aifc 0.7.0

Reader and writer for the AIFF and AIFF-C audio format
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256

use crate::cast;

/// Converts a 80-bit extended floating point value to a 64-bit floating point value.
pub fn f80_to_f64(bytes: &[u8]) -> f64 {
    assert_eq!(bytes.len(), 10);

    const POW_2_NEG16382: f64 = 0.0; // 2.powi(-16382) -> rounds to 0.0f64
    const POW_2_63: f64 = cast::u64_to_f64(1u64 << 63); // 2.powi(63)

    // see https://en.wikipedia.org/wiki/Extended_precision#x86_extended_precision_format

    let sign: f64 = if bytes[0] & 0x80 == 0 { 1.0 } else { -1.0 };
    let exponent = u16::from_be_bytes([ bytes[0], bytes[1] ]) & 0x7fff;
    let mut significand_buf = [0u8; 8];
    significand_buf.copy_from_slice(&bytes[2..10]);
    let significand = u64::from_be_bytes(significand_buf);

    let bit62 = significand >> 62 & 0x01;
    let bit63 = significand >> 63;

    let bits62 = significand & 0x3fffffff_ffffffff;
    let bits61 = significand & 0x1fffffff_ffffffff;

    // note: this may underflow, possibly losing accuracy
    let m = cast::u64_to_f64(significand) / POW_2_63;

    match (exponent, bit63, bit62) {
        (0x0000, 0, _) => {
            if bits62 == 0 { sign * 0.0 } // zero
            else { sign * m * POW_2_NEG16382 } // denormal
        },
        (0x0000, 1, _) => { // pseudo denormal
            sign * m * POW_2_NEG16382
        },

        (0x7fff, 0, 0) => {
            if bits61 == 0 { sign * f64::INFINITY } // pseudo-infinity
            else { f64::NAN } // pseudo Not a Number
        },
        (0x7fff, 0, 1) => { // pseudo Not a Number
            f64::NAN
        },
        (0x7fff, 1, 0) => {
            if bits61 == 0 { sign * f64::INFINITY } // infinity
            else { f64::NAN } // signalling Not a Number
        },
        (0x7fff, 1, 1) => { // floating-point indefinite or Quiet Not a Number
            f64::NAN
        },

        _ => { // unnormal or normalized value
            // powi() isn't available in no_std, so it isn't used:
            //sign * m * 2.0f64.powi(i32::from(exponent)-16383)

            if exponent > 16383+1024 { // overflows f64
                sign * f64::INFINITY
            } else if exponent < 16383-1022 { // underflows f64
                sign * 0.0
            } else if exponent >= 16383 { // positive exponent including zero exponent
                let f64exp = u64::from(exponent - 16383 + 1023);
                let pow = f64::from_bits(f64exp << 52);
                sign * m * pow
            } else { // negative exponent
                let f64exp = u64::from(16383 - exponent + 1023);
                let pow = f64::from_bits(f64exp << 52);
                sign * m / pow
            }
        }
    }
}

/// Converts a 64-bit floating point value to a 80-bit extended floating point value.
pub fn f64_to_f80(value: f64) -> [u8; 10] {
    if !value.is_finite() {
        if !value.is_nan() {
            if value.is_sign_positive() { // +inf
                return [127, 255, 0, 0, 0, 0, 0, 0, 0, 0];
            } else { // -inf
                return [255, 255, 0, 0, 0, 0, 0, 0, 0, 0];
            }
        } else { // nan
            return [127, 255, 255, 255, 255, 255, 255, 255, 255, 255];
        }
    }

    if value == 0.0 {
        if value.is_sign_positive() { // positive zero
            return [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ];
        } else { // negative zero
            return [ 128, 0, 0, 0, 0, 0, 0, 0, 0, 0 ];
        }
    }
    // shift f80 bits to match f64
    let bits = value.to_bits();
    let sign = bits & 0x80000000_00000000;
    let exponent = ((bits & 0x7ff00000_00000000) >> 52) + 16383 - 1023;
    let fraction = bits & 0x000fffff_ffffffff;
    [
        cast::u64_to_u8(sign >> 56 | (exponent >> 8)),
        cast::u64_to_u8(exponent & 0xff),
        // bit 63 is always 1 to match f64
        cast::u64_to_u8(0x80 | ((fraction & 0x000fe000_00000000) >> 45)),
        cast::u64_to_u8((fraction & 0x00001fe0_00000000) >> 37),
        cast::u64_to_u8((fraction & 0x0000001f_e0000000) >> 29),
        cast::u64_to_u8((fraction & 0x00000000_1fe00000) >> 21),

        cast::u64_to_u8((fraction & 0x00000000_001fe000) >> 13),
        cast::u64_to_u8((fraction & 0x00000000_00001fe0) >> 5),
        cast::u64_to_u8((fraction & 0x00000000_0000001f) << 3),
        0
    ]
}

#[cfg(test)]
mod tests {
    use super::*;

    /// Compares two 64-bit floating points for equality, checking the signs of signed zeros
    /// and making two NANs match.
    macro_rules! assert_eq_f64 {
        ($left:expr, $right:expr $(,)?) => {
            let l: f64 = $left;
            let r: f64 = $right;
            if (r.is_nan()) { assert!(l.is_nan()); }
            else if (l == 0.0 && r == 0.0) { assert_eq!(l.signum(), r.signum()); }
            else { assert_eq!($left, $right); }
        }
    }

    /// Casts u16 to u8, keeping only the lowest 8 bits.
    const fn u16_to_u8(value: u16) -> u8 {
        value as u8
    }

    /// Creates f80 [u8; 10] from the given parameters.
    fn f80(out: &mut [u8; 10], sign: i8, exponent: u16, bit63: u8, bit62: u8,
            significand: u64) -> &[u8; 10] {
        out[0] = (cast::i8_to_u8(sign) & 0x80) | u16_to_u8(exponent >> 8 & 0x7f);
        out[1] = u16_to_u8(exponent & 0xff);
        let sbytes = significand.to_be_bytes();
        out[2] = sbytes[0] | (bit63 << 7) | (bit62 << 6);
        out[3] = sbytes[1];
        out[4] = sbytes[2];
        out[5] = sbytes[3];
        out[6] = sbytes[4];
        out[7] = sbytes[5];
        out[8] = sbytes[6];
        out[9] = sbytes[7];
        out
    }

    #[test]
    fn test_f80_to_f64() {
        let mut f80buf: [u8; 10] = [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ];

        // exponent all zeros

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, 1, 0x0000,  0, 0,  0)), 0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x0000,  0, 0,  0)), -0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 0x0000,  0, 1,  1000000)), 0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 0x0000,  0, 1,  0x7fffffff_ffffffff)), 0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x0000,  1, 0,  1234)), -0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x0000,  1, 1,  0xffffffff_ffffffff)), -0.0);

        // exponent all ones: infinities and NANs

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 0x7fff,  0, 0,  0)), f64::INFINITY);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x7fff,  0, 0,  0)), -f64::INFINITY);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 0x7fff,  0, 0,  2)), f64::NAN);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x7fff,  0, 0,  3)), f64::NAN);

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 0x7fff,  0, 1,  0)), f64::NAN);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x7fff,  0, 1,  6)), f64::NAN);

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 0x7fff,  1, 0,  0)), f64::INFINITY);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x7fff,  1, 0,  0)), -f64::INFINITY);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 0x7fff,  1, 0,  2)), f64::NAN);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x7fff,  1, 0,  6)), f64::NAN);

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 0x7fff,  1, 1,  0)), f64::NAN);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 0x7fff,  1, 1,  6)), f64::NAN);

        // all other values

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16383,  0, 0,  0x00000000_00000000)), 0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16383,  0, 0,  0x80000000_00000000)), 1.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 16383,  0, 0,  0x80000000_00000000)), -1.0);

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16384,  0, 0,  0x00000000_00000000)), 0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16384,  0, 0,  0x80000000_00000000)), 2.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 16384,  0, 0,  0x80000000_00000000)), -2.0);

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16382,  0, 0,  0x00000000_00000000)), 0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16382,  0, 0,  0x80000000_00000000)), 0.5);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 16382,  0, 0,  0x80000000_00000000)), -0.5);

        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16385,  0, 0,  0x40000000_00000000)),
            2.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 16385,  0, 0,  0x40000000_00000000)),
            -2.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16381,  0, 0,  0x40000000_00000000)),
            0.125);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 16381,  0, 0,  0x40000000_00000000)),
            -0.125);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16383+1023,  0, 0,  0x80000000_00000000)),
            8.98846567431158e307);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 16383-1021,  0, 0,  0x80000000_00000000)),
            4.450147717014403e-308);
        assert_eq_f64!(f80_to_f64(&[0x40, 0x0e, 0xac, 0x44, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00]),
            44100.0);

        // large f80 value overflows f64
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 32766,  0, 0,  0xffffffff_ffffffff)),
            f64::INFINITY);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 32766,  0, 0,  0xffffffff_ffffffff)),
            -f64::INFINITY);

        // small f80 value underflows f64
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf,  1, 1,  0, 0,  0x00000000_00000000)), 0.0);
        assert_eq_f64!(f80_to_f64(f80(&mut f80buf, -1, 1,  0, 0,  0x00000000_00000000)), -0.0);
    }

    #[test]
    fn test_f64_to_f80() {
        assert_eq!(f64_to_f80(0.0), [ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]);
        assert_eq!(f64_to_f80(-0.0), [ 128, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]);

        assert_eq!(f64_to_f80(1.0), [63, 255, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(2.0), [64, 0, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(0.5), [63, 254, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(0.125), [63, 252, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(20000.0002), [64, 13, 156, 64, 0, 26, 54, 226, 232, 0]);
        assert_eq!(f64_to_f80(44100.0), [64, 14, 172, 68, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(1.341e152), [65, 248, 163, 221, 229, 184, 249, 213, 80, 0]);
        assert_eq!(f64_to_f80(1.341e-152), [62, 6, 179, 204, 119, 227, 61, 192, 40, 0]);
        assert_eq!(f64_to_f80(8.98846567431158e307), [67, 254, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(4.450147717014403e-308), [60, 2, 128, 0, 0, 0, 0, 0, 0, 0]);

        assert_eq!(f64_to_f80(-1.0), [191, 255, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(-2.0), [192, 0, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(-0.5), [191, 254, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(-0.125), [191, 252, 128, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(-20000.0002), [192, 13, 156, 64, 0, 26, 54, 226, 232, 0]);
        assert_eq!(f64_to_f80(-44100.0), [192, 14, 172, 68, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(-1.341e152), [193, 248, 163, 221, 229, 184, 249, 213, 80, 0]);
        assert_eq!(f64_to_f80(-1.341e-152), [190, 6, 179, 204, 119, 227, 61, 192, 40, 0]);

        assert_eq!(f64_to_f80(f64::MAX), [67, 254, 255, 255, 255, 255, 255, 255, 248, 0]);
        assert_eq!(f64_to_f80(f64::MIN), [195, 254, 255, 255, 255, 255, 255, 255, 248, 0]);
        assert_eq!(f64_to_f80(f64::MIN_POSITIVE), [60, 1, 128, 0, 0, 0, 0, 0, 0, 0]);

        assert_eq!(f64_to_f80(f64::INFINITY), [127, 255, 0, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(-f64::INFINITY), [255, 255, 0, 0, 0, 0, 0, 0, 0, 0]);
        assert_eq!(f64_to_f80(f64::NAN), [127, 255, 255, 255, 255, 255, 255, 255, 255, 255]);
    }
}