad = g_value_get_object (&data);
if (GST_PAD_IS_SRC (pad)) {
GstPad *peerpad = gst_pad_get_peer (pad);
/* see if the pad is linked and is really a pad of dest */
if (peerpad) {
GstElement *peerelem;
peerelem = gst_pad_get_parent_element (peerpad);
if (peerelem == dest) {
gst_pad_unlink (pad, peerpad);
}
if (peerelem)
gst_object_unref (peerelem);
gst_object_unref (peerpad);
}
}
g_value_reset (&data);
break;
}
case GST_ITERATOR_RESYNC:
gst_iterator_resync (pads);
break;
case GST_ITERATOR_DONE:
done = TRUE;
break;
default:
g_assert_not_reached ();
break;
}
}
g_value_unset (&data);
gst_iterator_free (pads);
}
/**
* gst_element_query_position:
* @element: a #GstElement to invoke the position query on.
* @format: the #GstFormat requested
* @cur: (out) (optional): a location in which to store the current
* position, or %NULL.
*
* Queries an element (usually top-level pipeline or playbin element) for the
* stream position in nanoseconds. This will be a value between 0 and the
* stream duration (if the stream duration is known). This query will usually
* only work once the pipeline is prerolled (i.e. reached PAUSED or PLAYING
* state). The application will receive an ASYNC_DONE message on the pipeline
* bus when that is the case.
*
* If one repeatedly calls this function one can also create a query and reuse
* it in gst_element_query().
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_element_query_position (GstElement * element, GstFormat format,
gint64 * cur)
{
GstQuery *query;
gboolean ret;
if (cur != NULL)
*cur = GST_CLOCK_TIME_NONE;
g_return_val_if_fail (GST_IS_ELEMENT (element), FALSE);
g_return_val_if_fail (format != GST_FORMAT_UNDEFINED, FALSE);
query = gst_query_new_position (format);
ret = gst_element_query (element, query);
if (ret)
gst_query_parse_position (query, NULL, cur);
gst_query_unref (query);
return ret;
}
/**
* gst_element_query_duration:
* @element: a #GstElement to invoke the duration query on.
* @format: the #GstFormat requested
* @duration: (out) (optional): A location in which to store the total duration, or %NULL.
*
* Queries an element (usually top-level pipeline or playbin element) for the
* total stream duration in nanoseconds. This query will only work once the
* pipeline is prerolled (i.e. reached PAUSED or PLAYING state). The application
* will receive an ASYNC_DONE message on the pipeline bus when that is the case.
*
* If the duration changes for some reason, you will get a DURATION_CHANGED
* message on the pipeline bus, in which case you should re-query the duration
* using this function.
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_element_query_duration (GstElement * element, GstFormat format,
gint64 * duration)
{
GstQuery *query;
gboolean ret;
if (duration != NULL)
*duration = GST_CLOCK_TIME_NONE;
g_return_val_if_fail (GST_IS_ELEMENT (element), FALSE);
g_return_val_if_fail (format != GST_FORMAT_UNDEFINED, FALSE);
query = gst_query_new_duration (format);
ret = gst_element_query (element, query);
if (ret)
gst_query_parse_duration (query, NULL, duration);
gst_query_unref (query);
return ret;
}
/**
* gst_element_query_convert:
* @element: a #GstElement to invoke the convert query on.
* @src_format: a #GstFormat to convert from.
* @src_val: a value to convert.
* @dest_format: the #GstFormat to convert to.
* @dest_val: (out): a pointer to the result.
*
* Queries an element to convert @src_val in @src_format to @dest_format.
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_element_query_convert (GstElement * element, GstFormat src_format,
gint64 src_val, GstFormat dest_format, gint64 * dest_val)
{
GstQuery *query;
gboolean ret;
g_return_val_if_fail (GST_IS_ELEMENT (element), FALSE);
g_return_val_if_fail (dest_format != GST_FORMAT_UNDEFINED, FALSE);
g_return_val_if_fail (dest_val != NULL, FALSE);
if (dest_format == src_format || src_val == -1) {
*dest_val = src_val;
return TRUE;
}
*dest_val = -1;
query = gst_query_new_convert (src_format, src_val, dest_format);
ret = gst_element_query (element, query);
if (ret)
gst_query_parse_convert (query, NULL, NULL, NULL, dest_val);
gst_query_unref (query);
return ret;
}
/**
* gst_element_seek_simple:
* @element: a #GstElement to seek on
* @format: a #GstFormat to execute the seek in, such as #GST_FORMAT_TIME
* @seek_flags: seek options; playback applications will usually want to use
* GST_SEEK_FLAG_FLUSH | GST_SEEK_FLAG_KEY_UNIT here
* @seek_pos: position to seek to (relative to the start); if you are doing
* a seek in #GST_FORMAT_TIME this value is in nanoseconds -
* multiply with #GST_SECOND to convert seconds to nanoseconds or
* with #GST_MSECOND to convert milliseconds to nanoseconds.
*
* Simple API to perform a seek on the given element, meaning it just seeks
* to the given position relative to the start of the stream. For more complex
* operations like segment seeks (e.g. for looping) or changing the playback
* rate or seeking relative to the last configured playback segment you should
* use gst_element_seek().
*
* In a completely prerolled PAUSED or PLAYING pipeline, seeking is always
* guaranteed to return %TRUE on a seekable media type or %FALSE when the media
* type is certainly not seekable (such as a live stream).
*
* Some elements allow for seeking in the READY state, in this
* case they will store the seek event and execute it when they are put to
* PAUSED. If the element supports seek in READY, it will always return %TRUE when
* it receives the event in the READY state.
*
* Returns: %TRUE if the seek operation succeeded. Flushing seeks will trigger a
* preroll, which will emit %GST_MESSAGE_ASYNC_DONE.
*/
gboolean
gst_element_seek_simple (GstElement * element, GstFormat format,
GstSeekFlags seek_flags, gint64 seek_pos)
{
g_return_val_if_fail (GST_IS_ELEMENT (element), FALSE);
g_return_val_if_fail (seek_pos >= 0, FALSE);
return gst_element_seek (element, 1.0, format, seek_flags,
GST_SEEK_TYPE_SET, seek_pos, GST_SEEK_TYPE_SET, GST_CLOCK_TIME_NONE);
}
/**
* gst_pad_use_fixed_caps:
* @pad: the pad to use
*
* A helper function you can use that sets the FIXED_CAPS flag
* This way the default CAPS query will always return the negotiated caps
* or in case the pad is not negotiated, the padtemplate caps.
*
* The negotiated caps are the caps of the last CAPS event that passed on the
* pad. Use this function on a pad that, once it negotiated to a CAPS, cannot
* be renegotiated to something else.
*/
void
gst_pad_use_fixed_caps (GstPad * pad)
{
GST_OBJECT_FLAG_SET (pad, GST_PAD_FLAG_FIXED_CAPS);
}
/**
* gst_pad_get_parent_element:
* @pad: a pad
*
* Gets the parent of @pad, cast to a #GstElement. If a @pad has no parent or
* its parent is not an element, return %NULL.
*
* Returns: (transfer full) (nullable): the parent of the pad. The
* caller has a reference on the parent, so unref when you're finished
* with it.
*
* MT safe.
*/
GstElement *
gst_pad_get_parent_element (GstPad * pad)
{
GstObject *p;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
p = gst_object_get_parent (GST_OBJECT_CAST (pad));
if (p && !GST_IS_ELEMENT (p)) {
gst_object_unref (p);
p = NULL;
}
return GST_ELEMENT_CAST (p);
}
/**
* gst_object_default_error:
* @source: the #GstObject that initiated the error.
* @error: (in): the GError.
* @debug: (in) (nullable): an additional debug information string, or %NULL
*
* A default error function that uses g_printerr() to display the error message
* and the optional debug string..
*
* The default handler will simply print the error string using g_print.
*/
void
gst_object_default_error (GstObject * source, const GError * error,
const gchar * debug)
{
gchar *name = gst_object_get_path_string (source);
g_printerr (_("ERROR: from element %s: %s\n"), name, error->message);
if (debug)
g_printerr (_("Additional debug info:\n%s\n"), debug);
g_free (name);
}
/**
* gst_bin_add_many: (skip)
* @bin: a #GstBin
* @element_1: (transfer floating): the #GstElement element to add to the bin
* @...: additional elements to add to the bin
*
* Adds a %NULL-terminated list of elements to a bin. This function is
* equivalent to calling gst_bin_add() for each member of the list. The return
* value of each gst_bin_add() is ignored.
*/
void
gst_bin_add_many (GstBin * bin, GstElement * element_1, ...)
{
va_list args;
g_return_if_fail (GST_IS_BIN (bin));
g_return_if_fail (GST_IS_ELEMENT (element_1));
va_start (args, element_1);
while (element_1) {
gst_bin_add (bin, element_1);
element_1 = va_arg (args, GstElement *);
}
va_end (args);
}
/**
* gst_bin_remove_many: (skip)
* @bin: a #GstBin
* @element_1: (transfer none): the first #GstElement to remove from the bin
* @...: (transfer none): %NULL-terminated list of elements to remove from the bin
*
* Removes a list of elements from a bin. This function is equivalent
* to calling gst_bin_remove() with each member of the list.
*/
void
gst_bin_remove_many (GstBin * bin, GstElement * element_1, ...)
{
va_list args;
g_return_if_fail (GST_IS_BIN (bin));
g_return_if_fail (GST_IS_ELEMENT (element_1));
va_start (args, element_1);
while (element_1) {
gst_bin_remove (bin, element_1);
element_1 = va_arg (args, GstElement *);
}
va_end (args);
}
typedef struct
{
GstQuery *query;
gboolean ret;
} QueryAcceptCapsData;
static gboolean
query_accept_caps_func (GstPad * pad, QueryAcceptCapsData * data)
{
if (G_LIKELY (gst_pad_peer_query (pad, data->query))) {
gboolean result;
gst_query_parse_accept_caps_result (data->query, &result);
data->ret &= result;
}
return FALSE;
}
/**
* gst_pad_proxy_query_accept_caps:
* @pad: a #GstPad to proxy.
* @query: an ACCEPT_CAPS #GstQuery.
*
* Checks if all internally linked pads of @pad accepts the caps in @query and
* returns the intersection of the results.
*
* This function is useful as a default accept caps query function for an element
* that can handle any stream format, but requires caps that are acceptable for
* all opposite pads.
*
* Returns: %TRUE if @query could be executed
*/
gboolean
gst_pad_proxy_query_accept_caps (GstPad * pad, GstQuery * query)
{
QueryAcceptCapsData data;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (GST_IS_QUERY (query), FALSE);
g_return_val_if_fail (GST_QUERY_TYPE (query) == GST_QUERY_ACCEPT_CAPS, FALSE);
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad,
"proxying accept caps query for %s:%s", GST_DEBUG_PAD_NAME (pad));
data.query = query;
/* value to hold the return, by default it holds TRUE */
/* FIXME: TRUE is wrong when there are no pads */
data.ret = TRUE;
gst_pad_forward (pad, (GstPadForwardFunction) query_accept_caps_func, &data);
gst_query_set_accept_caps_result (query, data.ret);
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad, "proxying accept caps query: %d",
data.ret);
return data.ret;
}
typedef struct
{
GstQuery *query;
GstCaps *ret;
} QueryCapsData;
static gboolean
query_caps_func (GstPad * pad, QueryCapsData * data)
{
gboolean empty = FALSE;
if (G_LIKELY (gst_pad_peer_query (pad, data->query))) {
GstCaps *peercaps, *intersection;
gst_query_parse_caps_result (data->query, &peercaps);
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad,
"intersect with result %" GST_PTR_FORMAT, peercaps);
intersection = gst_caps_intersect (data->ret, peercaps);
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad,
"intersected %" GST_PTR_FORMAT, intersection);
gst_caps_unref (data->ret);
data->ret = intersection;
/* stop when empty */
empty = gst_caps_is_empty (intersection);
}
return empty;
}
/**
* gst_pad_proxy_query_caps:
* @pad: a #GstPad to proxy.
* @query: a CAPS #GstQuery.
*
* Calls gst_pad_query_caps() for all internally linked pads of @pad and returns
* the intersection of the results.
*
* This function is useful as a default caps query function for an element
* that can handle any stream format, but requires all its pads to have
* the same caps. Two such elements are tee and adder.
*
* Returns: %TRUE if @query could be executed
*/
gboolean
gst_pad_proxy_query_caps (GstPad * pad, GstQuery * query)
{
GstCaps *filter, *templ, *result;
QueryCapsData data;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (GST_IS_QUERY (query), FALSE);
g_return_val_if_fail (GST_QUERY_TYPE (query) == GST_QUERY_CAPS, FALSE);
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad, "proxying caps query for %s:%s",
GST_DEBUG_PAD_NAME (pad));
data.query = query;
/* value to hold the return, by default it holds the filter or ANY */
gst_query_parse_caps (query, &filter);
data.ret = filter ? gst_caps_ref (filter) : gst_caps_new_any ();
gst_pad_forward (pad, (GstPadForwardFunction) query_caps_func, &data);
templ = gst_pad_get_pad_template_caps (pad);
result = gst_caps_intersect (data.ret, templ);
gst_caps_unref (data.ret);
gst_caps_unref (templ);
gst_query_set_caps_result (query, result);
gst_caps_unref (result);
/* FIXME: return something depending on the processing */
return TRUE;
}
/**
* gst_pad_query_position:
* @pad: a #GstPad to invoke the position query on.
* @format: the #GstFormat requested
* @cur: (out) (optional): A location in which to store the current position, or %NULL.
*
* Queries a pad for the stream position.
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_pad_query_position (GstPad * pad, GstFormat format, gint64 * cur)
{
GstQuery *query;
gboolean ret;
if (cur != NULL)
*cur = GST_CLOCK_TIME_NONE;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (format != GST_FORMAT_UNDEFINED, FALSE);
query = gst_query_new_position (format);
if ((ret = gst_pad_query (pad, query)))
gst_query_parse_position (query, NULL, cur);
gst_query_unref (query);
return ret;
}
/**
* gst_pad_peer_query_position:
* @pad: a #GstPad on whose peer to invoke the position query on.
* Must be a sink pad.
* @format: the #GstFormat requested
* @cur: (out) (optional): a location in which to store the current
* position, or %NULL.
*
* Queries the peer of a given sink pad for the stream position.
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_pad_peer_query_position (GstPad * pad, GstFormat format, gint64 * cur)
{
GstQuery *query;
gboolean ret = FALSE;
if (cur != NULL)
*cur = GST_CLOCK_TIME_NONE;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (format != GST_FORMAT_UNDEFINED, FALSE);
query = gst_query_new_position (format);
if ((ret = gst_pad_peer_query (pad, query)))
gst_query_parse_position (query, NULL, cur);
gst_query_unref (query);
return ret;
}
/**
* gst_pad_query_duration:
* @pad: a #GstPad to invoke the duration query on.
* @format: the #GstFormat requested
* @duration: (out) (optional): a location in which to store the total
* duration, or %NULL.
*
* Queries a pad for the total stream duration.
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_pad_query_duration (GstPad * pad, GstFormat format, gint64 * duration)
{
GstQuery *query;
gboolean ret;
if (duration != NULL)
*duration = GST_CLOCK_TIME_NONE;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (format != GST_FORMAT_UNDEFINED, FALSE);
query = gst_query_new_duration (format);
if ((ret = gst_pad_query (pad, query)))
gst_query_parse_duration (query, NULL, duration);
gst_query_unref (query);
return ret;
}
/**
* gst_pad_peer_query_duration:
* @pad: a #GstPad on whose peer pad to invoke the duration query on.
* Must be a sink pad.
* @format: the #GstFormat requested
* @duration: (out) (optional): a location in which to store the total
* duration, or %NULL.
*
* Queries the peer pad of a given sink pad for the total stream duration.
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_pad_peer_query_duration (GstPad * pad, GstFormat format, gint64 * duration)
{
GstQuery *query;
gboolean ret = FALSE;
if (duration != NULL)
*duration = GST_CLOCK_TIME_NONE;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (GST_PAD_IS_SINK (pad), FALSE);
g_return_val_if_fail (format != GST_FORMAT_UNDEFINED, FALSE);
query = gst_query_new_duration (format);
if ((ret = gst_pad_peer_query (pad, query)))
gst_query_parse_duration (query, NULL, duration);
gst_query_unref (query);
return ret;
}
/**
* gst_pad_query_convert:
* @pad: a #GstPad to invoke the convert query on.
* @src_format: a #GstFormat to convert from.
* @src_val: a value to convert.
* @dest_format: the #GstFormat to convert to.
* @dest_val: (out): a pointer to the result.
*
* Queries a pad to convert @src_val in @src_format to @dest_format.
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_pad_query_convert (GstPad * pad, GstFormat src_format, gint64 src_val,
GstFormat dest_format, gint64 * dest_val)
{
GstQuery *query;
gboolean ret;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (dest_format != GST_FORMAT_UNDEFINED, FALSE);
g_return_val_if_fail (dest_val != NULL, FALSE);
if (dest_format == src_format || src_val == -1) {
*dest_val = src_val;
return TRUE;
}
*dest_val = -1;
query = gst_query_new_convert (src_format, src_val, dest_format);
if ((ret = gst_pad_query (pad, query)))
gst_query_parse_convert (query, NULL, NULL, NULL, dest_val);
gst_query_unref (query);
return ret;
}
/**
* gst_pad_peer_query_convert:
* @pad: a #GstPad, on whose peer pad to invoke the convert query on.
* Must be a sink pad.
* @src_format: a #GstFormat to convert from.
* @src_val: a value to convert.
* @dest_format: the #GstFormat to convert to.
* @dest_val: (out): a pointer to the result.
*
* Queries the peer pad of a given sink pad to convert @src_val in @src_format
* to @dest_format.
*
* Returns: %TRUE if the query could be performed.
*/
gboolean
gst_pad_peer_query_convert (GstPad * pad, GstFormat src_format, gint64 src_val,
GstFormat dest_format, gint64 * dest_val)
{
GstQuery *query;
gboolean ret = FALSE;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (GST_PAD_IS_SINK (pad), FALSE);
g_return_val_if_fail (dest_format != GST_FORMAT_UNDEFINED, FALSE);
g_return_val_if_fail (dest_val != NULL, FALSE);
*dest_val = -1;
query = gst_query_new_convert (src_format, src_val, dest_format);
if ((ret = gst_pad_peer_query (pad, query)))
gst_query_parse_convert (query, NULL, NULL, NULL, dest_val);
gst_query_unref (query);
return ret;
}
/**
* gst_pad_query_caps:
* @pad: a #GstPad to get the capabilities of.
* @filter: (nullable): suggested #GstCaps, or %NULL
*
* Gets the capabilities this pad can produce or consume.
* Note that this method doesn't necessarily return the caps set by sending a
* gst_event_new_caps() - use gst_pad_get_current_caps() for that instead.
* gst_pad_query_caps returns all possible caps a pad can operate with, using
* the pad's CAPS query function, If the query fails, this function will return
* @filter, if not %NULL, otherwise ANY.
*
* When called on sinkpads @filter contains the caps that
* upstream could produce in the order preferred by upstream. When
* called on srcpads @filter contains the caps accepted by
* downstream in the preferred order. @filter might be %NULL but
* if it is not %NULL the returned caps will be a subset of @filter.
*
* Note that this function does not return writable #GstCaps, use
* gst_caps_make_writable() before modifying the caps.
*
* Returns: (transfer full): the caps of the pad with incremented ref-count.
*/
GstCaps *
gst_pad_query_caps (GstPad * pad, GstCaps * filter)
{
GstCaps *result = NULL;
GstQuery *query;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
g_return_val_if_fail (filter == NULL || GST_IS_CAPS (filter), NULL);
GST_CAT_DEBUG_OBJECT (GST_CAT_CAPS, pad,
"get pad caps with filter %" GST_PTR_FORMAT, filter);
query = gst_query_new_caps (filter);
if (gst_pad_query (pad, query)) {
gst_query_parse_caps_result (query, &result);
gst_caps_ref (result);
GST_CAT_DEBUG_OBJECT (GST_CAT_CAPS, pad,
"query returned %" GST_PTR_FORMAT, result);
} else if (filter) {
result = gst_caps_ref (filter);
} else {
result = gst_caps_new_any ();
}
gst_query_unref (query);
return result;
}
/**
* gst_pad_peer_query_caps:
* @pad: a #GstPad to get the capabilities of.
* @filter: (nullable): a #GstCaps filter, or %NULL.
*
* Gets the capabilities of the peer connected to this pad. Similar to
* gst_pad_query_caps().
*
* When called on srcpads @filter contains the caps that
* upstream could produce in the order preferred by upstream. When
* called on sinkpads @filter contains the caps accepted by
* downstream in the preferred order. @filter might be %NULL but
* if it is not %NULL the returned caps will be a subset of @filter.
*
* Returns: (transfer full): the caps of the peer pad with incremented
* ref-count. When there is no peer pad, this function returns @filter or,
* when @filter is %NULL, ANY caps.
*/
GstCaps *
gst_pad_peer_query_caps (GstPad * pad, GstCaps * filter)
{
GstCaps *result = NULL;
GstQuery *query;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
g_return_val_if_fail (filter == NULL || GST_IS_CAPS (filter), NULL);
GST_CAT_DEBUG_OBJECT (GST_CAT_CAPS, pad,
"get pad peer caps with filter %" GST_PTR_FORMAT, filter);
query = gst_query_new_caps (filter);
if (gst_pad_peer_query (pad, query)) {
gst_query_parse_caps_result (query, &result);
gst_caps_ref (result);
GST_CAT_DEBUG_OBJECT (GST_CAT_CAPS, pad,
"peer query returned %" GST_PTR_FORMAT, result);
} else if (filter) {
result = gst_caps_ref (filter);
} else {
result = gst_caps_new_any ();
}
gst_query_unref (query);
return result;
}
/**
* gst_pad_query_accept_caps:
* @pad: a #GstPad to check
* @caps: a #GstCaps to check on the pad
*
* Check if the given pad accepts the caps.
*
* Returns: %TRUE if the pad can accept the caps.
*/
gboolean
gst_pad_query_accept_caps (GstPad * pad, GstCaps * caps)
{
gboolean res = TRUE;
GstQuery *query;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (GST_IS_CAPS (caps), FALSE);
GST_CAT_DEBUG_OBJECT (GST_CAT_CAPS, pad, "accept caps of %"
GST_PTR_FORMAT, caps);
query = gst_query_new_accept_caps (caps);
if (gst_pad_query (pad, query)) {
gst_query_parse_accept_caps_result (query, &res);
GST_CAT_DEBUG_OBJECT (GST_CAT_CAPS, pad, "query returned %d", res);
}
gst_query_unref (query);
return res;
}
/**
* gst_pad_peer_query_accept_caps:
* @pad: a #GstPad to check the peer of
* @caps: a #GstCaps to check on the pad
*
* Check if the peer of @pad accepts @caps. If @pad has no peer, this function
* returns %TRUE.
*
* Returns: %TRUE if the peer of @pad can accept the caps or @pad has no peer.
*/
gboolean
gst_pad_peer_query_accept_caps (GstPad * pad, GstCaps * caps)
{
gboolean res = TRUE;
GstQuery *query;
g_return_val_if_fail (GST_IS_PAD (pad), FALSE);
g_return_val_if_fail (GST_IS_CAPS (caps), FALSE);
query = gst_query_new_accept_caps (caps);
if (gst_pad_peer_query (pad, query)) {
gst_query_parse_accept_caps_result (query, &res);
GST_CAT_DEBUG_OBJECT (GST_CAT_CAPS, pad, "query returned %d", res);
}
gst_query_unref (query);
return res;
}
static GstPad *
element_find_unlinked_pad (GstElement * element, GstPadDirection direction)
{
GstIterator *iter;
GstPad *unlinked_pad = NULL;
gboolean done;
GValue data = { 0, };
switch (direction) {
case GST_PAD_SRC:
iter = gst_element_iterate_src_pads (element);
break;
case GST_PAD_SINK:
iter = gst_element_iterate_sink_pads (element);
break;
default:
g_return_val_if_reached (NULL);
}
done = FALSE;
while (!done) {
switch (gst_iterator_next (iter, &data)) {
case GST_ITERATOR_OK:{
GstPad *peer;
GstPad *pad = g_value_get_object (&data);
GST_CAT_LOG_OBJECT (GST_CAT_ELEMENT_PADS, element,
"examining pad %s:%s", GST_DEBUG_PAD_NAME (pad));
peer = gst_pad_get_peer (pad);
if (peer == NULL) {
unlinked_pad = gst_object_ref (pad);
done = TRUE;
GST_CAT_DEBUG_OBJECT (GST_CAT_ELEMENT_PADS, element,
"found existing unlinked pad %s:%s",
GST_DEBUG_PAD_NAME (unlinked_pad));
} else {
gst_object_unref (peer);
}
g_value_reset (&data);
break;
}
case GST_ITERATOR_DONE:
done = TRUE;
break;
case GST_ITERATOR_RESYNC:
gst_iterator_resync (iter);
break;
case GST_ITERATOR_ERROR:
g_return_val_if_reached (NULL);
break;
}
}
g_value_unset (&data);
gst_iterator_free (iter);
return unlinked_pad;
}
/**
* gst_bin_find_unlinked_pad:
* @bin: bin in which to look for elements with unlinked pads
* @direction: whether to look for an unlinked source or sink pad
*
* Recursively looks for elements with an unlinked pad of the given
* direction within the specified bin and returns an unlinked pad
* if one is found, or %NULL otherwise. If a pad is found, the caller
* owns a reference to it and should use gst_object_unref() on the
* pad when it is not needed any longer.
*
* Returns: (transfer full) (nullable): unlinked pad of the given
* direction.
*/
GstPad *
gst_bin_find_unlinked_pad (GstBin * bin, GstPadDirection direction)
{
GstIterator *iter;
gboolean done;
GstPad *pad = NULL;
GValue data = { 0, };
g_return_val_if_fail (GST_IS_BIN (bin), NULL);
g_return_val_if_fail (direction != GST_PAD_UNKNOWN, NULL);
done = FALSE;
iter = gst_bin_iterate_recurse (bin);
while (!done) {
switch (gst_iterator_next (iter, &data)) {
case GST_ITERATOR_OK:{
GstElement *element = g_value_get_object (&data);
pad = element_find_unlinked_pad (element, direction);
if (pad != NULL)
done = TRUE;
g_value_reset (&data);
break;
}
case GST_ITERATOR_DONE:
done = TRUE;
break;
case GST_ITERATOR_RESYNC:
gst_iterator_resync (iter);
break;
case GST_ITERATOR_ERROR:
g_return_val_if_reached (NULL);
break;
}
}
g_value_unset (&data);
gst_iterator_free (iter);
return pad;
}
static void
gst_bin_sync_children_states_foreach (const GValue * value, gpointer user_data)
{
gboolean *success = user_data;
GstElement *element = g_value_get_object (value);
if (gst_element_is_locked_state (element)) {
*success = TRUE;
} else {
*success = *success && gst_element_sync_state_with_parent (element);
if (GST_IS_BIN (element))
*success = *success
&& gst_bin_sync_children_states (GST_BIN_CAST (element));
}
}
/**
* gst_bin_sync_children_states:
* @bin: a #GstBin
*
* Synchronizes the state of every child of @bin with the state
* of @bin. See also gst_element_sync_state_with_parent().
*
* Returns: %TRUE if syncing the state was successful for all children,
* otherwise %FALSE.
*
* Since: 1.6
*/
gboolean
gst_bin_sync_children_states (GstBin * bin)
{
GstIterator *it;
GstIteratorResult res = GST_ITERATOR_OK;
gboolean success = TRUE;
it = gst_bin_iterate_sorted (bin);
do {
if (res == GST_ITERATOR_RESYNC) {
success = TRUE;
gst_iterator_resync (it);
}
res =
gst_iterator_foreach (it, gst_bin_sync_children_states_foreach,
&success);
} while (res == GST_ITERATOR_RESYNC);
gst_iterator_free (it);
return success;
}
/**
* gst_parse_bin_from_description:
* @bin_description: command line describing the bin
* @ghost_unlinked_pads: whether to automatically create ghost pads
* for unlinked source or sink pads within the bin
* @err: where to store the error message in case of an error, or %NULL
*
* This is a convenience wrapper around gst_parse_launch() to create a
* #GstBin from a gst-launch-style pipeline description. See
* gst_parse_launch() and the gst-launch man page for details about the
* syntax. Ghost pads on the bin for unlinked source or sink pads
* within the bin can automatically be created (but only a maximum of
* one ghost pad for each direction will be created; if you expect
* multiple unlinked source pads or multiple unlinked sink pads
* and want them all ghosted, you will have to create the ghost pads
* yourself).
*
* Returns: (transfer floating) (type Gst.Bin): a
* newly-created bin, or %NULL if an error occurred.
*/
GstElement *
gst_parse_bin_from_description (const gchar * bin_description,
gboolean ghost_unlinked_pads, GError ** err)
{
return gst_parse_bin_from_description_full (bin_description,
ghost_unlinked_pads, NULL, GST_PARSE_FLAG_NONE, err);
}
/**
* gst_parse_bin_from_description_full:
* @bin_description: command line describing the bin
* @ghost_unlinked_pads: whether to automatically create ghost pads
* for unlinked source or sink pads within the bin
* @context: (transfer none) (nullable): a parse context allocated with
* gst_parse_context_new(), or %NULL
* @flags: parsing options, or #GST_PARSE_FLAG_NONE
* @err: where to store the error message in case of an error, or %NULL
*
* This is a convenience wrapper around gst_parse_launch() to create a
* #GstBin from a gst-launch-style pipeline description. See
* gst_parse_launch() and the gst-launch man page for details about the
* syntax. Ghost pads on the bin for unlinked source or sink pads
* within the bin can automatically be created (but only a maximum of
* one ghost pad for each direction will be created; if you expect
* multiple unlinked source pads or multiple unlinked sink pads
* and want them all ghosted, you will have to create the ghost pads
* yourself).
*
* Returns: (transfer floating) (type Gst.Element): a newly-created
* element, which is guaranteed to be a bin unless
* #GST_PARSE_FLAG_NO_SINGLE_ELEMENT_BINS was passed, or %NULL if an error
* occurred.
*/
GstElement *
gst_parse_bin_from_description_full (const gchar * bin_description,
gboolean ghost_unlinked_pads, GstParseContext * context,
GstParseFlags flags, GError ** err)
{
#ifndef GST_DISABLE_PARSE
GstPad *pad = NULL;
GstElement *element;
GstBin *bin;
gchar *desc;
g_return_val_if_fail (bin_description != NULL, NULL);
g_return_val_if_fail (err == NULL || *err == NULL, NULL);
GST_DEBUG ("Making bin from description '%s'", bin_description);
/* parse the pipeline to a bin */
if (flags & GST_PARSE_FLAG_NO_SINGLE_ELEMENT_BINS) {
element = gst_parse_launch_full (bin_description, context, flags, err);
} else {
desc = g_strdup_printf ("bin.( %s )", bin_description);
element = gst_parse_launch_full (desc, context, flags, err);
g_free (desc);
}
if (element == NULL || (err && *err != NULL)) {
if (element)
gst_object_unref (element);
return NULL;
}
if (GST_IS_BIN (element)) {
bin = GST_BIN (element);
} else {
return element;
}
/* find pads and ghost them if necessary */
if (ghost_unlinked_pads) {
if ((pad = gst_bin_find_unlinked_pad (bin, GST_PAD_SRC))) {
gst_element_add_pad (GST_ELEMENT (bin), gst_ghost_pad_new ("src", pad));
gst_object_unref (pad);
}
if ((pad = gst_bin_find_unlinked_pad (bin, GST_PAD_SINK))) {
gst_element_add_pad (GST_ELEMENT (bin), gst_ghost_pad_new ("sink", pad));
gst_object_unref (pad);
}
}
return GST_ELEMENT (bin);
#else
gchar *msg;
GST_WARNING ("Disabled API called");
msg = gst_error_get_message (GST_CORE_ERROR, GST_CORE_ERROR_DISABLED);
g_set_error (err, GST_CORE_ERROR, GST_CORE_ERROR_DISABLED, "%s", msg);
g_free (msg);
return NULL;
#endif
}
/**
* gst_util_get_timestamp:
*
* Get a timestamp as GstClockTime to be used for interval measurements.
* The timestamp should not be interpreted in any other way.
*
* Returns: the timestamp
*/
GstClockTime
gst_util_get_timestamp (void)
{
#if defined(G_OS_WIN32) && !defined(GST_STATIC_COMPILATION)
/* priv_gst_clock_init() is called in DllMain */
#else
priv_gst_clock_init ();
#endif
return priv_gst_get_monotonic_time ();
}
/**
* gst_util_array_binary_search:
* @array: the sorted input array
* @num_elements: number of elements in the array
* @element_size: size of every element in bytes
* @search_func: (scope call): function to compare two elements, @search_data will always be passed as second argument
* @mode: search mode that should be used
* @search_data: element that should be found
* @user_data: (closure): data to pass to @search_func
*
* Searches inside @array for @search_data by using the comparison function
* @search_func. @array must be sorted ascending.
*
* As @search_data is always passed as second argument to @search_func it's
* not required that @search_data has the same type as the array elements.
*
* The complexity of this search function is O(log (num_elements)).
*
* Returns: (transfer none) (nullable): The address of the found
* element or %NULL if nothing was found
*/
gpointer
gst_util_array_binary_search (gpointer array, guint num_elements,
gsize element_size, GCompareDataFunc search_func, GstSearchMode mode,
gconstpointer search_data, gpointer user_data)
{
glong left = 0, right = num_elements - 1, m;
gint ret;
guint8 *data = (guint8 *) array;
g_return_val_if_fail (array != NULL, NULL);
g_return_val_if_fail (element_size > 0, NULL);
g_return_val_if_fail (search_func != NULL, NULL);
/* 0. No elements => return NULL */
if (num_elements == 0)
return NULL;
/* 1. If search_data is before the 0th element return the 0th element */
ret = search_func (data, search_data, user_data);
if ((ret >= 0 && mode == GST_SEARCH_MODE_AFTER) || ret == 0)
return data;
else if (ret > 0)
return NULL;
/* 2. If search_data is after the last element return the last element */
ret =
search_func (data + (num_elements - 1) * element_size, search_data,
user_data);
if ((ret <= 0 && mode == GST_SEARCH_MODE_BEFORE) || ret == 0)
return data + (num_elements - 1) * element_size;
else if (ret < 0)
return NULL;
/* 3. else binary search */
while (TRUE) {
m = left + (right - left) / 2;
ret = search_func (data + m * element_size, search_data, user_data);
if (ret == 0) {
return data + m * element_size;
} else if (ret < 0) {
left = m + 1;
} else {
right = m - 1;
}
/* No exact match found */
if (right < left) {
if (mode == GST_SEARCH_MODE_EXACT) {
return NULL;
} else if (mode == GST_SEARCH_MODE_AFTER) {
if (ret < 0)
return (m < num_elements) ? data + (m + 1) * element_size : NULL;
else
return data + m * element_size;
} else {
if (ret < 0)
return data + m * element_size;
else
return (m > 0) ? data + (m - 1) * element_size : NULL;
}
}
}
}
/* Finds the greatest common divisor.
* Returns 1 if none other found.
* This is Euclid's algorithm. */
/**
* gst_util_greatest_common_divisor:
* @a: First value as #gint
* @b: Second value as #gint
*
* Calculates the greatest common divisor of @a
* and @b.
*
* Returns: Greatest common divisor of @a and @b
*/
gint
gst_util_greatest_common_divisor (gint a, gint b)
{
while (b != 0) {
int temp = a;
a = b;
b = temp % b;
}
return ABS (a);
}
/**
* gst_util_greatest_common_divisor_int64:
* @a: First value as #gint64
* @b: Second value as #gint64
*
* Calculates the greatest common divisor of @a
* and @b.
*
* Returns: Greatest common divisor of @a and @b
*/
gint64
gst_util_greatest_common_divisor_int64 (gint64 a, gint64 b)
{
while (b != 0) {
gint64 temp = a;
a = b;
b = temp % b;
}
return ABS (a);
}
/**
* gst_util_simplify_fraction:
* @numerator: First value as #gint
* @denominator: Second value as #gint
* @n_terms: non-significative terms (typical value: 8)
* @threshold: threshold (typical value: 333)
*
* Calculates the simpler representation of @numerator and @denominator and
* update both values with the resulting simplified fraction.
*
* Simplify a fraction using a simple continued fraction decomposition.
* The idea here is to convert fractions such as 333333/10000000 to 1/30
* using 32 bit arithmetic only. The algorithm is not perfect and relies
* upon two arbitrary parameters to remove non-significative terms from
* the simple continued fraction decomposition. Using 8 and 333 for
* @n_terms and @threshold respectively seems to give nice results.
*
* Since: 1.24
*/
void
gst_util_simplify_fraction (gint * numerator, gint * denominator,
guint n_terms, guint threshold)
{
guint *an;
guint x, y, r;
guint i, n;
an = g_malloc_n (n_terms, sizeof (*an));
if (an == NULL)
return;
/*
* Convert the fraction to a simple continued fraction. See
* https://en.wikipedia.org/wiki/Continued_fraction
* Stop if the current term is bigger than or equal to the given
* threshold.
*/
x = *numerator;
y = *denominator;
for (n = 0; n < n_terms && y != 0; ++n) {
an[n] = x / y;
if (an[n] >= threshold) {
if (n < 2)
n++;
break;
}
r = x - an[n] * y;
x = y;
y = r;
}
/* Expand the simple continued fraction back to an integer fraction. */
x = 0;
y = 1;
for (i = n; i > 0; --i) {
r = y;
y = an[i - 1] * y + x;
x = r;
}
*numerator = y;
*denominator = x;
g_free (an);
}
/**
* gst_util_fraction_to_double:
* @src_n: Fraction numerator as #gint
* @src_d: Fraction denominator #gint
* @dest: (out): pointer to a #gdouble for the result
*
* Transforms a fraction to a #gdouble.
*/
void
gst_util_fraction_to_double (gint src_n, gint src_d, gdouble * dest)
{
g_return_if_fail (dest != NULL);
g_return_if_fail (src_d != 0);
*dest = ((gdouble) src_n) / ((gdouble) src_d);
}
#define MAX_TERMS 30
#define MIN_DIVISOR 1.0e-10
#define MAX_ERROR 1.0e-20
/* use continued fractions to transform a double into a fraction,
* see http://mathforum.org/dr.math/faq/faq.fractions.html#decfrac.
* This algorithm takes care of overflows.
*/
/**
* gst_util_double_to_fraction:
* @src: #gdouble to transform
* @dest_n: (out): pointer to a #gint to hold the result numerator
* @dest_d: (out): pointer to a #gint to hold the result denominator
*
* Transforms a #gdouble to a fraction and simplifies
* the result.
*/
void
gst_util_double_to_fraction (gdouble src, gint * dest_n, gint * dest_d)
{
gdouble V, F; /* double being converted */
gint N, D; /* will contain the result */
gint A; /* current term in continued fraction */
gint64 N1, D1; /* numerator, denominator of last approx */
gint64 N2, D2; /* numerator, denominator of previous approx */
gint i;
gint gcd;
gboolean negative = FALSE;
g_return_if_fail (dest_n != NULL);
g_return_if_fail (dest_d != NULL);
/* initialize fraction being converted */
F = src;
if (F < 0.0) {
F = -F;
negative = TRUE;
}
V = F;
/* initialize fractions with 1/0, 0/1 */
N1 = 1;
D1 = 0;
N2 = 0;
D2 = 1;
N = 1;
D = 1;
for (i = 0; i < MAX_TERMS; i++) {
/* get next term */
A = (gint) F; /* no floor() needed, F is always >= 0 */
/* get new divisor */
F = F - A;
/* calculate new fraction in temp */
N2 = N1 * A + N2;
D2 = D1 * A + D2;
/* guard against overflow */
if (N2 > G_MAXINT || D2 > G_MAXINT) {
break;
}
N = N2;
D = D2;
/* save last two fractions */
N2 = N1;
D2 = D1;
N1 = N;
D1 = D;
/* quit if dividing by zero or close enough to target */
if (F < MIN_DIVISOR || fabs (V - ((gdouble) N) / D) < MAX_ERROR) {
break;
}
/* Take reciprocal */
F = 1 / F;
}
/* fix for overflow */
if (D == 0) {
N = G_MAXINT;
D = 1;
}
/* fix for negative */
if (negative)
N = -N;
/* simplify */
gcd = gst_util_greatest_common_divisor (N, D);
if (gcd) {
N /= gcd;
D /= gcd;
}
/* set results */
*dest_n = N;
*dest_d = D;
}
/**
* gst_util_fraction_multiply:
* @a_n: Numerator of first value
* @a_d: Denominator of first value
* @b_n: Numerator of second value
* @b_d: Denominator of second value
* @res_n: (out): Pointer to #gint to hold the result numerator
* @res_d: (out): Pointer to #gint to hold the result denominator
*
* Multiplies the fractions @a_n/@a_d and @b_n/@b_d and stores
* the result in @res_n and @res_d.
*
* Returns: %FALSE on overflow, %TRUE otherwise.
*/
gboolean
gst_util_fraction_multiply (gint a_n, gint a_d, gint b_n, gint b_d,
gint * res_n, gint * res_d)
{
gint gcd;
g_return_val_if_fail (res_n != NULL, FALSE);
g_return_val_if_fail (res_d != NULL, FALSE);
g_return_val_if_fail (a_d != 0, FALSE);
g_return_val_if_fail (b_d != 0, FALSE);
/* early out if either is 0, as its gcd would be 0 */
if (a_n == 0 || b_n == 0) {
*res_n = 0;
*res_d = 1;
return TRUE;
}
gcd = gst_util_greatest_common_divisor (a_n, a_d);
a_n /= gcd;
a_d /= gcd;
gcd = gst_util_greatest_common_divisor (b_n, b_d);
b_n /= gcd;
b_d /= gcd;
gcd = gst_util_greatest_common_divisor (a_n, b_d);
a_n /= gcd;
b_d /= gcd;
gcd = gst_util_greatest_common_divisor (a_d, b_n);
a_d /= gcd;
b_n /= gcd;
/* This would result in overflow */
if (a_n != 0 && G_MAXINT / ABS (a_n) < ABS (b_n))
return FALSE;
if (G_MAXINT / ABS (a_d) < ABS (b_d))
return FALSE;
*res_n = a_n * b_n;
*res_d = a_d * b_d;
gcd = gst_util_greatest_common_divisor (*res_n, *res_d);
*res_n /= gcd;
*res_d /= gcd;
return TRUE;
}
/**
* gst_util_fraction_add:
* @a_n: Numerator of first value
* @a_d: Denominator of first value
* @b_n: Numerator of second value
* @b_d: Denominator of second value
* @res_n: (out): Pointer to #gint to hold the result numerator
* @res_d: (out): Pointer to #gint to hold the result denominator
*
* Adds the fractions @a_n/@a_d and @b_n/@b_d and stores
* the result in @res_n and @res_d.
*
* Returns: %FALSE on overflow, %TRUE otherwise.
*/
gboolean
gst_util_fraction_add (gint a_n, gint a_d, gint b_n, gint b_d, gint * res_n,
gint * res_d)
{
gint gcd;
g_return_val_if_fail (res_n != NULL, FALSE);
g_return_val_if_fail (res_d != NULL, FALSE);
g_return_val_if_fail (a_d != 0, FALSE);
g_return_val_if_fail (b_d != 0, FALSE);
gcd = gst_util_greatest_common_divisor (a_n, a_d);
a_n /= gcd;
a_d /= gcd;
gcd = gst_util_greatest_common_divisor (b_n, b_d);
b_n /= gcd;
b_d /= gcd;
if (a_n == 0) {
*res_n = b_n;
*res_d = b_d;
return TRUE;
}
if (b_n == 0) {
*res_n = a_n;
*res_d = a_d;
return TRUE;
}
/* This would result in overflow */
if (G_MAXINT / ABS (a_n) < ABS (b_n))
return FALSE;
if (G_MAXINT / ABS (a_d) < ABS (b_d))
return FALSE;
*res_n = (a_n * b_d) + (a_d * b_n);
*res_d = a_d * b_d;
gcd = gst_util_greatest_common_divisor (*res_n, *res_d);
if (gcd) {
*res_n /= gcd;
*res_d /= gcd;
} else {
/* res_n == 0 */
*res_d = 1;
}
return TRUE;
}
/**
* gst_util_fraction_compare:
* @a_n: Numerator of first value
* @a_d: Denominator of first value
* @b_n: Numerator of second value
* @b_d: Denominator of second value
*
* Compares the fractions @a_n/@a_d and @b_n/@b_d and returns
* -1 if a < b, 0 if a = b and 1 if a > b.
*
* Returns: -1 if a < b; 0 if a = b; 1 if a > b.
*/
gint
gst_util_fraction_compare (gint a_n, gint a_d, gint b_n, gint b_d)
{
gint64 new_num_1;
gint64 new_num_2;
gint gcd;
g_return_val_if_fail (a_d != 0 && b_d != 0, 0);
/* Simplify */
gcd = gst_util_greatest_common_divisor (a_n, a_d);
a_n /= gcd;
a_d /= gcd;
gcd = gst_util_greatest_common_divisor (b_n, b_d);
b_n /= gcd;
b_d /= gcd;
/* fractions are reduced when set, so we can quickly see if they're equal */
if (a_n == b_n && a_d == b_d)
return 0;
/* extend to 64 bits */
new_num_1 = ((gint64) a_n) * b_d;
new_num_2 = ((gint64) b_n) * a_d;
if (new_num_1 < new_num_2)
return -1;
if (new_num_1 > new_num_2)
return 1;
/* Should not happen because a_d and b_d are not 0 */
g_return_val_if_reached (0);
}
static gchar *
gst_element_decorate_stream_id_internal (GstElement * element,
const gchar * stream_id)
{
GstEvent *upstream_event;
gchar *upstream_stream_id = NULL, *new_stream_id;
GstPad *sinkpad;
/* First try to get the upstream stream-start stream-id from the sinkpad.
* This will only work for non-source elements */
sinkpad = gst_element_get_static_pad (element, "sink");
if (sinkpad) {
upstream_event =
gst_pad_get_sticky_event (sinkpad, GST_EVENT_STREAM_START, 0);
if (upstream_event) {
const gchar *tmp;
gst_event_parse_stream_start (upstream_event, &tmp);
if (tmp)
upstream_stream_id = g_strdup (tmp);
gst_event_unref (upstream_event);
}
gst_object_unref (sinkpad);
}
/* The only case where we don't have an upstream stream-start event
* here is for source elements */
if (!upstream_stream_id) {
GstQuery *query;
gchar *uri = NULL;
/* Try to generate one from the URI query and
* if it fails take a random number instead */
query = gst_query_new_uri ();
if (gst_element_query (element, query)) {
gst_query_parse_uri (query, &uri);
}
if (uri) {
GChecksum *cs;
/* And then generate an SHA256 sum of the URI */
cs = g_checksum_new (G_CHECKSUM_SHA256);
g_checksum_update (cs, (const guchar *) uri, strlen (uri));
g_free (uri);
upstream_stream_id = g_strdup (g_checksum_get_string (cs));
g_checksum_free (cs);
} else {
/* Just get some random number if the URI query fails */
GST_FIXME_OBJECT (element, "Creating random stream-id, consider "
"implementing a deterministic way of creating a stream-id");
upstream_stream_id =
g_strdup_printf ("%08x%08x%08x%08x", g_random_int (), g_random_int (),
g_random_int (), g_random_int ());
}
gst_query_unref (query);
}
if (stream_id) {
new_stream_id = g_strconcat (upstream_stream_id, "/", stream_id, NULL);
} else {
new_stream_id = g_strdup (upstream_stream_id);
}
g_free (upstream_stream_id);
return new_stream_id;
}
/**
* gst_element_decorate_stream_id_printf_valist:
* @element: The #GstElement to create a stream-id for
* @format: (not nullable): The stream-id
* @var_args: parameters for the @format string
*
* Creates a stream-id for @element by combining the upstream information with
* the @format.
*
* This function generates an unique stream-id by getting the upstream
* stream-start event stream ID and appending @format to it. If the element
* has no sinkpad it will generate an upstream stream-id by doing an URI query
* on the element and in the worst case just uses a random number. Source
* elements that don't implement the URI handler interface should ideally
* generate a unique, deterministic stream-id manually instead.
*
* Since stream IDs are sorted alphabetically, any numbers in the stream ID
* should be printed with a fixed number of characters, preceded by 0's, such as
* by using the format \%03u instead of \%u.
*
* Returns: (transfer full): A stream-id for @element.
*
* Since: 1.24
*/
gchar *
gst_element_decorate_stream_id_printf_valist (GstElement * element,
const gchar * format, va_list var_args)
{
gchar *stream_id, *res;
g_return_val_if_fail (format, NULL);
stream_id = g_strdup_vprintf (format, var_args);
res = gst_element_decorate_stream_id_internal (element, stream_id);
g_free (stream_id);
return res;
}
/**
* gst_element_decorate_stream_id_printf:
* @element: The #GstElement to create a stream-id for
* @format: (not nullable): The stream-id
*
* Creates a stream-id for @element by combining the upstream information with
* the @format.
*
* This function generates an unique stream-id by getting the upstream
* stream-start event stream ID and appending the stream-id to it. If the element
* has no sinkpad it will generate an upstream stream-id by doing an URI query
* on the element and in the worst case just uses a random number. Source
* elements that don't implement the URI handler interface should ideally
* generate a unique, deterministic stream-id manually instead.
*
* Since stream IDs are sorted alphabetically, any numbers in the stream ID
* should be printed with a fixed number of characters, preceded by 0's, such as
* by using the format \%03u instead of \%u.
*
* Returns: (transfer full): A stream-id for @element.
*
* Since: 1.24
*/
gchar *
gst_element_decorate_stream_id_printf (GstElement * element,
const gchar * format, ...)
{
gchar *res;
va_list var_args;
g_return_val_if_fail (format, NULL);
va_start (var_args, format);
res =
gst_element_decorate_stream_id_printf_valist (element, format, var_args);
va_end (var_args);
return res;
}
/**
* gst_element_decorate_stream_id:
* @element: The #GstElement to create a stream-id for
* @stream_id: (not nullable): The stream-id
*
* Creates a stream-id for @element by combining the upstream information with
* the @stream_id.
*
* This function generates an unique stream-id by getting the upstream
* stream-start event stream ID and appending @stream_id to it. If the element
* has no sinkpad it will generate an upstream stream-id by doing an URI query
* on the element and in the worst case just uses a random number. Source
* elements that don't implement the URI handler interface should ideally
* generate a unique, deterministic stream-id manually instead.
*
* Since stream IDs are sorted alphabetically, any numbers in the stream ID
* should be printed with a fixed number of characters, preceded by 0's, such as
* by using the format \%03u instead of \%u.
*
* Returns: (transfer full): A stream-id for @element.
*
* Since: 1.24
*/
gchar *
gst_element_decorate_stream_id (GstElement * element, const gchar * stream_id)
{
g_return_val_if_fail (stream_id, NULL);
g_return_val_if_fail (GST_IS_ELEMENT (element), NULL);
return gst_element_decorate_stream_id_internal (element, stream_id);
}
static gchar *
gst_pad_create_stream_id_internal (GstPad * pad, GstElement * parent,
const gchar * stream_id)
{
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
g_return_val_if_fail (GST_PAD_IS_SRC (pad), NULL);
g_return_val_if_fail (GST_IS_ELEMENT (parent), NULL);
g_return_val_if_fail (parent->numsinkpads <= 1, NULL);
/* If the element has multiple source pads it must
* provide a stream-id for every source pad, otherwise
* all source pads will have the same and are not
* distinguishable */
g_return_val_if_fail (parent->numsrcpads <= 1 || stream_id, NULL);
return gst_element_decorate_stream_id_internal (parent, stream_id);
}
/**
* gst_pad_create_stream_id_printf_valist:
* @pad: A source #GstPad
* @parent: Parent #GstElement of @pad
* @stream_id: (nullable): The stream-id
* @var_args: parameters for the @stream_id format string
*
* Creates a stream-id for the source #GstPad @pad by combining the
* upstream information with the optional @stream_id of the stream
* of @pad. @pad must have a parent #GstElement and which must have zero
* or one sinkpad. @stream_id can only be %NULL if the parent element
* of @pad has only a single source pad.
*
* This function generates an unique stream-id by getting the upstream
* stream-start event stream ID and appending @stream_id to it. If the
* element has no sinkpad it will generate an upstream stream-id by
* doing an URI query on the element and in the worst case just uses
* a random number. Source elements that don't implement the URI
* handler interface should ideally generate a unique, deterministic
* stream-id manually instead.
*
* Returns: A stream-id for @pad. g_free() after usage.
*/
gchar *
gst_pad_create_stream_id_printf_valist (GstPad * pad, GstElement * parent,
const gchar * stream_id, va_list var_args)
{
gchar *expanded = NULL, *new_stream_id;
if (stream_id)
expanded = g_strdup_vprintf (stream_id, var_args);
new_stream_id = gst_pad_create_stream_id_internal (pad, parent, expanded);
g_free (expanded);
return new_stream_id;
}
/**
* gst_pad_create_stream_id_printf:
* @pad: A source #GstPad
* @parent: Parent #GstElement of @pad
* @stream_id: (nullable): The stream-id
* @...: parameters for the @stream_id format string
*
* Creates a stream-id for the source #GstPad @pad by combining the
* upstream information with the optional @stream_id of the stream
* of @pad. @pad must have a parent #GstElement and which must have zero
* or one sinkpad. @stream_id can only be %NULL if the parent element
* of @pad has only a single source pad.
*
* This function generates an unique stream-id by getting the upstream
* stream-start event stream ID and appending @stream_id to it. If the
* element has no sinkpad it will generate an upstream stream-id by
* doing an URI query on the element and in the worst case just uses
* a random number. Source elements that don't implement the URI
* handler interface should ideally generate a unique, deterministic
* stream-id manually instead.
*
* Returns: A stream-id for @pad. g_free() after usage.
*/
gchar *
gst_pad_create_stream_id_printf (GstPad * pad, GstElement * parent,
const gchar * stream_id, ...)
{
va_list var_args;
gchar *new_stream_id;
va_start (var_args, stream_id);
new_stream_id =
gst_pad_create_stream_id_printf_valist (pad, parent, stream_id, var_args);
va_end (var_args);
return new_stream_id;
}
/**
* gst_pad_create_stream_id:
* @pad: A source #GstPad
* @parent: Parent #GstElement of @pad
* @stream_id: (nullable): The stream-id
*
* Creates a stream-id for the source #GstPad @pad by combining the
* upstream information with the optional @stream_id of the stream
* of @pad. @pad must have a parent #GstElement and which must have zero
* or one sinkpad. @stream_id can only be %NULL if the parent element
* of @pad has only a single source pad.
*
* This function generates an unique stream-id by getting the upstream
* stream-start event stream ID and appending @stream_id to it. If the
* element has no sinkpad it will generate an upstream stream-id by
* doing an URI query on the element and in the worst case just uses
* a random number. Source elements that don't implement the URI
* handler interface should ideally generate a unique, deterministic
* stream-id manually instead.
*
* Since stream IDs are sorted alphabetically, any numbers in the
* stream ID should be printed with a fixed number of characters,
* preceded by 0's, such as by using the format \%03u instead of \%u.
*
* Returns: A stream-id for @pad. g_free() after usage.
*/
gchar *
gst_pad_create_stream_id (GstPad * pad, GstElement * parent,
const gchar * stream_id)
{
return gst_pad_create_stream_id_internal (pad, parent, stream_id);
}
/**
* gst_pad_get_stream_id:
* @pad: A source #GstPad
*
* Returns the current stream-id for the @pad, or %NULL if none has been
* set yet, i.e. the pad has not received a stream-start event yet.
*
* This is a convenience wrapper around gst_pad_get_sticky_event() and
* gst_event_parse_stream_start().
*
* The returned stream-id string should be treated as an opaque string, its
* contents should not be interpreted.
*
* Returns: (nullable): a newly-allocated copy of the stream-id for
* @pad, or %NULL. g_free() the returned string when no longer
* needed.
*
* Since: 1.2
*/
gchar *
gst_pad_get_stream_id (GstPad * pad)
{
const gchar *stream_id = NULL;
GstEvent *event;
gchar *ret = NULL;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
event = gst_pad_get_sticky_event (pad, GST_EVENT_STREAM_START, 0);
if (event != NULL) {
gst_event_parse_stream_start (event, &stream_id);
ret = g_strdup (stream_id);
gst_event_unref (event);
GST_CAT_LOG_OBJECT (GST_CAT_PADS, pad, "pad has stream-id '%s'", ret);
} else {
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad,
"pad has not received a stream-start event yet");
}
return ret;
}
/**
* gst_pad_get_stream:
* @pad: A source #GstPad
*
* Returns the current #GstStream for the @pad, or %NULL if none has been
* set yet, i.e. the pad has not received a stream-start event yet.
*
* This is a convenience wrapper around gst_pad_get_sticky_event() and
* gst_event_parse_stream().
*
* Returns: (nullable) (transfer full): the current #GstStream for @pad, or %NULL.
* unref the returned stream when no longer needed.
*
* Since: 1.10
*/
GstStream *
gst_pad_get_stream (GstPad * pad)
{
GstStream *stream = NULL;
GstEvent *event;
g_return_val_if_fail (GST_IS_PAD (pad), NULL);
event = gst_pad_get_sticky_event (pad, GST_EVENT_STREAM_START, 0);
if (event != NULL) {
gst_event_parse_stream (event, &stream);
gst_event_unref (event);
GST_CAT_LOG_OBJECT (GST_CAT_PADS, pad, "pad has stream object %p", stream);
} else {
GST_CAT_DEBUG_OBJECT (GST_CAT_PADS, pad,
"pad has not received a stream-start event yet");
}
return stream;
}
/**
* gst_util_group_id_next:
*
* Return a constantly incrementing group id.
*
* This function is used to generate a new group-id for the
* stream-start event.
*
* This function never returns %GST_GROUP_ID_INVALID (which is 0)
*
* Returns: A constantly incrementing unsigned integer, which might
* overflow back to 0 at some point.
*/
guint
gst_util_group_id_next (void)
{
static gint counter = 1;
gint ret = g_atomic_int_add (&counter, 1);
/* Make sure we don't return GST_GROUP_ID_INVALID */
if (G_UNLIKELY (ret == GST_GROUP_ID_INVALID))
ret = g_atomic_int_add (&counter, 1);
return ret;
}
/* Compute the number of bits needed at least to store `in` */
static guint
gst_bit_storage_uint64 (guint64 in)
{
#if defined(__GNUC__) && __GNUC__ >= 4
return in ? 64 - __builtin_clzll (in) : 1;
#else
/* integer log2(v) from:
<http://graphics.stanford.edu/~seander/bithacks.html#IntegerLog> */
const guint64 b[] =
{ 0x2, 0xC, 0xF0, 0xFF00, 0xFFFF0000, 0xFFFFFFFF00000000LL };
const guint64 S[] = { 1, 2, 4, 8, 16, 32 };
int i;
guint count = 0;
for (i = 5; i >= 0; i--) {
if (in & b[i]) {
in >>= S[i];
count |= S[i];
}
}
return count + 1; // + 1 to get the number of storage bits needed
#endif
}
/**
* gst_util_ceil_log2:
* @v: a #guint32 value.
*
* Return a max num of log2.
*
* Returns: a computed #guint val.
*
* Since: 1.24
*/
guint
gst_util_ceil_log2 (guint32 v)
{
/* Compute Ceil(Log2(v)) */
/* Derived from branchless code for integer log2(v) from:
<http://graphics.stanford.edu/~seander/bithacks.html#IntegerLog> */
guint r, shift;
v--;
r = (v > 0xFFFF) << 4;
v >>= r;
shift = (v > 0xFF) << 3;
v >>= shift;
r |= shift;
shift = (v > 0xF) << 2;
v >>= shift;
r |= shift;
shift = (v > 0x3) << 1;
v >>= shift;
r |= shift;
r |= (v >> 1);
return r + 1;
}
/**
* gst_calculate_linear_regression: (skip)
* @xy: Pairs of (x,y) values
* @temp: Temporary scratch space used by the function
* @n: number of (x,y) pairs
* @m_num: (out): numerator of calculated slope
* @m_denom: (out): denominator of calculated slope
* @b: (out): Offset at Y-axis
* @xbase: (out): Offset at X-axis
* @r_squared: (out): R-squared
*
* Calculates the linear regression of the values @xy and places the
* result in @m_num, @m_denom, @b and @xbase, representing the function
* y(x) = m_num/m_denom * (x - xbase) + b
* that has the least-square distance from all points @x and @y.
*
* @r_squared will contain the remaining error.
*
* If @temp is not %NULL, it will be used as temporary space for the function,
* in which case the function works without any allocation at all. If @temp is
* %NULL, an allocation will take place. @temp should have at least the same
* amount of memory allocated as @xy, i.e. 2*n*sizeof(GstClockTime).
*
* > This function assumes (x,y) values with reasonable large differences
* > between them. It will not calculate the exact results if the differences
* > between neighbouring values are too small due to not being able to
* > represent sub-integer values during the calculations.
*
* Returns: %TRUE if the linear regression was successfully calculated
*
* Since: 1.12
*/
/* http://mathworld.wolfram.com/LeastSquaresFitting.html
* with SLAVE_LOCK
*/
gboolean
gst_calculate_linear_regression (const GstClockTime * xy,
GstClockTime * temp, guint n,
GstClockTime * m_num, GstClockTime * m_denom,
GstClockTime * b, GstClockTime * xbase, gdouble * r_squared)
{
const GstClockTime *x, *y;
GstClockTime *newx, *newy;
GstClockTime xmin, ymin, xbar, ybar, xbar4, ybar4;
GstClockTime xmax, ymax;
GstClockTimeDiff sxx, sxy, syy;
gint i, j;
gint pshift = 0;
gint max_bits;
g_return_val_if_fail (xy != NULL, FALSE);
g_return_val_if_fail (m_num != NULL, FALSE);
g_return_val_if_fail (m_denom != NULL, FALSE);
g_return_val_if_fail (b != NULL, FALSE);
g_return_val_if_fail (xbase != NULL, FALSE);
g_return_val_if_fail (r_squared != NULL, FALSE);
x = xy;
y = xy + 1;
xbar = ybar = sxx = syy = sxy = 0;
xmin = ymin = G_MAXUINT64;
xmax = ymax = 0;
for (i = j = 0; i < n; i++, j += 2) {
xmin = MIN (xmin, x[j]);
ymin = MIN (ymin, y[j]);
xmax = MAX (xmax, x[j]);
ymax = MAX (ymax, y[j]);
}
if (temp == NULL) {
/* Allocate up to 1kb on the stack, otherwise heap */
newx = n > 64 ? g_new (GstClockTime, 2 * n) : g_newa (GstClockTime, 2 * n);
newy = newx + 1;
} else {
newx = temp;
newy = temp + 1;
}
/* strip off unnecessary bits of precision */
for (i = j = 0; i < n; i++, j += 2) {
newx[j] = x[j] - xmin;
newy[j] = y[j] - ymin;
}
#ifdef DEBUGGING_ENABLED
GST_CAT_DEBUG (GST_CAT_CLOCK, "reduced numbers:");
for (i = j = 0; i < n; i++, j += 2)
GST_CAT_DEBUG (GST_CAT_CLOCK,
" %" G_GUINT64_FORMAT " %" G_GUINT64_FORMAT, newx[j], newy[j]);
#endif
/* have to do this precisely otherwise the results are pretty much useless.
* should guarantee that none of these accumulators can overflow */
/* quantities on the order of 1e10 to 1e13 -> 30-35 bits;
* window size a max of 2^10, so
this addition could end up around 2^45 or so -- ample headroom */
for (i = j = 0; i < n; i++, j += 2) {
/* Just in case assumptions about headroom prove false, let's check */
if ((newx[j] > 0 && G_MAXUINT64 - xbar <= newx[j]) ||
(newy[j] > 0 && G_MAXUINT64 - ybar <= newy[j])) {
GST_CAT_WARNING (GST_CAT_CLOCK,
"Regression overflowed in clock slaving! xbar %"
G_GUINT64_FORMAT " newx[j] %" G_GUINT64_FORMAT " ybar %"
G_GUINT64_FORMAT " newy[j] %" G_GUINT64_FORMAT, xbar, newx[j], ybar,
newy[j]);
if (temp == NULL && n > 64)
g_free (newx);
return FALSE;
}
xbar += newx[j];
ybar += newy[j];
}
xbar /= n;
ybar /= n;
/* multiplying directly would give quantities on the order of 1e20-1e26 ->
* 60 bits to 70 bits times the window size that's 80 which is too much.
* Instead we (1) subtract off the xbar*ybar in the loop instead of after,
* to avoid accumulation; (2) shift off some estimated number of bits from
* each multiplicand to limit the expected ceiling. For strange
* distributions of input values, things can still overflow, in which
* case we drop precision and retry - at most a few times, in practice rarely
*/
/* Guess how many bits we might need for the usual distribution of input,
* with a fallback loop that drops precision if things go pear-shaped.
*
* Each calculation of tmp during the iteration is multiplying two numbers and
* then adding them together, or adding them together and then multiplying
* them. The second case is worse and means we need at least twice
* (multiplication) as many bits as the biggest number needs, plus another bit
* (addition). At most 63 bits (signed 64 bit integer) are available.
*
* That means that each number must require at most (63 - 1) / 2 bits = 31
* bits of storage.
*/
max_bits = gst_bit_storage_uint64 (MAX (1, MAX (xmax - xmin, ymax - ymin)));
if (max_bits > 31)
pshift = max_bits - 31;
i = 0;
do {
#ifdef DEBUGGING_ENABLED
GST_CAT_DEBUG (GST_CAT_CLOCK,
"Restarting regression with precision shift %u", pshift);
#endif
xbar4 = xbar >> pshift;
ybar4 = ybar >> pshift;
sxx = syy = sxy = 0;
for (i = j = 0; i < n; i++, j += 2) {
GstClockTime newx4, newy4;
GstClockTimeDiff tmp;
newx4 = newx[j] >> pshift;
newy4 = newy[j] >> pshift;
tmp = (newx4 + xbar4) * (newx4 - xbar4);
if (G_UNLIKELY (tmp > 0 && sxx > 0 && (G_MAXINT64 - sxx <= tmp))) {
do {
/* Drop some precision and restart */
pshift++;
sxx /= 4;
tmp /= 4;
} while (G_MAXINT64 - sxx <= tmp);
break;
} else if (G_UNLIKELY (tmp < 0 && sxx < 0 && (G_MININT64 - sxx >= tmp))) {
do {
/* Drop some precision and restart */
pshift++;
sxx /= 4;
tmp /= 4;
} while (G_MININT64 - sxx >= tmp);
break;
}
sxx += tmp;
tmp = newy4 * newy4 - ybar4 * ybar4;
if (G_UNLIKELY (tmp > 0 && syy > 0 && (G_MAXINT64 - syy <= tmp))) {
do {
pshift++;
syy /= 4;
tmp /= 4;
} while (G_MAXINT64 - syy <= tmp);
break;
} else if (G_UNLIKELY (tmp < 0 && syy < 0 && (G_MININT64 - syy >= tmp))) {
do {
pshift++;
syy /= 4;
tmp /= 4;
} while (G_MININT64 - syy >= tmp);
break;
}
syy += tmp;
tmp = newx4 * newy4 - xbar4 * ybar4;
if (G_UNLIKELY (tmp > 0 && sxy > 0 && (G_MAXINT64 - sxy <= tmp))) {
do {
pshift++;
sxy /= 4;
tmp /= 4;
} while (G_MAXINT64 - sxy <= tmp);
break;
} else if (G_UNLIKELY (tmp < 0 && sxy < 0 && (G_MININT64 - sxy >= tmp))) {
do {
pshift++;
sxy /= 4;
tmp /= 4;
} while (G_MININT64 - sxy >= tmp);
break;
}
sxy += tmp;
}
} while (i < n);
if (G_UNLIKELY (sxx == 0))
goto invalid;
*m_num = sxy;
*m_denom = sxx;
*b = (ymin + ybar) - gst_util_uint64_scale_round (xbar, *m_num, *m_denom);
/* Report base starting from the most recent observation */
*xbase = xmax;
*b += gst_util_uint64_scale_round (xmax - xmin, *m_num, *m_denom);
*r_squared = ((double) sxy * (double) sxy) / ((double) sxx * (double) syy);
#ifdef DEBUGGING_ENABLED
GST_CAT_DEBUG (GST_CAT_CLOCK, " m = %g", ((double) *m_num) / *m_denom);
GST_CAT_DEBUG (GST_CAT_CLOCK, " b = %" G_GUINT64_FORMAT, *b);
GST_CAT_DEBUG (GST_CAT_CLOCK, " xbase = %" G_GUINT64_FORMAT, *xbase);
GST_CAT_DEBUG (GST_CAT_CLOCK, " r2 = %g", *r_squared);
#endif
if (temp == NULL && n > 64)
g_free (newx);
return TRUE;
invalid:
{
GST_CAT_DEBUG (GST_CAT_CLOCK, "sxx == 0, regression failed");
if (temp == NULL && n > 64)
g_free (newx);
return FALSE;
}
}
/**
* gst_type_mark_as_plugin_api:
* @type: a GType
* @flags: a set of #GstPluginAPIFlags to further inform cache generation.
*
* Marks @type as plugin API. This should be called in `class_init` of
* elements that expose new types (i.e. enums, flags or internal GObjects) via
* properties, signals or pad templates.
*
* Types exposed by plugins are not automatically added to the documentation
* as they might originate from another library and should in that case be
* documented via that library instead.
*
* By marking a type as plugin API it will be included in the documentation of
* the plugin that defines it.
*
* Since: 1.18
*/
void
gst_type_mark_as_plugin_api (GType type, GstPluginAPIFlags flags)
{
g_type_set_qdata (type, GST_QUARK (PLUGIN_API), GINT_TO_POINTER (TRUE));
g_type_set_qdata (type, GST_QUARK (PLUGIN_API_FLAGS),
GINT_TO_POINTER (flags));
}
/**
* gst_type_is_plugin_api:
* @type: a GType
* @flags: (out) (optional): What #GstPluginAPIFlags the plugin was marked with
*
* Checks if @type is plugin API. See gst_type_mark_as_plugin_api() for
* details.
*
* Returns: %TRUE if @type is plugin API or %FALSE otherwise.
*
* Since: 1.18
*/
gboolean
gst_type_is_plugin_api (GType type, GstPluginAPIFlags * flags)
{
gboolean ret =
!!GPOINTER_TO_INT (g_type_get_qdata (type, GST_QUARK (PLUGIN_API)));
if (ret && flags) {
*flags =
GPOINTER_TO_INT (g_type_get_qdata (type, GST_QUARK (PLUGIN_API_FLAGS)));
}
return ret;
}
/**
* gst_util_filename_compare:
* @a: (type filename): a filename to compare with @b
* @b: (type filename): a filename to compare with @a
*
* Compares the given filenames using natural ordering.
*
* Since: 1.24
*/
gint
gst_util_filename_compare (const gchar * a, const gchar * b)
{
gchar *a_utf8, *b_utf8;
gchar *a1, *b1;
gint ret;
/* Filenames in GLib are only guaranteed to be UTF-8 on Windows */
a_utf8 = g_filename_to_utf8 (a, -1, NULL, NULL, NULL);
b_utf8 = g_filename_to_utf8 (b, -1, NULL, NULL, NULL);
if (a_utf8 == NULL || b_utf8 == NULL) {
return strcmp (a, b);
}
a1 = g_utf8_collate_key_for_filename (a_utf8, -1);
b1 = g_utf8_collate_key_for_filename (b_utf8, -1);
ret = strcmp (a1, b1);
g_free (a1);
g_free (b1);
g_free (a_utf8);
g_free (b_utf8);
return ret;
}
Reference in New Issue View Git Blame Copy Permalink