File: juce_SingleThreadedAbstractFifo.h

package info (click to toggle)
juce 8.0.10%2Bds-2
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 78,768 kB
  • sloc: cpp: 526,464; ansic: 159,952; java: 3,038; javascript: 847; xml: 269; python: 224; sh: 167; makefile: 84
file content (138 lines) | stat: -rw-r--r-- 5,878 bytes parent folder | download | duplicates (2)
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
/*
  ==============================================================================

   This file is part of the JUCE framework.
   Copyright (c) Raw Material Software Limited

   JUCE is an open source framework subject to commercial or open source
   licensing.

   By downloading, installing, or using the JUCE framework, or combining the
   JUCE framework with any other source code, object code, content or any other
   copyrightable work, you agree to the terms of the JUCE End User Licence
   Agreement, and all incorporated terms including the JUCE Privacy Policy and
   the JUCE Website Terms of Service, as applicable, which will bind you. If you
   do not agree to the terms of these agreements, we will not license the JUCE
   framework to you, and you must discontinue the installation or download
   process and cease use of the JUCE framework.

   JUCE End User Licence Agreement: https://juce.com/legal/juce-8-licence/
   JUCE Privacy Policy: https://juce.com/juce-privacy-policy
   JUCE Website Terms of Service: https://juce.com/juce-website-terms-of-service/

   Or:

   You may also use this code under the terms of the AGPLv3:
   https://www.gnu.org/licenses/agpl-3.0.en.html

   THE JUCE FRAMEWORK IS PROVIDED "AS IS" WITHOUT ANY WARRANTY, AND ALL
   WARRANTIES, WHETHER EXPRESSED OR IMPLIED, INCLUDING WARRANTY OF
   MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, ARE DISCLAIMED.

  ==============================================================================
*/

namespace juce
{

//==============================================================================
/**
    Encapsulates the logic for a single-threaded FIFO.

    This might be useful for building buffers which can be written and read in
    blocks of different sizes. For example, in an audio effect we might wish to
    run some processing on fixed-size blocks of audio input, but the host may
    provide input blocks of varying sizes. In this situation, we might want to
    store the previous input in a buffer, and extract a fixed-size block
    whenever there are enough samples available. The SingleThreadedAbstractFifo
    implements logic suitable for this use-case.

    This class is quite similar to AbstractFifo, in that it only keeps track of
    the current read/write locations. The user is responsible for providing the
    actual buffer that will be read/written.

    The intended usage of this class is as follows:
    - Create some backing storage in a vector, AudioBuffer etc.
    - Construct a SingleThreadedAbstractFifo to manage the buffer, passing the
      number of items in the buffer.
    - Each time new input is ready, call write(), passing the number of items
      you wish to write into the buffer. This function returns a pair of ranges
      describing which indices in the backing storage should be written.
    - Call getNumReadable() to find out how many items are ready to read from
      the buffer.
    - If there are enough items ready to read, call read(), passing the number
      of items you require. This function returns a pair of ranges describing
      which indices in the backing storage may be read.

    Unlike AbstractFifo, the SingleThreadedAbstractFifo is intended for use
    from a single thread. It is not safe to call any non-const member function
    of SingleThreadedAbstractFifo concurrently with any other member function.

    @see AbstractFifo

    @tags{Core}
*/
class SingleThreadedAbstractFifo
{
public:
    /** Creates a SingleThreadedAbstractFifo with no size. */
    SingleThreadedAbstractFifo() = default;

    /** Creates a SingleThreadedAbstractFifo that can manage a buffer of the specified size. */
    explicit SingleThreadedAbstractFifo (int sizeIn)
        : size (sizeIn)
    {
        // This class only works properly when the size is a power of two.
        // Use nextPowerOfTwo() to find a good size, and ensure that your
        // backing storage is the same size.
        jassert (isPowerOfTwo (sizeIn));
    }

    /** Returns the number of unused elements present in the buffer. */
    int getRemainingSpace() const   { return size - numReadable; }

    /** Returns the number of pending elements present in the buffer. */
    int getNumReadable() const      { return numReadable; }

    /** Returns the size of the managed buffer. */
    int getSize() const             { return size; }

    /** Returns two blocks in the buffer where new items may be written.

        Note that if the buffer is running low on free space, the sum of the lengths of
        the returned ranges may be less than num!
    */
    std::array<Range<int>, 2> write (int num)
    {
        const auto startPos = (readPos + numReadable) & (size - 1);
        const auto maxToWrite = jmin (getRemainingSpace(), num);
        const auto firstBlockSize = jmin (maxToWrite, size - startPos);

        numReadable += maxToWrite;

        return { { { startPos, startPos + firstBlockSize }, { 0, maxToWrite - firstBlockSize } } };
    }

    /** Returns two blocks in the buffer from which new items may be read.

        Note that if the buffer doesn't have the requested number of items available,
        the sum of the lengths of the returned ranges may be less than num!
    */
    std::array<Range<int>, 2> read (int num)
    {
        const auto startPos = readPos;
        const auto maxToRead = jmin (numReadable, num);
        const auto firstBlockSize = jmin (maxToRead, size - startPos);

        readPos = (startPos + maxToRead) & (size - 1);
        numReadable -= maxToRead;

        return { { { startPos, startPos + firstBlockSize }, { 0, maxToRead - firstBlockSize } } };
    }

private:
    int size = 0, readPos = 0, numReadable = 0;
};


} // namespace juce