format_parser 2.7.2 → 2.9.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 41515e692a4ba72edc213859230083a47f91b649f05fe1c4b1aeccdfd7eac149
-  data.tar.gz: 71829cb1c447121aa4384237b83732d1bcb6b29bd9aa460950963029c8fdc1b3
+  metadata.gz: 04b14799c012037cf53b949b0920e0f8e2d5a42e30182363fee0d283d41eade4
+  data.tar.gz: 78ff4a354e05fb0e67a7b8dd8368768fceb313afc4dd736d5fa41592dfbc7b35
 SHA512:
-  metadata.gz: cafea2033f3f00535897be3ca2c1c15918b1a826978be14af4e9156249908e5950b1077908eb275c7f11e7f61f9c15507a5823a50623bdb410c6b274cb27a3e8
-  data.tar.gz: 9164177d753a9258d3966cf07b38ecf3121975ba9f19e42cc93834b2fa26b1d60e6756fcde11f1832e8c2b2cc7accc3926ed752b5ff1a1bb88baf9bb8ac03d1a
+  metadata.gz: e5792c23191b2d4b4d63a5d9dfbfe5a3b28862f9e8ba977c49e39f3ebaf6bd222a31c296f224743a00fc877f8dbbfb6f97cdd19a0b6d712db6e3a00d29dd30d8
+  data.tar.gz: 6fd2ac968375090928b407afa3234c623c933150a36a00da965e75bb97d01ee2dcb502124e71fa100f0f2321b16a0f268c8a8fa94691cd195fbeb65a1fbab177

data/.github/workflows/main.yml

@@ -17,6 +17,8 @@ jobs:
           - 2.7
           - 3.0
           - 3.1
+          - 3.2
+          - 3.3
           - jruby
     steps:
       - name: Checkout
@@ -64,6 +66,8 @@ jobs:
           - 2.7
           - 3.0
           - 3.1
+          - 3.2
+          - 3.3
           - jruby
         experimental: [false]
     steps:

data/CHANGELOG.md

@@ -1,3 +1,18 @@
+## 2.9.0
+* Improve WAV parser by performing a best-effort when extracting metadata from files that do not strictly follow the format spec.
+
+## 2.8.0
+* Add support for Ruby 3.2 and 3.3.
+
+## 2.7.2
+* Improved stability for mp4 parser when dealing with corrupted FTYP boxes.
+
+## 2.7.1
+* Fixed bug with WAV file wrongly parsed as MP3.
+
+## 2.7.0
+* JSON format support.
+
 ## 2.6.0
 * Prevent the default loading of thumbnails on TIFF-based formats to improve I/O.
 

data/lib/format_parser/version.rb

@@ -1,3 +1,3 @@
 module FormatParser
-  VERSION = '2.7.2'
+  VERSION = '2.9.0'
 end

data/lib/parsers/wav_parser.rb

@@ -20,75 +20,60 @@ class FormatParser::WAVParser
     # 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
+    # https://www.mmsp.ece.mcgill.ca/Documents/AudioFormats/WAVE/WAVE.html
     fmt_data = {}
-    total_sample_frames = 0
+    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)
-        return process_non_pcm(fmt_data, total_sample_frames) if fmt_data[:audio_format] != 1 and fact_processed
-        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)
+        data_size = chunk_size
       when 'fact'
         total_sample_frames = safe_read(io, 4).unpack('l').first
         safe_skip(io, chunk_size - 4)
-        return process_non_pcm(fmt_data, total_sample_frames) if fmt_processed and fmt_data[:audio_format] != 1
-        fact_processed = true
       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
 
   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:
+    # 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
 
-    fmt_info = safe_read(io, 16).unpack('S_2I2S_2')
+    _, 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
 
     {
-      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],
+      channels: channels,
+      sample_rate: sample_rate,
+      byte_rate: byte_rate,
+      bits_per_sample: bits_per_sample,
     }
   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
+  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],

data/spec/parsers/wav_parser_spec.rb

@@ -46,9 +46,14 @@ describe FormatParser::WAVParser do
     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.call(File.open(__dir__ + '/../fixtures/WAV/invalid_d_6_Channel_ID.wav', 'rb'))
-    }.to raise_error(FormatParser::IOUtils::InvalidRead)
+  it 'returns correct info about non pcm files with no fact chunk' do
+    parse_result = subject.call(File.open(__dir__ + '/../fixtures/WAV/d_6_Channel_ID.wav', 'rb'))
+
+    expect(parse_result.nature).to eq(:audio)
+    expect(parse_result.format).to eq(:wav)
+    expect(parse_result.num_audio_channels).to eq(6)
+    expect(parse_result.audio_sample_rate_hz).to eq(44100)
+    expect(parse_result.media_duration_frames).to eq(257411)
+    expect(parse_result.media_duration_seconds).to be_within(0.01).of(5.83)
   end
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: format_parser
 version: !ruby/object:Gem::Version
-  version: 2.7.2
+  version: 2.9.0
 platform: ruby
 authors:
 - Noah Berman
@@ -9,7 +9,7 @@ authors:
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2023-09-19 00:00:00.000000000 Z
+date: 2024-07-09 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: exifr
@@ -334,7 +334,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.3.7
+rubygems_version: 3.3.26
 signing_key:
 specification_version: 4
 summary: A library for efficient parsing of file metadata