format_parser 0.1.3 → 0.1.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: b0ef6923a01b8fbe52f4491979a3be1224ea7018
-  data.tar.gz: a19859f36a81154f73d1071857834cd0e7e1ca74
+  metadata.gz: 99d5518139da4b6ee5131ec57009f9c3982a3a3d
+  data.tar.gz: 386ca02d14fac20a0d9d74bce6a43a65d862b73c
 SHA512:
-  metadata.gz: ca6bd5d8324a4dcb41d6f3137a1ab621a016acf8e9fafae5983bb8c325f883ecc72b4046d774e423e5e4ca7f35d048dae20a1e6ca2287d842c111c2714e2d606
-  data.tar.gz: 99d6517341e8b48635c8540e7f92cb48e772ed02047eb6cd56412b21de2411c06d64d43878ffa1a3ce43bf8e82b7e886eef4f00b221bfdfe42bafc19435f3f35
+  metadata.gz: 791c835fbd9a2cbbcb8dce9dc07c34417967880fff6c3e044ae07142c5da5e106be378041461b28673136a2639ae42ae24022438a00d4e53366edd126bc7ac1a
+  data.tar.gz: 196a9d2137cee58e07621cd9e6140320863358830c4ca20407c11e5d523dce08790a9e132c97045e44a7d48ab82105499b6a0f9aa21314ccb5682136e87922c7

data/README.md

@@ -9,7 +9,7 @@ and [dimensions,](https://github.com/sstephenson/dimensions) borrowing from them
 
 ## Currently supported filetypes:
 
-`TIFF, PSD, PNG, MP3, JPEG, GIF, DPX, AIFF`
+`TIFF, PSD, PNG, MP3, JPEG, GIF, DPX, AIFF, WAV`
 
 ...with more on the way!
 
@@ -56,5 +56,12 @@ Therefore we adapt the following approaches:
 
 ## Fixture Sources
 
-- MIT licensed fixture files from the FastImage and Dimensions projects
+Unless specified otherwise in this section the fixture files are MIT licensed and from the FastImage and Dimensions projects.
+
+### AIFF
 - fixture.aiff was created by one of the project maintainers and is MIT licensed
+
+### WAV
+- c_11k16bitpcm.wav and c_8kmp316.wav are from [Wikipedia WAV](https://en.wikipedia.org/wiki/WAV#Comparison_of_coding_schemes), retrieved January 7, 2018
+- c_39064__alienbomb__atmo-truck.wav is from [freesound](https://freesound.org/people/alienbomb/sounds/39064/) and is CC0 licensed
+- c_M1F1-Alaw-AFsp.wav and d_6_Channel_ID.wav are from a [McGill Engineering site](http://www-mmsp.ece.mcgill.ca/Documents/AudioFormats/WAVE/Samples.html)

data/lib/format_parser/version.rb

@@ -1,3 +1,3 @@
 module FormatParser
-  VERSION = '0.1.3'
+  VERSION = '0.1.4'
 end

data/lib/parsers/wav_parser.rb

@@ -0,0 +1,102 @@
+class FormatParser::WAVParser
+  include FormatParser::IOUtils
+
+  def information_from_io(io)
+    io.seek(0)
+
+    # 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.
+    # http://soundfile.sapp.org/doc/WaveFormat/
+    # For WAVE files containing PCM audio format we parse the 'fmt ' and
+    # 'data' chunks while for non PCM audio formats the 'fmt ' and 'fact'
+    # chunks. In the latter case the order fo appearence of the chunks is
+    # arbitrary.
+    fmt_processed = false
+    fact_processed = false
+    fmt_data = {}
+    total_sample_frames = 0
+    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)
+        if fmt_data[:audio_format] != 1 and fact_processed
+          return process_non_pcm(fmt_data, total_sample_frames)
+        end
+        fmt_processed = true
+      when 'data'
+        return unless fmt_processed # the 'data' chunk cannot preceed the 'fmt ' chunk
+        return process_pcm(fmt_data, chunk_size) if fmt_data[:audio_format] == 1
+        safe_skip(io, chunk_size)
+      when 'fact'
+        total_sample_frames = safe_read(io, 4).unpack('l').first
+        safe_skip(io, chunk_size - 4)
+        if fmt_processed and fmt_data[:audio_format] != 1
+          return process_non_pcm(fmt_data, total_sample_frames)
+        end
+        fact_processed = true
+      else # Skip this chunk until a known chunk is encountered
+        safe_skip(io, chunk_size)
+      end
+    end
+  end
+
+  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 last 4 fields of the fmt tag are always:
+    # * unsigned short     channels
+    # * unsigned long      samples per sec
+    # * unsigned long      average bytes per sec
+    # * unsigned short     block align
+    # * unsigned short     bits per sample
+
+    fmt_info = safe_read(io, 16).unpack('S_2I2S_2')
+    safe_skip(io, chunk_size - 16) # skip the extra fields
+
+    {
+      audio_format:    fmt_info[0],
+      channels:        fmt_info[1],
+      sample_rate:     fmt_info[2],
+      byte_rate:       fmt_info[3],
+      block_align:     fmt_info[4],
+      bits_per_sample: fmt_info[5],
+    }
+  end
+
+  def process_pcm(fmt_data, data_size)
+    return unless fmt_data[:channels] > 0 and fmt_data[:bits_per_sample] > 0
+    sample_frames = data_size / (fmt_data[:channels] * fmt_data[:bits_per_sample] / 8)
+    file_info(fmt_data, sample_frames)
+  end
+
+  def process_non_pcm(fmt_data, total_sample_frames)
+    file_info(fmt_data, total_sample_frames)
+  end
+
+  def file_info(fmt_data, sample_frames)
+    return unless fmt_data[:sample_rate] > 0
+    duration_in_seconds = sample_frames / fmt_data[:sample_rate].to_f
+    FormatParser::FileInformation.new(
+      file_nature: :audio,
+      file_type: :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,
+    )
+  end
+
+  FormatParser.register_parser_constructor self
+end

data/spec/parsers/wav_parser_spec.rb

@@ -0,0 +1,53 @@
+require 'spec_helper'
+
+describe FormatParser::WAVParser do
+  # Fixtures prefixed with c_ are considered canonical
+  # while fixtures prefixed with d_ deviate from the standard.
+  Dir.glob(fixtures_dir + '/WAV/c_*.*').each do |wav_path|
+    it "is able to parse #{File.basename(wav_path)}" do
+      parse_result = subject.information_from_io(File.open(wav_path, 'rb'))
+
+      expect(parse_result.file_nature).to eq(:audio)
+      expect(parse_result.file_type).to eq(:wav)
+    end
+  end
+
+  it "returns correct info about pcm files" do
+    parse_result = subject.information_from_io(File.open(__dir__ + '/../fixtures/WAV/c_8kmp316.wav', 'rb'))
+
+    expect(parse_result.file_nature).to eq(:audio)
+    expect(parse_result.file_type).to eq(:wav)
+    expect(parse_result.num_audio_channels).to eq(1)
+    expect(parse_result.audio_sample_rate_hz).to eq(8000)
+    expect(parse_result.media_duration_frames).to eq(110488)
+    expect(parse_result.media_duration_seconds).to be_within(0.01).of(13.81)
+  end
+
+  it "returns correct info about pcm files with more channels" do
+    parse_result = subject.information_from_io(File.open(__dir__ + '/../fixtures/WAV/c_39064__alienbomb__atmo-truck.wav', 'rb'))
+
+    expect(parse_result.file_nature).to eq(:audio)
+    expect(parse_result.file_type).to eq(:wav)
+    expect(parse_result.num_audio_channels).to eq(2)
+    expect(parse_result.audio_sample_rate_hz).to eq(44100)
+    expect(parse_result.media_duration_frames).to eq(162832)
+    expect(parse_result.media_duration_seconds).to be_within(0.01).of(3.69)
+  end
+
+  it "returns correct info about non pcm files" do
+    parse_result = subject.information_from_io(File.open(__dir__ + '/../fixtures/WAV/c_11k16bitpcm.wav', 'rb'))
+
+    expect(parse_result.file_nature).to eq(:audio)
+    expect(parse_result.file_type).to eq(:wav)
+    expect(parse_result.num_audio_channels).to eq(1)
+    expect(parse_result.audio_sample_rate_hz).to eq(11025)
+    expect(parse_result.media_duration_frames).to eq(152267)
+    expect(parse_result.media_duration_seconds).to be_within(0.01).of(13.81)
+  end
+
+  it "cannot parse file with audio format different from 1 and no 'fact' chunk" do
+    expect {
+      subject.information_from_io(File.open(__dir__ + '/../fixtures/WAV/d_6_Channel_ID.wav', 'rb'))
+    }.to raise_error(FormatParser::IOUtils::InvalidRead)
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: format_parser
 version: !ruby/object:Gem::Version
-  version: 0.1.3
+  version: 0.1.4
 platform: ruby
 authors:
 - Noah Berman
@@ -161,6 +161,7 @@ files:
 - lib/parsers/png_parser.rb
 - lib/parsers/psd_parser.rb
 - lib/parsers/tiff_parser.rb
+- lib/parsers/wav_parser.rb
 - lib/read_limiter.rb
 - lib/remote_io.rb
 - spec/aiff_parser_spec.rb
@@ -176,6 +177,7 @@ files:
 - spec/parsers/png_parser_spec.rb
 - spec/parsers/psd_parser_spec.rb
 - spec/parsers/tiff_parser_spec.rb
+- spec/parsers/wav_parser_spec.rb
 - spec/read_limiter_spec.rb
 - spec/remote_fetching_spec.rb
 - spec/remote_io_spec.rb
@@ -201,7 +203,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.5.2
+rubygems_version: 2.5.1
 signing_key: 
 specification_version: 4
 summary: A library for efficient parsing of file metadata