Class: FormatParser::WAVParser

Inherits:
Object
  • Object
show all
Includes:
IOUtils
Defined in:
lib/parsers/wav_parser.rb

Constant Summary collapse

WAV_MIME_TYPE = 
'audio/x-wav'

Constants included from IOUtils

IOUtils::INTEGER_DIRECTIVES

Instance Method Summary collapse

Methods included from IOUtils

#read_bytes, #read_fixed_point, #read_int, #safe_read, #safe_skip, #skip_bytes

Instance Method Details

#call(io) ⇒ Object 



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
 
# File 'lib/parsers/wav_parser.rb', line 10

def call(io)
  # Read the RIFF header. Chunk descriptor should be RIFF, the size should
  # contain the size of the entire file in bytes minus 8 bytes for the
  # two fields not included in this count: chunk_id and size.
  chunk_id, _size, riff_type = safe_read(io, 12).unpack('a4la4')

  # The chunk_id and riff_type should be RIFF and WAVE respectively
  return unless chunk_id == 'RIFF' && riff_type == 'WAVE'

  # There are no restrictions upon the order of the chunks within a WAVE file,
  # with the exception that the Format chunk must precede the Data chunk.
  # The specification does not require the Format chunk to be the first chunk
  # after the RIFF header.
  # https://www.mmsp.ece.mcgill.ca/Documents/AudioFormats/WAVE/WAVE.html
  fmt_data = {}
  data_size = 0
  total_sample_frames = nil
  loop do
    chunk_type, chunk_size = safe_read(io, 8).unpack('a4l')
    case chunk_type
    when 'fmt ' # watch out: the chunk ID of the format chunk ends with a space
      fmt_data = unpack_fmt_chunk(io, chunk_size)
    when 'data'
      data_size = chunk_size
    when 'fact'
      total_sample_frames = safe_read(io, 4).unpack('l').first
      safe_skip(io, chunk_size - 4)
    else
      # Skip this chunk until a known chunk is encountered
      safe_skip(io, chunk_size)
    end
  rescue FormatParser::IOUtils::InvalidRead
    # We've reached EOF, so it's time to make the most out of the metadata we
    # managed to parse
    break
  end

  file_info(fmt_data, data_size, total_sample_frames)
end

#file_info(fmt_data, data_size, sample_frames) ⇒ Object 



73
74
75
76
77
78
79
80
81
82
83
84
85
 
# File 'lib/parsers/wav_parser.rb', line 73

def file_info(fmt_data, data_size, sample_frames)
  # NOTE: Each sample includes information for each channel
  sample_frames ||= data_size / (fmt_data[:channels] * fmt_data[:bits_per_sample] / 8) if fmt_data[:channels] > 0 && fmt_data[:bits_per_sample] > 0
  duration_in_seconds = sample_frames / fmt_data[:sample_rate].to_f if fmt_data[:sample_rate] > 0
  FormatParser::Audio.new(
    format: :wav,
    num_audio_channels: fmt_data[:channels],
    audio_sample_rate_hz: fmt_data[:sample_rate],
    media_duration_frames: sample_frames,
    media_duration_seconds: duration_in_seconds,
    content_type: WAV_MIME_TYPE,
  )
end

#likely_match?(filename) ⇒ Boolean

Returns:

  • (Boolean) 


6
7
8
 
# File 'lib/parsers/wav_parser.rb', line 6

def likely_match?(filename)
  filename =~ /\.wav$/i
end

#unpack_fmt_chunk(io, chunk_size) ⇒ Object 



50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
 
# File 'lib/parsers/wav_parser.rb', line 50

def unpack_fmt_chunk(io, chunk_size)
  # The size of the fmt chunk is at least 16 bytes. If the format tag's value is not
  # 1 compression might be in use for storing the data
  # and the fmt chunk might contain extra fields appended to it.
  # The first 6 fields of the fmt tag are always:
  # * unsigned short     audio format
  # * unsigned short     channels
  # * unsigned long      samples per sec
  # * unsigned long      average bytes per sec
  # * unsigned short     block align
  # * unsigned short     bits per sample

  _, channels, sample_rate, byte_rate, _, bits_per_sample = safe_read(io, 16).unpack('S_2I2S_2')
  safe_skip(io, chunk_size - 16) # skip the extra fields

  {
    channels: channels,
    sample_rate: sample_rate,
    byte_rate: byte_rate,
    bits_per_sample: bits_per_sample,
  }
end