format_parser 0.25.0 → 0.25.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: c24092543b6c98c713b07a45c3e3a2990332858397e11f6709470c32343a62f3
-  data.tar.gz: 0222b05ddfa1efa7cee364db7ad61ae7d44806b9622d6d6012fbb50e1b2e8138
+  metadata.gz: 87adbfef15c2281ab6a13f151b857409f0ffad0ecc5270d9d0bbc5cebe207cdb
+  data.tar.gz: 332e3c4efd4ae01b3cf47c669debba7cc1aee5c264bef503720f632f6d801054
 SHA512:
-  metadata.gz: 67a3a64166115ef70b2043c05e9c105c2939752b1de34086421a2b96311880673d1c668927eecf904e06db368cc9979218d666004b59a187f52e569f47e9d2a3
-  data.tar.gz: 5f4034dbcc2a92cb4908c6609465fbb81ec5011f42f22e9c5477b72a26ee4b59bbaa1191a13a5405d71f68ac8b8c0ceac652befd9eb2c1ea9fc5af000b221828
+  metadata.gz: 231d768d4b69b2c2f29bcb861888a7fbb0f4242eec5c9313d6428c9053b4fb4e7d20b8615d731a695c278ac59d3bf07b4977d217892aa1fce8fa7adc9d415efa
+  data.tar.gz: 732fae92e71f6a25fa98b7d88fa70b82404f00c234a6debd641e20ad8ffd9e45c78270f20adb8ad5593cd815e128a2a409ec639351c985b79763d4bfd821fec9

data/CHANGELOG.md

@@ -1,3 +1,8 @@
+## 0.25.1
+* MOV: Fix error "negative length"
+* MOV: Fix reading dimensions in multi-track files
+* MP3: Fix parse of the Xing header to not raise errors
+
 ## 0.25.0
 * MP3: add suport to id3 v2.4.x
 * JPEG: Update gem exifr to 1.3.8 to fix a bug

data/lib/format_parser/version.rb

@@ -1,3 +1,3 @@
 module FormatParser
-  VERSION = '0.25.0'
+  VERSION = '0.25.1'
 end

data/lib/parsers/moov_parser.rb

@@ -38,14 +38,8 @@ class FormatParser::MOOVParser
     ftyp_atom = decoder.find_first_atom_by_path(atom_tree, 'ftyp')
     file_type = ftyp_atom.field_value(:major_brand)
 
-    width = nil
-    height = nil
-
     # Try to find the width and height in the tkhd
-    if tkhd = decoder.find_first_atom_by_path(atom_tree, 'moov', 'trak', 'tkhd')
-      width = tkhd.field_value(:track_width).first
-      height = tkhd.field_value(:track_height).first
-    end
+    width, height = parse_dimensions(decoder, atom_tree)
 
     # Try to find the "topmost" duration (respecting edits)
     if mdhd = decoder.find_first_atom_by_path(atom_tree, 'moov', 'mvhd')
@@ -78,6 +72,31 @@ class FormatParser::MOOVParser
     FTYP_MAP.fetch(file_type.downcase, :mov)
   end
 
+  # The dimensions are located in tkhd atom, but in some files it is necessary
+  # to get it below the video track, because it can have other tracks such as
+  # audio which does not have the dimensions.
+  # More details in https://developer.apple.com/library/archive/documentation/QuickTime/QTFF/QTFFChap2/qtff2.html#//apple_ref/doc/uid/TP40000939-CH204-DontLinkElementID_147
+  #
+  # Returns [width, height] if the dimension is found
+  # Returns [nil, nil] if the dimension is not found
+  def parse_dimensions(decoder, atom_tree)
+    video_trak_atom = decoder.find_video_trak_atom(atom_tree)
+
+    tkhd = begin
+      if video_trak_atom
+        decoder.find_first_atom_by_path([video_trak_atom], 'trak', 'tkhd')
+      else
+        decoder.find_first_atom_by_path(atom_tree, 'moov', 'trak', 'tkhd')
+      end
+    end
+
+    if tkhd
+      [tkhd.field_value(:track_width).first, tkhd.field_value(:track_height).first]
+    else
+      [nil, nil]
+    end
+  end
+
   # An MPEG4/MOV/M4A will start with the "ftyp" atom. The atom must have a length
   # of at least 8 (to accomodate the atom size and the atom type itself) plus the major
   # and minor version fields. If we cannot find it we can be certain this is not our file.

data/lib/parsers/moov_parser/decoder.rb

@@ -1,6 +1,7 @@
 # Handles decoding of MOV/MPEG4 atoms/boxes in a stream. Will recursively
 # read atoms and parse their data fields if applicable. Also contains
 # a few utility functions for finding atoms in a list etc.
+# To know more about Atoms: https://developer.apple.com/library/archive/documentation/QuickTime/QTFF/QTFFChap2/qtff2.html
 class FormatParser::MOOVParser::Decoder
   include FormatParser::IOUtils
 
@@ -47,6 +48,34 @@ class FormatParser::MOOVParser::Decoder
     find_first_atom_by_path(requisite.children || [], *atom_types)
   end
 
+  def find_atoms_by_path(atoms, atom_types)
+    type_to_find = atom_types.shift
+    requisites = atoms.select { |e| e.atom_type == type_to_find }
+
+    # Return if we found our match
+    return requisites if atom_types.empty?
+
+    # Return nil if we didn't find the match at this nesting level
+    return unless requisites
+
+    # ...otherwise drill further down
+    find_atoms_by_path(requisites.flat_map(&:children).compact || [], atom_types)
+  end
+
+  # A file can have multiple tracks. To identify the type it is necessary to check
+  # the fields `omponent_subtype` in hdlr atom under the trak atom
+  # More details in https://developer.apple.com/library/archive/documentation/QuickTime/QTFF/QTFFChap2/qtff2.html#//apple_ref/doc/uid/TP40000939-CH204-DontLinkElementID_147
+  def find_video_trak_atom(atoms)
+    trak_atoms = find_atoms_by_path(atoms, ['moov', 'trak'])
+
+    return [] if trak_atoms.empty?
+
+    trak_atoms.find do |trak_atom|
+      hdlr_atom = find_first_atom_by_path([trak_atom], 'trak', 'mdia', 'hdlr')
+      hdlr_atom.atom_fields[:component_type] == 'mhlr' && hdlr_atom.atom_fields[:component_subtype] == 'vide'
+    end
+  end
+
   def parse_ftyp_atom(io, atom_size)
     # Subtract 8 for the atom_size+atom_type,
     # and 8 once more for the major_brand and minor_version. The remaining
@@ -194,6 +223,8 @@ class FormatParser::MOOVParser::Decoder
   end
 
   def parse_meta_atom(io, atom_size)
+    return if atom_size == 0 # this atom can be empty
+
     parse_hdlr_atom(io, atom_size)
   end
 

data/lib/parsers/mp3_parser.rb

@@ -249,16 +249,16 @@ class FormatParser::MP3Parser
     io.seek(xing_offset + 4) # Include the length of "Xing" itself
 
     # https://www.codeproject.com/Articles/8295/MPEG-Audio-Frame-Header#XINGHeader
-    header_flags, _ = io.read(4).unpack('s>s>')
+    header_flags, _ = io.read(4).unpack('i>')
     frames = byte_count = toc = vbr_scale = nil
 
-    frames = io.read(4).unpack('N1').first if header_flags & 1 # FRAMES FLAG
+    frames = io.read(4).unpack('N1').first if header_flags & 1 != 0   # FRAMES FLAG
 
-    byte_count = io.read(4).unpack('N1').first if header_flags & 2 # BYTES FLAG
+    byte_count = io.read(4).unpack('N1').first if header_flags & 2 != 0   # BYTES FLAG
 
-    toc = io.read(100).unpack('C100') if header_flags & 4 # TOC FLAG
+    toc = io.read(100).unpack('C100') if header_flags & 4 != 0   # TOC FLAG
 
-    vbr_scale = io.read(4).unpack('N1').first if header_flags & 8 # VBR SCALE FLAG
+    vbr_scale = io.read(4).unpack('N1').first if header_flags & 8 != 0   # VBR SCALE FLAG
 
     VBRHeader.new(frames: frames, byte_count: byte_count, toc_entries: toc, vbr_scale: vbr_scale)
   end

data/spec/parsers/moov_parser_spec.rb

@@ -108,4 +108,24 @@ describe FormatParser::MOOVParser do
   it 'provides filename hints' do
     expect(subject).to be_likely_match('file.m4v')
   end
+
+  it 'reads correctly the video dimensions' do
+    mov_path = fixtures_dir + '/MOOV/MOV/Test_Dimensions.mov'
+
+    result = subject.call(File.open(mov_path, 'rb'))
+
+    expect(result).not_to be_nil
+    expect(result.nature).to eq(:video)
+    expect(result.format).to eq(:mov)
+    expect(result.width_px).to eq(640)
+    expect(result.height_px).to eq(360)
+  end
+
+  it 'does not raise error when a meta atom has size 0' do
+    mov_path = fixtures_dir + '/MOOV/MOV/Test_Meta_Atom_With_Size_Zero.mov'
+
+    result = subject.call(File.open(mov_path, 'rb'))
+    expect(result).not_to be_nil
+    expect(result.format).to eq(:mov)
+  end
 end

data/spec/parsers/mp3_parser_spec.rb

@@ -15,6 +15,20 @@ describe FormatParser::MP3Parser do
     expect(parsed.media_duration_seconds).to be_within(0.1).of(0.836)
   end
 
+  it 'reads the Xing header without raising errors' do
+    fpath = fixtures_dir + '/MP3/test_xing_header.mp3'
+    parsed = subject.call(File.open(fpath, 'rb'))
+
+    expect(parsed).not_to be_nil
+
+    expect(parsed.nature).to eq(:audio)
+    expect(parsed.format).to eq(:mp3)
+    expect(parsed.num_audio_channels).to eq(2)
+    expect(parsed.audio_sample_rate_hz).to eq(48000)
+    expect(parsed.intrinsics).not_to be_nil
+    expect(parsed.media_duration_seconds).to be_within(0.1).of(0.0342)
+  end
+
   it 'does not misdetect a PNG' do
     fpath = fixtures_dir + '/PNG/anim.png'
     parsed = subject.call(File.open(fpath, 'rb'))

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: format_parser
 version: !ruby/object:Gem::Version
-  version: 0.25.0
+  version: 0.25.1
 platform: ruby
 authors:
 - Noah Berman
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2020-09-30 00:00:00.000000000 Z
+date: 2020-10-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ks