#![allow(unused_mut, unused_assignments, unused_variables, clippy::all)]
#[inline(always)]
fn clip(min: i32, max: i32, v: i32) -> i32 { v.clamp(min, max) }
#[inline(always)]
fn idx(o: isize, s: isize, k: isize) -> usize { (o + k * s) as usize }
fn dct4_internal(c: &mut [i32], o: isize, s: isize, min: i32, max: i32, tx64: bool) {
let in0: i32 = c[idx(o, s, 0)];
let in1: i32 = c[idx(o, s, 1)];
let mut t0: i32;
let mut t1: i32;
let mut t2: i32;
let mut t3: i32;
if tx64 {
t1 = (in0 * 181 + 128) >> 8;
t0 = t1;
t2 = (in1 * 1567 + 2048) >> 12;
t3 = (in1 * 3784 + 2048) >> 12;
} else {
let in2: i32 = c[idx(o, s, 2)];
let in3: i32 = c[idx(o, s, 3)];
t0 = ((in0 + in2) * 181 + 128) >> 8;
t1 = ((in0 - in2) * 181 + 128) >> 8;
t2 = ((in1 * 1567 - in3 * (3784 - 4096) + 2048) >> 12) - in3;
t3 = ((in1 * (3784 - 4096) + in3 * 1567 + 2048) >> 12) + in1;
}
c[idx(o, s, 0)] = clip(min, max, t0 + t3);
c[idx(o, s, 1)] = clip(min, max, t1 + t2);
c[idx(o, s, 2)] = clip(min, max, t1 - t2);
c[idx(o, s, 3)] = clip(min, max, t0 - t3);
}
fn dct8_internal(c: &mut [i32], o: isize, s: isize, min: i32, max: i32, tx64: bool) {
dct4_internal(c, o, s << 1, min, max, tx64);
let in1: i32 = c[idx(o, s, 1)];
let in3: i32 = c[idx(o, s, 3)];
let mut t4a: i32;
let mut t5a: i32;
let mut t6a: i32;
let mut t7a: i32;
if tx64 {
t4a = (in1 * 799 + 2048) >> 12;
t5a = (in3 * -2276 + 2048) >> 12;
t6a = (in3 * 3406 + 2048) >> 12;
t7a = (in1 * 4017 + 2048) >> 12;
} else {
let in5: i32 = c[idx(o, s, 5)];
let in7: i32 = c[idx(o, s, 7)];
t4a = ((in1 * 799 - in7 * (4017 - 4096) + 2048) >> 12) - in7;
t5a = (in5 * 1703 - in3 * 1138 + 1024) >> 11;
t6a = (in5 * 1138 + in3 * 1703 + 1024) >> 11;
t7a = ((in1 * (4017 - 4096) + in7 * 799 + 2048) >> 12) + in1;
}
let t4: i32 = clip(min, max, t4a + t5a);
t5a = clip(min, max, t4a - t5a);
let t7: i32 = clip(min, max, t7a + t6a);
t6a = clip(min, max, t7a - t6a);
let t5: i32 = ((t6a - t5a) * 181 + 128) >> 8;
let t6: i32 = ((t6a + t5a) * 181 + 128) >> 8;
let t0: i32 = c[idx(o, s, 0)];
let t1: i32 = c[idx(o, s, 2)];
let t2: i32 = c[idx(o, s, 4)];
let t3: i32 = c[idx(o, s, 6)];
c[idx(o, s, 0)] = clip(min, max, t0 + t7);
c[idx(o, s, 1)] = clip(min, max, t1 + t6);
c[idx(o, s, 2)] = clip(min, max, t2 + t5);
c[idx(o, s, 3)] = clip(min, max, t3 + t4);
c[idx(o, s, 4)] = clip(min, max, t3 - t4);
c[idx(o, s, 5)] = clip(min, max, t2 - t5);
c[idx(o, s, 6)] = clip(min, max, t1 - t6);
c[idx(o, s, 7)] = clip(min, max, t0 - t7);
}
fn dct16_internal(c: &mut [i32], o: isize, s: isize, min: i32, max: i32, tx64: bool) {
dct8_internal(c, o, s << 1, min, max, tx64);
let in1: i32 = c[idx(o, s, 1)];
let in3: i32 = c[idx(o, s, 3)];
let in5: i32 = c[idx(o, s, 5)];
let in7: i32 = c[idx(o, s, 7)];
let mut t8a: i32;
let mut t9a: i32;
let mut t10a: i32;
let mut t11a: i32;
let mut t12a: i32;
let mut t13a: i32;
let mut t14a: i32;
let mut t15a: i32;
if tx64 {
t8a = (in1 * 401 + 2048) >> 12;
t9a = (in7 * -2598 + 2048) >> 12;
t10a = (in5 * 1931 + 2048) >> 12;
t11a = (in3 * -1189 + 2048) >> 12;
t12a = (in3 * 3920 + 2048) >> 12;
t13a = (in5 * 3612 + 2048) >> 12;
t14a = (in7 * 3166 + 2048) >> 12;
t15a = (in1 * 4076 + 2048) >> 12;
} else {
let in9: i32 = c[idx(o, s, 9)];
let in11: i32 = c[idx(o, s, 11)];
let in13: i32 = c[idx(o, s, 13)];
let in15: i32 = c[idx(o, s, 15)];
t8a = ((in1 * 401 - in15 * (4076 - 4096) + 2048) >> 12) - in15;
t9a = (in9 * 1583 - in7 * 1299 + 1024) >> 11;
t10a = ((in5 * 1931 - in11 * (3612 - 4096) + 2048) >> 12) - in11;
t11a = ((in13 * (3920 - 4096) - in3 * 1189 + 2048) >> 12) + in13;
t12a = ((in13 * 1189 + in3 * (3920 - 4096) + 2048) >> 12) + in3;
t13a = ((in5 * (3612 - 4096) + in11 * 1931 + 2048) >> 12) + in5;
t14a = (in9 * 1299 + in7 * 1583 + 1024) >> 11;
t15a = ((in1 * (4076 - 4096) + in15 * 401 + 2048) >> 12) + in1;
}
let mut t8: i32 = clip(min, max, t8a + t9a);
let mut t9: i32 = clip(min, max, t8a - t9a);
let mut t10: i32 = clip(min, max, t11a - t10a);
let mut t11: i32 = clip(min, max, t11a + t10a);
let mut t12: i32 = clip(min, max, t12a + t13a);
let mut t13: i32 = clip(min, max, t12a - t13a);
let mut t14: i32 = clip(min, max, t15a - t14a);
let mut t15: i32 = clip(min, max, t15a + t14a);
t9a = (( t14 * 1567 - t9 * (3784 - 4096) + 2048) >> 12) - t9;
t14a = (( t14 * (3784 - 4096) + t9 * 1567 + 2048) >> 12) + t14;
t10a = ((-(t13 * (3784 - 4096) + t10 * 1567) + 2048) >> 12) - t13;
t13a = (( t13 * 1567 - t10 * (3784 - 4096) + 2048) >> 12) - t10;
t8a = clip(min, max, t8 + t11);
t9 = clip(min, max, t9a + t10a);
t10 = clip(min, max, t9a - t10a);
t11a = clip(min, max, t8 - t11);
t12a = clip(min, max, t15 - t12);
t13 = clip(min, max, t14a - t13a);
t14 = clip(min, max, t14a + t13a);
t15a = clip(min, max, t15 + t12);
t10a = ((t13 - t10) * 181 + 128) >> 8;
t13a = ((t13 + t10) * 181 + 128) >> 8;
t11 = ((t12a - t11a) * 181 + 128) >> 8;
t12 = ((t12a + t11a) * 181 + 128) >> 8;
let t0: i32 = c[idx(o, s, 0)];
let t1: i32 = c[idx(o, s, 2)];
let t2: i32 = c[idx(o, s, 4)];
let t3: i32 = c[idx(o, s, 6)];
let t4: i32 = c[idx(o, s, 8)];
let t5: i32 = c[idx(o, s, 10)];
let t6: i32 = c[idx(o, s, 12)];
let t7: i32 = c[idx(o, s, 14)];
c[idx(o, s, 0)] = clip(min, max, t0 + t15a);
c[idx(o, s, 1)] = clip(min, max, t1 + t14);
c[idx(o, s, 2)] = clip(min, max, t2 + t13a);
c[idx(o, s, 3)] = clip(min, max, t3 + t12);
c[idx(o, s, 4)] = clip(min, max, t4 + t11);
c[idx(o, s, 5)] = clip(min, max, t5 + t10a);
c[idx(o, s, 6)] = clip(min, max, t6 + t9);
c[idx(o, s, 7)] = clip(min, max, t7 + t8a);
c[idx(o, s, 8)] = clip(min, max, t7 - t8a);
c[idx(o, s, 9)] = clip(min, max, t6 - t9);
c[idx(o, s, 10)] = clip(min, max, t5 - t10a);
c[idx(o, s, 11)] = clip(min, max, t4 - t11);
c[idx(o, s, 12)] = clip(min, max, t3 - t12);
c[idx(o, s, 13)] = clip(min, max, t2 - t13a);
c[idx(o, s, 14)] = clip(min, max, t1 - t14);
c[idx(o, s, 15)] = clip(min, max, t0 - t15a);
}
fn dct32_internal(c: &mut [i32], o: isize, s: isize, min: i32, max: i32, tx64: bool) {
dct16_internal(c, o, s << 1, min, max, tx64);
let in1: i32 = c[idx(o, s, 1)];
let in3: i32 = c[idx(o, s, 3)];
let in5: i32 = c[idx(o, s, 5)];
let in7: i32 = c[idx(o, s, 7)];
let in9: i32 = c[idx(o, s, 9)];
let in11: i32 = c[idx(o, s, 11)];
let in13: i32 = c[idx(o, s, 13)];
let in15: i32 = c[idx(o, s, 15)];
let mut t16a: i32;
let mut t17a: i32;
let mut t18a: i32;
let mut t19a: i32;
let mut t20a: i32;
let mut t21a: i32;
let mut t22a: i32;
let mut t23a: i32;
let mut t24a: i32;
let mut t25a: i32;
let mut t26a: i32;
let mut t27a: i32;
let mut t28a: i32;
let mut t29a: i32;
let mut t30a: i32;
let mut t31a: i32;
if tx64 {
t16a = (in1 * 201 + 2048) >> 12;
t17a = (in15 * -2751 + 2048) >> 12;
t18a = (in9 * 1751 + 2048) >> 12;
t19a = (in7 * -1380 + 2048) >> 12;
t20a = (in5 * 995 + 2048) >> 12;
t21a = (in11 * -2106 + 2048) >> 12;
t22a = (in13 * 2440 + 2048) >> 12;
t23a = (in3 * -601 + 2048) >> 12;
t24a = (in3 * 4052 + 2048) >> 12;
t25a = (in13 * 3290 + 2048) >> 12;
t26a = (in11 * 3513 + 2048) >> 12;
t27a = (in5 * 3973 + 2048) >> 12;
t28a = (in7 * 3857 + 2048) >> 12;
t29a = (in9 * 3703 + 2048) >> 12;
t30a = (in15 * 3035 + 2048) >> 12;
t31a = (in1 * 4091 + 2048) >> 12;
} else {
let in17: i32 = c[idx(o, s, 17)];
let in19: i32 = c[idx(o, s, 19)];
let in21: i32 = c[idx(o, s, 21)];
let in23: i32 = c[idx(o, s, 23)];
let in25: i32 = c[idx(o, s, 25)];
let in27: i32 = c[idx(o, s, 27)];
let in29: i32 = c[idx(o, s, 29)];
let in31: i32 = c[idx(o, s, 31)];
t16a = ((in1 * 201 - in31 * (4091 - 4096) + 2048) >> 12) - in31;
t17a = ((in17 * (3035 - 4096) - in15 * 2751 + 2048) >> 12) + in17;
t18a = ((in9 * 1751 - in23 * (3703 - 4096) + 2048) >> 12) - in23;
t19a = ((in25 * (3857 - 4096) - in7 * 1380 + 2048) >> 12) + in25;
t20a = ((in5 * 995 - in27 * (3973 - 4096) + 2048) >> 12) - in27;
t21a = ((in21 * (3513 - 4096) - in11 * 2106 + 2048) >> 12) + in21;
t22a = (in13 * 1220 - in19 * 1645 + 1024) >> 11;
t23a = ((in29 * (4052 - 4096) - in3 * 601 + 2048) >> 12) + in29;
t24a = ((in29 * 601 + in3 * (4052 - 4096) + 2048) >> 12) + in3;
t25a = (in13 * 1645 + in19 * 1220 + 1024) >> 11;
t26a = ((in21 * 2106 + in11 * (3513 - 4096) + 2048) >> 12) + in11;
t27a = ((in5 * (3973 - 4096) + in27 * 995 + 2048) >> 12) + in5;
t28a = ((in25 * 1380 + in7 * (3857 - 4096) + 2048) >> 12) + in7;
t29a = ((in9 * (3703 - 4096) + in23 * 1751 + 2048) >> 12) + in9;
t30a = ((in17 * 2751 + in15 * (3035 - 4096) + 2048) >> 12) + in15;
t31a = ((in1 * (4091 - 4096) + in31 * 201 + 2048) >> 12) + in1;
}
let mut t16: i32 = clip(min, max, t16a + t17a);
let mut t17: i32 = clip(min, max, t16a - t17a);
let mut t18: i32 = clip(min, max, t19a - t18a);
let mut t19: i32 = clip(min, max, t19a + t18a);
let mut t20: i32 = clip(min, max, t20a + t21a);
let mut t21: i32 = clip(min, max, t20a - t21a);
let mut t22: i32 = clip(min, max, t23a - t22a);
let mut t23: i32 = clip(min, max, t23a + t22a);
let mut t24: i32 = clip(min, max, t24a + t25a);
let mut t25: i32 = clip(min, max, t24a - t25a);
let mut t26: i32 = clip(min, max, t27a - t26a);
let mut t27: i32 = clip(min, max, t27a + t26a);
let mut t28: i32 = clip(min, max, t28a + t29a);
let mut t29: i32 = clip(min, max, t28a - t29a);
let mut t30: i32 = clip(min, max, t31a - t30a);
let mut t31: i32 = clip(min, max, t31a + t30a);
t17a = (( t30 * 799 - t17 * (4017 - 4096) + 2048) >> 12) - t17;
t30a = (( t30 * (4017 - 4096) + t17 * 799 + 2048) >> 12) + t30;
t18a = ((-(t29 * (4017 - 4096) + t18 * 799) + 2048) >> 12) - t29;
t29a = (( t29 * 799 - t18 * (4017 - 4096) + 2048) >> 12) - t18;
t21a = ( t26 * 1703 - t21 * 1138 + 1024) >> 11;
t26a = ( t26 * 1138 + t21 * 1703 + 1024) >> 11;
t22a = (-(t25 * 1138 + t22 * 1703 ) + 1024) >> 11;
t25a = ( t25 * 1703 - t22 * 1138 + 1024) >> 11;
t16a = clip(min, max, t16 + t19);
t17 = clip(min, max, t17a + t18a);
t18 = clip(min, max, t17a - t18a);
t19a = clip(min, max, t16 - t19);
t20a = clip(min, max, t23 - t20);
t21 = clip(min, max, t22a - t21a);
t22 = clip(min, max, t22a + t21a);
t23a = clip(min, max, t23 + t20);
t24a = clip(min, max, t24 + t27);
t25 = clip(min, max, t25a + t26a);
t26 = clip(min, max, t25a - t26a);
t27a = clip(min, max, t24 - t27);
t28a = clip(min, max, t31 - t28);
t29 = clip(min, max, t30a - t29a);
t30 = clip(min, max, t30a + t29a);
t31a = clip(min, max, t31 + t28);
t18a = (( t29 * 1567 - t18 * (3784 - 4096) + 2048) >> 12) - t18;
t29a = (( t29 * (3784 - 4096) + t18 * 1567 + 2048) >> 12) + t29;
t19 = (( t28a * 1567 - t19a * (3784 - 4096) + 2048) >> 12) - t19a;
t28 = (( t28a * (3784 - 4096) + t19a * 1567 + 2048) >> 12) + t28a;
t20 = ((-(t27a * (3784 - 4096) + t20a * 1567) + 2048) >> 12) - t27a;
t27 = (( t27a * 1567 - t20a * (3784 - 4096) + 2048) >> 12) - t20a;
t21a = ((-(t26 * (3784 - 4096) + t21 * 1567) + 2048) >> 12) - t26;
t26a = (( t26 * 1567 - t21 * (3784 - 4096) + 2048) >> 12) - t21;
t16 = clip(min, max, t16a + t23a);
t17a = clip(min, max, t17 + t22);
t18 = clip(min, max, t18a + t21a);
t19a = clip(min, max, t19 + t20);
t20a = clip(min, max, t19 - t20);
t21 = clip(min, max, t18a - t21a);
t22a = clip(min, max, t17 - t22);
t23 = clip(min, max, t16a - t23a);
t24 = clip(min, max, t31a - t24a);
t25a = clip(min, max, t30 - t25);
t26 = clip(min, max, t29a - t26a);
t27a = clip(min, max, t28 - t27);
t28a = clip(min, max, t28 + t27);
t29 = clip(min, max, t29a + t26a);
t30a = clip(min, max, t30 + t25);
t31 = clip(min, max, t31a + t24a);
t20 = ((t27a - t20a) * 181 + 128) >> 8;
t27 = ((t27a + t20a) * 181 + 128) >> 8;
t21a = ((t26 - t21 ) * 181 + 128) >> 8;
t26a = ((t26 + t21 ) * 181 + 128) >> 8;
t22 = ((t25a - t22a) * 181 + 128) >> 8;
t25 = ((t25a + t22a) * 181 + 128) >> 8;
t23a = ((t24 - t23 ) * 181 + 128) >> 8;
t24a = ((t24 + t23 ) * 181 + 128) >> 8;
let t0: i32 = c[idx(o, s, 0)];
let t1: i32 = c[idx(o, s, 2)];
let t2: i32 = c[idx(o, s, 4)];
let t3: i32 = c[idx(o, s, 6)];
let t4: i32 = c[idx(o, s, 8)];
let t5: i32 = c[idx(o, s, 10)];
let t6: i32 = c[idx(o, s, 12)];
let t7: i32 = c[idx(o, s, 14)];
let t8: i32 = c[idx(o, s, 16)];
let t9: i32 = c[idx(o, s, 18)];
let t10: i32 = c[idx(o, s, 20)];
let t11: i32 = c[idx(o, s, 22)];
let t12: i32 = c[idx(o, s, 24)];
let t13: i32 = c[idx(o, s, 26)];
let t14: i32 = c[idx(o, s, 28)];
let t15: i32 = c[idx(o, s, 30)];
c[idx(o, s, 0)] = clip(min, max, t0 + t31);
c[idx(o, s, 1)] = clip(min, max, t1 + t30a);
c[idx(o, s, 2)] = clip(min, max, t2 + t29);
c[idx(o, s, 3)] = clip(min, max, t3 + t28a);
c[idx(o, s, 4)] = clip(min, max, t4 + t27);
c[idx(o, s, 5)] = clip(min, max, t5 + t26a);
c[idx(o, s, 6)] = clip(min, max, t6 + t25);
c[idx(o, s, 7)] = clip(min, max, t7 + t24a);
c[idx(o, s, 8)] = clip(min, max, t8 + t23a);
c[idx(o, s, 9)] = clip(min, max, t9 + t22);
c[idx(o, s, 10)] = clip(min, max, t10 + t21a);
c[idx(o, s, 11)] = clip(min, max, t11 + t20);
c[idx(o, s, 12)] = clip(min, max, t12 + t19a);
c[idx(o, s, 13)] = clip(min, max, t13 + t18);
c[idx(o, s, 14)] = clip(min, max, t14 + t17a);
c[idx(o, s, 15)] = clip(min, max, t15 + t16);
c[idx(o, s, 16)] = clip(min, max, t15 - t16);
c[idx(o, s, 17)] = clip(min, max, t14 - t17a);
c[idx(o, s, 18)] = clip(min, max, t13 - t18);
c[idx(o, s, 19)] = clip(min, max, t12 - t19a);
c[idx(o, s, 20)] = clip(min, max, t11 - t20);
c[idx(o, s, 21)] = clip(min, max, t10 - t21a);
c[idx(o, s, 22)] = clip(min, max, t9 - t22);
c[idx(o, s, 23)] = clip(min, max, t8 - t23a);
c[idx(o, s, 24)] = clip(min, max, t7 - t24a);
c[idx(o, s, 25)] = clip(min, max, t6 - t25);
c[idx(o, s, 26)] = clip(min, max, t5 - t26a);
c[idx(o, s, 27)] = clip(min, max, t4 - t27);
c[idx(o, s, 28)] = clip(min, max, t3 - t28a);
c[idx(o, s, 29)] = clip(min, max, t2 - t29);
c[idx(o, s, 30)] = clip(min, max, t1 - t30a);
c[idx(o, s, 31)] = clip(min, max, t0 - t31);
}
fn dct64(c: &mut [i32], o: isize, s: isize, min: i32, max: i32) {
dct32_internal(c, o, s << 1, min, max, true);
let in1: i32 = c[idx(o, s, 1)];
let in3: i32 = c[idx(o, s, 3)];
let in5: i32 = c[idx(o, s, 5)];
let in7: i32 = c[idx(o, s, 7)];
let in9: i32 = c[idx(o, s, 9)];
let in11: i32 = c[idx(o, s, 11)];
let in13: i32 = c[idx(o, s, 13)];
let in15: i32 = c[idx(o, s, 15)];
let in17: i32 = c[idx(o, s, 17)];
let in19: i32 = c[idx(o, s, 19)];
let in21: i32 = c[idx(o, s, 21)];
let in23: i32 = c[idx(o, s, 23)];
let in25: i32 = c[idx(o, s, 25)];
let in27: i32 = c[idx(o, s, 27)];
let in29: i32 = c[idx(o, s, 29)];
let in31: i32 = c[idx(o, s, 31)];
let mut t32a: i32 = (in1 * 101 + 2048) >> 12;
let mut t33a: i32 = (in31 * -2824 + 2048) >> 12;
let mut t34a: i32 = (in17 * 1660 + 2048) >> 12;
let mut t35a: i32 = (in15 * -1474 + 2048) >> 12;
let mut t36a: i32 = (in9 * 897 + 2048) >> 12;
let mut t37a: i32 = (in23 * -2191 + 2048) >> 12;
let mut t38a: i32 = (in25 * 2359 + 2048) >> 12;
let mut t39a: i32 = (in7 * -700 + 2048) >> 12;
let mut t40a: i32 = (in5 * 501 + 2048) >> 12;
let mut t41a: i32 = (in27 * -2520 + 2048) >> 12;
let mut t42a: i32 = (in21 * 2019 + 2048) >> 12;
let mut t43a: i32 = (in11 * -1092 + 2048) >> 12;
let mut t44a: i32 = (in13 * 1285 + 2048) >> 12;
let mut t45a: i32 = (in19 * -1842 + 2048) >> 12;
let mut t46a: i32 = (in29 * 2675 + 2048) >> 12;
let mut t47a: i32 = (in3 * -301 + 2048) >> 12;
let mut t48a: i32 = (in3 * 4085 + 2048) >> 12;
let mut t49a: i32 = (in29 * 3102 + 2048) >> 12;
let mut t50a: i32 = (in19 * 3659 + 2048) >> 12;
let mut t51a: i32 = (in13 * 3889 + 2048) >> 12;
let mut t52a: i32 = (in11 * 3948 + 2048) >> 12;
let mut t53a: i32 = (in21 * 3564 + 2048) >> 12;
let mut t54a: i32 = (in27 * 3229 + 2048) >> 12;
let mut t55a: i32 = (in5 * 4065 + 2048) >> 12;
let mut t56a: i32 = (in7 * 4036 + 2048) >> 12;
let mut t57a: i32 = (in25 * 3349 + 2048) >> 12;
let mut t58a: i32 = (in23 * 3461 + 2048) >> 12;
let mut t59a: i32 = (in9 * 3996 + 2048) >> 12;
let mut t60a: i32 = (in15 * 3822 + 2048) >> 12;
let mut t61a: i32 = (in17 * 3745 + 2048) >> 12;
let mut t62a: i32 = (in31 * 2967 + 2048) >> 12;
let mut t63a: i32 = (in1 * 4095 + 2048) >> 12;
let mut t32: i32 = clip(min, max, t32a + t33a);
let mut t33: i32 = clip(min, max, t32a - t33a);
let mut t34: i32 = clip(min, max, t35a - t34a);
let mut t35: i32 = clip(min, max, t35a + t34a);
let mut t36: i32 = clip(min, max, t36a + t37a);
let mut t37: i32 = clip(min, max, t36a - t37a);
let mut t38: i32 = clip(min, max, t39a - t38a);
let mut t39: i32 = clip(min, max, t39a + t38a);
let mut t40: i32 = clip(min, max, t40a + t41a);
let mut t41: i32 = clip(min, max, t40a - t41a);
let mut t42: i32 = clip(min, max, t43a - t42a);
let mut t43: i32 = clip(min, max, t43a + t42a);
let mut t44: i32 = clip(min, max, t44a + t45a);
let mut t45: i32 = clip(min, max, t44a - t45a);
let mut t46: i32 = clip(min, max, t47a - t46a);
let mut t47: i32 = clip(min, max, t47a + t46a);
let mut t48: i32 = clip(min, max, t48a + t49a);
let mut t49: i32 = clip(min, max, t48a - t49a);
let mut t50: i32 = clip(min, max, t51a - t50a);
let mut t51: i32 = clip(min, max, t51a + t50a);
let mut t52: i32 = clip(min, max, t52a + t53a);
let mut t53: i32 = clip(min, max, t52a - t53a);
let mut t54: i32 = clip(min, max, t55a - t54a);
let mut t55: i32 = clip(min, max, t55a + t54a);
let mut t56: i32 = clip(min, max, t56a + t57a);
let mut t57: i32 = clip(min, max, t56a - t57a);
let mut t58: i32 = clip(min, max, t59a - t58a);
let mut t59: i32 = clip(min, max, t59a + t58a);
let mut t60: i32 = clip(min, max, t60a + t61a);
let mut t61: i32 = clip(min, max, t60a - t61a);
let mut t62: i32 = clip(min, max, t63a - t62a);
let mut t63: i32 = clip(min, max, t63a + t62a);
t33a = ((t33 * (4096 - 4076) + t62 * 401 + 2048) >> 12) - t33;
t34a = ((t34 * -401 + t61 * (4096 - 4076) + 2048) >> 12) - t61;
t37a = (t37 * -1299 + t58 * 1583 + 1024) >> 11;
t38a = (t38 * -1583 + t57 * -1299 + 1024) >> 11;
t41a = ((t41 * (4096 - 3612) + t54 * 1931 + 2048) >> 12) - t41;
t42a = ((t42 * -1931 + t53 * (4096 - 3612) + 2048) >> 12) - t53;
t45a = ((t45 * -1189 + t50 * (3920 - 4096) + 2048) >> 12) + t50;
t46a = ((t46 * (4096 - 3920) + t49 * -1189 + 2048) >> 12) - t46;
t49a = ((t46 * -1189 + t49 * (3920 - 4096) + 2048) >> 12) + t49;
t50a = ((t45 * (3920 - 4096) + t50 * 1189 + 2048) >> 12) + t45;
t53a = ((t42 * (4096 - 3612) + t53 * 1931 + 2048) >> 12) - t42;
t54a = ((t41 * 1931 + t54 * (3612 - 4096) + 2048) >> 12) + t54;
t57a = (t38 * -1299 + t57 * 1583 + 1024) >> 11;
t58a = (t37 * 1583 + t58 * 1299 + 1024) >> 11;
t61a = ((t34 * (4096 - 4076) + t61 * 401 + 2048) >> 12) - t34;
t62a = ((t33 * 401 + t62 * (4076 - 4096) + 2048) >> 12) + t62;
t32a = clip(min, max, t32 + t35);
t33 = clip(min, max, t33a + t34a);
t34 = clip(min, max, t33a - t34a);
t35a = clip(min, max, t32 - t35);
t36a = clip(min, max, t39 - t36);
t37 = clip(min, max, t38a - t37a);
t38 = clip(min, max, t38a + t37a);
t39a = clip(min, max, t39 + t36);
t40a = clip(min, max, t40 + t43);
t41 = clip(min, max, t41a + t42a);
t42 = clip(min, max, t41a - t42a);
t43a = clip(min, max, t40 - t43);
t44a = clip(min, max, t47 - t44);
t45 = clip(min, max, t46a - t45a);
t46 = clip(min, max, t46a + t45a);
t47a = clip(min, max, t47 + t44);
t48a = clip(min, max, t48 + t51);
t49 = clip(min, max, t49a + t50a);
t50 = clip(min, max, t49a - t50a);
t51a = clip(min, max, t48 - t51);
t52a = clip(min, max, t55 - t52);
t53 = clip(min, max, t54a - t53a);
t54 = clip(min, max, t54a + t53a);
t55a = clip(min, max, t55 + t52);
t56a = clip(min, max, t56 + t59);
t57 = clip(min, max, t57a + t58a);
t58 = clip(min, max, t57a - t58a);
t59a = clip(min, max, t56 - t59);
t60a = clip(min, max, t63 - t60);
t61 = clip(min, max, t62a - t61a);
t62 = clip(min, max, t62a + t61a);
t63a = clip(min, max, t63 + t60);
t34a = ((t34 * (4096 - 4017) + t61 * 799 + 2048) >> 12) - t34;
t35 = ((t35a * (4096 - 4017) + t60a * 799 + 2048) >> 12) - t35a;
t36 = ((t36a * -799 + t59a * (4096 - 4017) + 2048) >> 12) - t59a;
t37a = ((t37 * -799 + t58 * (4096 - 4017) + 2048) >> 12) - t58;
t42a = (t42 * -1138 + t53 * 1703 + 1024) >> 11;
t43 = (t43a * -1138 + t52a * 1703 + 1024) >> 11;
t44 = (t44a * -1703 + t51a * -1138 + 1024) >> 11;
t45a = (t45 * -1703 + t50 * -1138 + 1024) >> 11;
t50a = (t45 * -1138 + t50 * 1703 + 1024) >> 11;
t51 = (t44a * -1138 + t51a * 1703 + 1024) >> 11;
t52 = (t43a * 1703 + t52a * 1138 + 1024) >> 11;
t53a = (t42 * 1703 + t53 * 1138 + 1024) >> 11;
t58a = ((t37 * (4096 - 4017) + t58 * 799 + 2048) >> 12) - t37;
t59 = ((t36a * (4096 - 4017) + t59a * 799 + 2048) >> 12) - t36a;
t60 = ((t35a * 799 + t60a * (4017 - 4096) + 2048) >> 12) + t60a;
t61a = ((t34 * 799 + t61 * (4017 - 4096) + 2048) >> 12) + t61;
t32 = clip(min, max, t32a + t39a);
t33a = clip(min, max, t33 + t38);
t34 = clip(min, max, t34a + t37a);
t35a = clip(min, max, t35 + t36);
t36a = clip(min, max, t35 - t36);
t37 = clip(min, max, t34a - t37a);
t38a = clip(min, max, t33 - t38);
t39 = clip(min, max, t32a - t39a);
t40 = clip(min, max, t47a - t40a);
t41a = clip(min, max, t46 - t41);
t42 = clip(min, max, t45a - t42a);
t43a = clip(min, max, t44 - t43);
t44a = clip(min, max, t44 + t43);
t45 = clip(min, max, t45a + t42a);
t46a = clip(min, max, t46 + t41);
t47 = clip(min, max, t47a + t40a);
t48 = clip(min, max, t48a + t55a);
t49a = clip(min, max, t49 + t54);
t50 = clip(min, max, t50a + t53a);
t51a = clip(min, max, t51 + t52);
t52a = clip(min, max, t51 - t52);
t53 = clip(min, max, t50a - t53a);
t54a = clip(min, max, t49 - t54);
t55 = clip(min, max, t48a - t55a);
t56 = clip(min, max, t63a - t56a);
t57a = clip(min, max, t62 - t57);
t58 = clip(min, max, t61a - t58a);
t59a = clip(min, max, t60 - t59);
t60a = clip(min, max, t60 + t59);
t61 = clip(min, max, t61a + t58a);
t62a = clip(min, max, t62 + t57);
t63 = clip(min, max, t63a + t56a);
t36 = ((t36a * (4096 - 3784) + t59a * 1567 + 2048) >> 12) - t36a;
t37a = ((t37 * (4096 - 3784) + t58 * 1567 + 2048) >> 12) - t37;
t38 = ((t38a * (4096 - 3784) + t57a * 1567 + 2048) >> 12) - t38a;
t39a = ((t39 * (4096 - 3784) + t56 * 1567 + 2048) >> 12) - t39;
t40a = ((t40 * -1567 + t55 * (4096 - 3784) + 2048) >> 12) - t55;
t41 = ((t41a * -1567 + t54a * (4096 - 3784) + 2048) >> 12) - t54a;
t42a = ((t42 * -1567 + t53 * (4096 - 3784) + 2048) >> 12) - t53;
t43 = ((t43a * -1567 + t52a * (4096 - 3784) + 2048) >> 12) - t52a;
t52 = ((t43a * (4096 - 3784) + t52a * 1567 + 2048) >> 12) - t43a;
t53a = ((t42 * (4096 - 3784) + t53 * 1567 + 2048) >> 12) - t42;
t54 = ((t41a * (4096 - 3784) + t54a * 1567 + 2048) >> 12) - t41a;
t55a = ((t40 * (4096 - 3784) + t55 * 1567 + 2048) >> 12) - t40;
t56a = ((t39 * 1567 + t56 * (3784 - 4096) + 2048) >> 12) + t56;
t57 = ((t38a * 1567 + t57a * (3784 - 4096) + 2048) >> 12) + t57a;
t58a = ((t37 * 1567 + t58 * (3784 - 4096) + 2048) >> 12) + t58;
t59 = ((t36a * 1567 + t59a * (3784 - 4096) + 2048) >> 12) + t59a;
t32a = clip(min, max, t32 + t47);
t33 = clip(min, max, t33a + t46a);
t34a = clip(min, max, t34 + t45);
t35 = clip(min, max, t35a + t44a);
t36a = clip(min, max, t36 + t43);
t37 = clip(min, max, t37a + t42a);
t38a = clip(min, max, t38 + t41);
t39 = clip(min, max, t39a + t40a);
t40 = clip(min, max, t39a - t40a);
t41a = clip(min, max, t38 - t41);
t42 = clip(min, max, t37a - t42a);
t43a = clip(min, max, t36 - t43);
t44 = clip(min, max, t35a - t44a);
t45a = clip(min, max, t34 - t45);
t46 = clip(min, max, t33a - t46a);
t47a = clip(min, max, t32 - t47);
t48a = clip(min, max, t63 - t48);
t49 = clip(min, max, t62a - t49a);
t50a = clip(min, max, t61 - t50);
t51 = clip(min, max, t60a - t51a);
t52a = clip(min, max, t59 - t52);
t53 = clip(min, max, t58a - t53a);
t54a = clip(min, max, t57 - t54);
t55 = clip(min, max, t56a - t55a);
t56 = clip(min, max, t56a + t55a);
t57a = clip(min, max, t57 + t54);
t58 = clip(min, max, t58a + t53a);
t59a = clip(min, max, t59 + t52);
t60 = clip(min, max, t60a + t51a);
t61a = clip(min, max, t61 + t50);
t62 = clip(min, max, t62a + t49a);
t63a = clip(min, max, t63 + t48);
t40a = ((t55 - t40 ) * 181 + 128) >> 8;
t41 = ((t54a - t41a) * 181 + 128) >> 8;
t42a = ((t53 - t42 ) * 181 + 128) >> 8;
t43 = ((t52a - t43a) * 181 + 128) >> 8;
t44a = ((t51 - t44 ) * 181 + 128) >> 8;
t45 = ((t50a - t45a) * 181 + 128) >> 8;
t46a = ((t49 - t46 ) * 181 + 128) >> 8;
t47 = ((t48a - t47a) * 181 + 128) >> 8;
t48 = ((t47a + t48a) * 181 + 128) >> 8;
t49a = ((t46 + t49 ) * 181 + 128) >> 8;
t50 = ((t45a + t50a) * 181 + 128) >> 8;
t51a = ((t44 + t51 ) * 181 + 128) >> 8;
t52 = ((t43a + t52a) * 181 + 128) >> 8;
t53a = ((t42 + t53 ) * 181 + 128) >> 8;
t54 = ((t41a + t54a) * 181 + 128) >> 8;
t55a = ((t40 + t55 ) * 181 + 128) >> 8;
let t0: i32 = c[idx(o, s, 0)];
let t1: i32 = c[idx(o, s, 2)];
let t2: i32 = c[idx(o, s, 4)];
let t3: i32 = c[idx(o, s, 6)];
let t4: i32 = c[idx(o, s, 8)];
let t5: i32 = c[idx(o, s, 10)];
let t6: i32 = c[idx(o, s, 12)];
let t7: i32 = c[idx(o, s, 14)];
let t8: i32 = c[idx(o, s, 16)];
let t9: i32 = c[idx(o, s, 18)];
let t10: i32 = c[idx(o, s, 20)];
let t11: i32 = c[idx(o, s, 22)];
let t12: i32 = c[idx(o, s, 24)];
let t13: i32 = c[idx(o, s, 26)];
let t14: i32 = c[idx(o, s, 28)];
let t15: i32 = c[idx(o, s, 30)];
let t16: i32 = c[idx(o, s, 32)];
let t17: i32 = c[idx(o, s, 34)];
let t18: i32 = c[idx(o, s, 36)];
let t19: i32 = c[idx(o, s, 38)];
let t20: i32 = c[idx(o, s, 40)];
let t21: i32 = c[idx(o, s, 42)];
let t22: i32 = c[idx(o, s, 44)];
let t23: i32 = c[idx(o, s, 46)];
let t24: i32 = c[idx(o, s, 48)];
let t25: i32 = c[idx(o, s, 50)];
let t26: i32 = c[idx(o, s, 52)];
let t27: i32 = c[idx(o, s, 54)];
let t28: i32 = c[idx(o, s, 56)];
let t29: i32 = c[idx(o, s, 58)];
let t30: i32 = c[idx(o, s, 60)];
let t31: i32 = c[idx(o, s, 62)];
c[idx(o, s, 0)] = clip(min, max, t0 + t63a);
c[idx(o, s, 1)] = clip(min, max, t1 + t62);
c[idx(o, s, 2)] = clip(min, max, t2 + t61a);
c[idx(o, s, 3)] = clip(min, max, t3 + t60);
c[idx(o, s, 4)] = clip(min, max, t4 + t59a);
c[idx(o, s, 5)] = clip(min, max, t5 + t58);
c[idx(o, s, 6)] = clip(min, max, t6 + t57a);
c[idx(o, s, 7)] = clip(min, max, t7 + t56);
c[idx(o, s, 8)] = clip(min, max, t8 + t55a);
c[idx(o, s, 9)] = clip(min, max, t9 + t54);
c[idx(o, s, 10)] = clip(min, max, t10 + t53a);
c[idx(o, s, 11)] = clip(min, max, t11 + t52);
c[idx(o, s, 12)] = clip(min, max, t12 + t51a);
c[idx(o, s, 13)] = clip(min, max, t13 + t50);
c[idx(o, s, 14)] = clip(min, max, t14 + t49a);
c[idx(o, s, 15)] = clip(min, max, t15 + t48);
c[idx(o, s, 16)] = clip(min, max, t16 + t47);
c[idx(o, s, 17)] = clip(min, max, t17 + t46a);
c[idx(o, s, 18)] = clip(min, max, t18 + t45);
c[idx(o, s, 19)] = clip(min, max, t19 + t44a);
c[idx(o, s, 20)] = clip(min, max, t20 + t43);
c[idx(o, s, 21)] = clip(min, max, t21 + t42a);
c[idx(o, s, 22)] = clip(min, max, t22 + t41);
c[idx(o, s, 23)] = clip(min, max, t23 + t40a);
c[idx(o, s, 24)] = clip(min, max, t24 + t39);
c[idx(o, s, 25)] = clip(min, max, t25 + t38a);
c[idx(o, s, 26)] = clip(min, max, t26 + t37);
c[idx(o, s, 27)] = clip(min, max, t27 + t36a);
c[idx(o, s, 28)] = clip(min, max, t28 + t35);
c[idx(o, s, 29)] = clip(min, max, t29 + t34a);
c[idx(o, s, 30)] = clip(min, max, t30 + t33);
c[idx(o, s, 31)] = clip(min, max, t31 + t32a);
c[idx(o, s, 32)] = clip(min, max, t31 - t32a);
c[idx(o, s, 33)] = clip(min, max, t30 - t33);
c[idx(o, s, 34)] = clip(min, max, t29 - t34a);
c[idx(o, s, 35)] = clip(min, max, t28 - t35);
c[idx(o, s, 36)] = clip(min, max, t27 - t36a);
c[idx(o, s, 37)] = clip(min, max, t26 - t37);
c[idx(o, s, 38)] = clip(min, max, t25 - t38a);
c[idx(o, s, 39)] = clip(min, max, t24 - t39);
c[idx(o, s, 40)] = clip(min, max, t23 - t40a);
c[idx(o, s, 41)] = clip(min, max, t22 - t41);
c[idx(o, s, 42)] = clip(min, max, t21 - t42a);
c[idx(o, s, 43)] = clip(min, max, t20 - t43);
c[idx(o, s, 44)] = clip(min, max, t19 - t44a);
c[idx(o, s, 45)] = clip(min, max, t18 - t45);
c[idx(o, s, 46)] = clip(min, max, t17 - t46a);
c[idx(o, s, 47)] = clip(min, max, t16 - t47);
c[idx(o, s, 48)] = clip(min, max, t15 - t48);
c[idx(o, s, 49)] = clip(min, max, t14 - t49a);
c[idx(o, s, 50)] = clip(min, max, t13 - t50);
c[idx(o, s, 51)] = clip(min, max, t12 - t51a);
c[idx(o, s, 52)] = clip(min, max, t11 - t52);
c[idx(o, s, 53)] = clip(min, max, t10 - t53a);
c[idx(o, s, 54)] = clip(min, max, t9 - t54);
c[idx(o, s, 55)] = clip(min, max, t8 - t55a);
c[idx(o, s, 56)] = clip(min, max, t7 - t56);
c[idx(o, s, 57)] = clip(min, max, t6 - t57a);
c[idx(o, s, 58)] = clip(min, max, t5 - t58);
c[idx(o, s, 59)] = clip(min, max, t4 - t59a);
c[idx(o, s, 60)] = clip(min, max, t3 - t60);
c[idx(o, s, 61)] = clip(min, max, t2 - t61a);
c[idx(o, s, 62)] = clip(min, max, t1 - t62);
c[idx(o, s, 63)] = clip(min, max, t0 - t63a);
}
fn adst4_internal(c: &mut [i32], io: isize, is: isize, min: i32, max: i32, oo: isize, os: isize) {
let in0: i32 = c[idx(io, is, 0)];
let in1: i32 = c[idx(io, is, 1)];
let in2: i32 = c[idx(io, is, 2)];
let in3: i32 = c[idx(io, is, 3)];
c[idx(oo, os, 0)] = (( 1321 * in0 + (3803 - 4096) * in2 + (2482 - 4096) * in3 + (3344 - 4096) * in1 + 2048) >> 12) + in2 + in3 + in1;
c[idx(oo, os, 1)] = (((2482 - 4096) * in0 - 1321 * in2 - (3803 - 4096) * in3 + (3344 - 4096) * in1 + 2048) >> 12) + in0 - in3 + in1;
c[idx(oo, os, 2)] = (209 * (in0 - in2 + in3) + 128) >> 8;
c[idx(oo, os, 3)] = (((3803 - 4096) * in0 + (2482 - 4096) * in2 - 1321 * in3 - (3344 - 4096) * in1 + 2048) >> 12) + in0 + in2 - in1;
}
fn adst8_internal(c: &mut [i32], io: isize, is: isize, min: i32, max: i32, oo: isize, os: isize) {
let in0: i32 = c[idx(io, is, 0)];
let in1: i32 = c[idx(io, is, 1)];
let in2: i32 = c[idx(io, is, 2)];
let in3: i32 = c[idx(io, is, 3)];
let in4: i32 = c[idx(io, is, 4)];
let in5: i32 = c[idx(io, is, 5)];
let in6: i32 = c[idx(io, is, 6)];
let in7: i32 = c[idx(io, is, 7)];
let t0a: i32 = (((4076 - 4096) * in7 + 401 * in0 + 2048) >> 12) + in7;
let t1a: i32 = (( 401 * in7 - (4076 - 4096) * in0 + 2048) >> 12) - in0;
let t2a: i32 = (((3612 - 4096) * in5 + 1931 * in2 + 2048) >> 12) + in5;
let t3a: i32 = (( 1931 * in5 - (3612 - 4096) * in2 + 2048) >> 12) - in2;
let mut t4a: i32 = ( 1299 * in3 + 1583 * in4 + 1024) >> 11;
let mut t5a: i32 = ( 1583 * in3 - 1299 * in4 + 1024) >> 11;
let mut t6a: i32 = (( 1189 * in1 + (3920 - 4096) * in6 + 2048) >> 12) + in6;
let mut t7a: i32 = (((3920 - 4096) * in1 - 1189 * in6 + 2048) >> 12) + in1;
let t0: i32 = clip(min, max, t0a + t4a);
let t1: i32 = clip(min, max, t1a + t5a);
let mut t2: i32 = clip(min, max, t2a + t6a);
let mut t3: i32 = clip(min, max, t3a + t7a);
let t4: i32 = clip(min, max, t0a - t4a);
let t5: i32 = clip(min, max, t1a - t5a);
let mut t6: i32 = clip(min, max, t2a - t6a);
let mut t7: i32 = clip(min, max, t3a - t7a);
t4a = (((3784 - 4096) * t4 + 1567 * t5 + 2048) >> 12) + t4;
t5a = (( 1567 * t4 - (3784 - 4096) * t5 + 2048) >> 12) - t5;
t6a = (((3784 - 4096) * t7 - 1567 * t6 + 2048) >> 12) + t7;
t7a = (( 1567 * t7 + (3784 - 4096) * t6 + 2048) >> 12) + t6;
c[idx(oo, os, 0)] = clip(min, max, t0 + t2 );
c[idx(oo, os, 7)] = -clip(min, max, t1 + t3 );
t2 = clip(min, max, t0 - t2 );
t3 = clip(min, max, t1 - t3 );
c[idx(oo, os, 1)] = -clip(min, max, t4a + t6a);
c[idx(oo, os, 6)] = clip(min, max, t5a + t7a);
t6 = clip(min, max, t4a - t6a);
t7 = clip(min, max, t5a - t7a);
c[idx(oo, os, 3)] = -(((t2 + t3) * 181 + 128) >> 8);
c[idx(oo, os, 4)] = ((t2 - t3) * 181 + 128) >> 8;
c[idx(oo, os, 2)] = ((t6 + t7) * 181 + 128) >> 8;
c[idx(oo, os, 5)] = -(((t6 - t7) * 181 + 128) >> 8);
}
fn adst16_internal(c: &mut [i32], io: isize, is: isize, min: i32, max: i32, oo: isize, os: isize) {
let in0: i32 = c[idx(io, is, 0)];
let in1: i32 = c[idx(io, is, 1)];
let in2: i32 = c[idx(io, is, 2)];
let in3: i32 = c[idx(io, is, 3)];
let in4: i32 = c[idx(io, is, 4)];
let in5: i32 = c[idx(io, is, 5)];
let in6: i32 = c[idx(io, is, 6)];
let in7: i32 = c[idx(io, is, 7)];
let in8: i32 = c[idx(io, is, 8)];
let in9: i32 = c[idx(io, is, 9)];
let in10: i32 = c[idx(io, is, 10)];
let in11: i32 = c[idx(io, is, 11)];
let in12: i32 = c[idx(io, is, 12)];
let in13: i32 = c[idx(io, is, 13)];
let in14: i32 = c[idx(io, is, 14)];
let in15: i32 = c[idx(io, is, 15)];
let mut t0: i32 = ((in15 * (4091 - 4096) + in0 * 201 + 2048) >> 12) + in15;
let mut t1: i32 = ((in15 * 201 - in0 * (4091 - 4096) + 2048) >> 12) - in0;
let mut t2: i32 = ((in13 * (3973 - 4096) + in2 * 995 + 2048) >> 12) + in13;
let mut t3: i32 = ((in13 * 995 - in2 * (3973 - 4096) + 2048) >> 12) - in2;
let mut t4: i32 = ((in11 * (3703 - 4096) + in4 * 1751 + 2048) >> 12) + in11;
let mut t5: i32 = ((in11 * 1751 - in4 * (3703 - 4096) + 2048) >> 12) - in4;
let mut t6: i32 = (in9 * 1645 + in6 * 1220 + 1024) >> 11;
let mut t7: i32 = (in9 * 1220 - in6 * 1645 + 1024) >> 11;
let mut t8: i32 = ((in7 * 2751 + in8 * (3035 - 4096) + 2048) >> 12) + in8;
let mut t9: i32 = ((in7 * (3035 - 4096) - in8 * 2751 + 2048) >> 12) + in7;
let mut t10: i32 = ((in5 * 2106 + in10 * (3513 - 4096) + 2048) >> 12) + in10;
let mut t11: i32 = ((in5 * (3513 - 4096) - in10 * 2106 + 2048) >> 12) + in5;
let mut t12: i32 = ((in3 * 1380 + in12 * (3857 - 4096) + 2048) >> 12) + in12;
let mut t13: i32 = ((in3 * (3857 - 4096) - in12 * 1380 + 2048) >> 12) + in3;
let mut t14: i32 = ((in1 * 601 + in14 * (4052 - 4096) + 2048) >> 12) + in14;
let mut t15: i32 = ((in1 * (4052 - 4096) - in14 * 601 + 2048) >> 12) + in1;
let mut t0a: i32 = clip(min, max, t0 + t8 );
let mut t1a: i32 = clip(min, max, t1 + t9 );
let mut t2a: i32 = clip(min, max, t2 + t10);
let mut t3a: i32 = clip(min, max, t3 + t11);
let mut t4a: i32 = clip(min, max, t4 + t12);
let mut t5a: i32 = clip(min, max, t5 + t13);
let mut t6a: i32 = clip(min, max, t6 + t14);
let mut t7a: i32 = clip(min, max, t7 + t15);
let mut t8a: i32 = clip(min, max, t0 - t8 );
let mut t9a: i32 = clip(min, max, t1 - t9 );
let mut t10a: i32 = clip(min, max, t2 - t10);
let mut t11a: i32 = clip(min, max, t3 - t11);
let mut t12a: i32 = clip(min, max, t4 - t12);
let mut t13a: i32 = clip(min, max, t5 - t13);
let mut t14a: i32 = clip(min, max, t6 - t14);
let mut t15a: i32 = clip(min, max, t7 - t15);
t8 = ((t8a * (4017 - 4096) + t9a * 799 + 2048) >> 12) + t8a;
t9 = ((t8a * 799 - t9a * (4017 - 4096) + 2048) >> 12) - t9a;
t10 = ((t10a * 2276 + t11a * (3406 - 4096) + 2048) >> 12) + t11a;
t11 = ((t10a * (3406 - 4096) - t11a * 2276 + 2048) >> 12) + t10a;
t12 = ((t13a * (4017 - 4096) - t12a * 799 + 2048) >> 12) + t13a;
t13 = ((t13a * 799 + t12a * (4017 - 4096) + 2048) >> 12) + t12a;
t14 = ((t15a * 2276 - t14a * (3406 - 4096) + 2048) >> 12) - t14a;
t15 = ((t15a * (3406 - 4096) + t14a * 2276 + 2048) >> 12) + t15a;
t0 = clip(min, max, t0a + t4a);
t1 = clip(min, max, t1a + t5a);
t2 = clip(min, max, t2a + t6a);
t3 = clip(min, max, t3a + t7a);
t4 = clip(min, max, t0a - t4a);
t5 = clip(min, max, t1a - t5a);
t6 = clip(min, max, t2a - t6a);
t7 = clip(min, max, t3a - t7a);
t8a = clip(min, max, t8 + t12);
t9a = clip(min, max, t9 + t13);
t10a = clip(min, max, t10 + t14);
t11a = clip(min, max, t11 + t15);
t12a = clip(min, max, t8 - t12);
t13a = clip(min, max, t9 - t13);
t14a = clip(min, max, t10 - t14);
t15a = clip(min, max, t11 - t15);
t4a = ((t4 * (3784 - 4096) + t5 * 1567 + 2048) >> 12) + t4;
t5a = ((t4 * 1567 - t5 * (3784 - 4096) + 2048) >> 12) - t5;
t6a = ((t7 * (3784 - 4096) - t6 * 1567 + 2048) >> 12) + t7;
t7a = ((t7 * 1567 + t6 * (3784 - 4096) + 2048) >> 12) + t6;
t12 = ((t12a * (3784 - 4096) + t13a * 1567 + 2048) >> 12) + t12a;
t13 = ((t12a * 1567 - t13a * (3784 - 4096) + 2048) >> 12) - t13a;
t14 = ((t15a * (3784 - 4096) - t14a * 1567 + 2048) >> 12) + t15a;
t15 = ((t15a * 1567 + t14a * (3784 - 4096) + 2048) >> 12) + t14a;
c[idx(oo, os, 0)] = clip(min, max, t0 + t2 );
c[idx(oo, os, 15)] = -clip(min, max, t1 + t3 );
t2a = clip(min, max, t0 - t2 );
t3a = clip(min, max, t1 - t3 );
c[idx(oo, os, 3)] = -clip(min, max, t4a + t6a );
c[idx(oo, os, 12)] = clip(min, max, t5a + t7a );
t6 = clip(min, max, t4a - t6a );
t7 = clip(min, max, t5a - t7a );
c[idx(oo, os, 1)] = -clip(min, max, t8a + t10a);
c[idx(oo, os, 14)] = clip(min, max, t9a + t11a);
t10 = clip(min, max, t8a - t10a);
t11 = clip(min, max, t9a - t11a);
c[idx(oo, os, 2)] = clip(min, max, t12 + t14 );
c[idx(oo, os, 13)] = -clip(min, max, t13 + t15 );
t14a = clip(min, max, t12 - t14 );
t15a = clip(min, max, t13 - t15 );
c[idx(oo, os, 7)] = -(((t2a + t3a) * 181 + 128) >> 8);
c[idx(oo, os, 8)] = ((t2a - t3a) * 181 + 128) >> 8;
c[idx(oo, os, 4)] = ((t6 + t7) * 181 + 128) >> 8;
c[idx(oo, os, 11)] = -(((t6 - t7) * 181 + 128) >> 8);
c[idx(oo, os, 6)] = ((t10 + t11) * 181 + 128) >> 8;
c[idx(oo, os, 9)] = -(((t10 - t11) * 181 + 128) >> 8);
c[idx(oo, os, 5)] = -(((t14a + t15a) * 181 + 128) >> 8);
c[idx(oo, os, 10)] = ((t14a - t15a) * 181 + 128) >> 8;
}
pub const DCT: u8 = 0;
pub const ADST: u8 = 1;
pub const FLIPADST: u8 = 2;
pub const IDENTITY: u8 = 3;
fn identity(c: &mut [i32], o: isize, s: isize, n: usize) {
for i in 0..n as isize {
let p = idx(o, s, i);
let v = c[p];
c[p] = match n {
4 => v + ((v * 1697 + 2048) >> 12),
8 => v * 2,
16 => 2 * v + ((v * 1697 + 1024) >> 11),
_ => v * 4,
};
}
}
pub fn tx1d(lsz: usize, typ: u8, c: &mut [i32], o: isize, s: isize, min: i32, max: i32) {
let n = 4usize << lsz;
match (lsz, typ) {
(0, DCT) => dct4_internal(c, o, s, min, max, false),
(1, DCT) => dct8_internal(c, o, s, min, max, false),
(2, DCT) => dct16_internal(c, o, s, min, max, false),
(3, DCT) => dct32_internal(c, o, s, min, max, false),
(4, DCT) => dct64(c, o, s, min, max),
(0, ADST) => adst4_internal(c, o, s, min, max, o, s),
(1, ADST) => adst8_internal(c, o, s, min, max, o, s),
(2, ADST) => adst16_internal(c, o, s, min, max, o, s),
(0, FLIPADST) => adst4_internal(c, o, s, min, max, o + 3 * s, -s),
(1, FLIPADST) => adst8_internal(c, o, s, min, max, o + 7 * s, -s),
(2, FLIPADST) => adst16_internal(c, o, s, min, max, o + 15 * s, -s),
(_, IDENTITY) => identity(c, o, s, n),
_ => unreachable!("1-D transform {typ} of size {n}"),
}
}
pub fn wht4(c: &mut [i32], o: isize, s: isize) {
let in0 = c[idx(o, s, 0)];
let in1 = c[idx(o, s, 1)];
let in2 = c[idx(o, s, 2)];
let in3 = c[idx(o, s, 3)];
let t0 = in0 + in1;
let t2 = in2 - in3;
let t4 = (t0 - t2) >> 1;
let t3 = t4 - in3;
let t1 = t4 - in1;
c[idx(o, s, 0)] = t0 - t3;
c[idx(o, s, 1)] = t3;
c[idx(o, s, 2)] = t1;
c[idx(o, s, 3)] = t2 + t1;
}
pub const TX1D_TYPES: [[u8; 2]; 16] = [[DCT, DCT], [ADST, DCT], [DCT, ADST], [ADST, ADST], [FLIPADST, DCT], [DCT, FLIPADST], [FLIPADST, FLIPADST], [ADST, FLIPADST], [FLIPADST, ADST], [IDENTITY, IDENTITY], [DCT, IDENTITY], [IDENTITY, DCT], [ADST, IDENTITY], [IDENTITY, ADST], [FLIPADST, IDENTITY], [IDENTITY, FLIPADST]];