webvtt-ruby 0.3.2 → 0.4.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: e0ecae0ceefe5452aa14107dcb45c892ff9d02a6
-  data.tar.gz: 173216b97d99db10103bc59207916ac5474748d8
+SHA256:
+  metadata.gz: 378a27bdabf62c13b0446de29e0aefc7267ac38f67401f1bab668f1ea673edab
+  data.tar.gz: 9b68d17e5435a1ef088bc06ec1c5346cc3526dc3af5b56578c14afca6c922656
 SHA512:
-  metadata.gz: 252d353160e8055adf98277d1c581cec2d13d975be165a8ec7108b10574606173e71946f670242c71def21e7f5a8eab7487834c0bcd06612f4f4b1c95d1b8c77
-  data.tar.gz: dcb306b34ac391d51bcdec9ad9677a7cef97c6aa699a3b4cbf620b699d213f05f1960f839bc811752d96fd25784960f7291633dbd02cdf0d43683834ee277b16
+  metadata.gz: 5b7e76f6306f2df4fcc2f45a194583eb288b4822f2f414723fb6b1675023adbc9ba088be5bbf6b0e6062ef13c3d649dcd0794ab5ff0bca4bbb87012f01c318e6
+  data.tar.gz: ea5a3ea7d52dc8f61bc1ee66fe495e1721f11c4430a1eda17a73b54fd279758eb861b716e2fda1ad5a2e1154400dde4694f83aefe8669995fee3cb0574c528d2

data/README.md

@@ -116,6 +116,6 @@ Usage: bin/webvtt-segmenter [--arg]
 
 **Bruno Celeste**
 
-* http://www.heywatchencoding.com
-* bruno@heywatch.com
-* [@sadikzzz](http://twitter.com/sadikzzz)
+* http://coconut.co
+* bruno@coconut.co
+* [@brunoceleste](http://twitter.com/brunoceleste)

data/lib/webvtt/parser.rb

@@ -4,14 +4,20 @@ module WebVTT
     File.new(file)
   end
 
+  def self.from_blob(content)
+    Blob.new(content)
+  end
+
   def self.convert_from_srt(srt_file, output=nil)
-    if !::File.exists?(srt_file)
+    if !::File.exist?(srt_file)
       raise InputError, "SRT file not found"
     end
 
     srt = ::File.read(srt_file)
     output ||= srt_file.gsub(".srt", ".vtt")
 
+    # normalize timestamps in srt
+    srt.gsub!(/(:|^)(\d)(,|:)/, '\10\2\3')
     # convert timestamps and save the file
     srt.gsub!(/([0-9]{2}:[0-9]{2}:[0-9]{2})([,])([0-9]{3})/, '\1.\3')
     # normalize new line character
@@ -23,19 +29,20 @@ module WebVTT
     return File.new(output)
   end
 
-  class File
-    attr_reader :header, :path, :filename
+  class Blob
+    attr_reader :header
     attr_accessor :cues
 
-    def initialize(webvtt_file)
-      if !::File.exists?(webvtt_file)
-        raise InputError, "WebVTT file not found"
-      end
+    def initialize(content = nil)
+      @cues = []
 
-      @path = webvtt_file
-      @filename = ::File.basename(@path)
-      @content = ::File.read(webvtt_file).gsub("\r\n", "\n") # normalizing new line character
-      parse
+      if content
+        parse(
+          content.gsub("\r\n", "\n").gsub("\r","\n") # normalizing new line character
+        )
+      else
+        @header = 'WEBVTT'
+      end
     end
 
     def to_webvtt
@@ -50,20 +57,12 @@ module WebVTT
       @cues.last.end_in_sec - @cues.first.start_in_sec
     end
 
-    def save(output=nil)
-      output ||= @path.gsub(".srt", ".vtt")
-
-      ::File.open(output, "w") do |f|
-        f.write(to_webvtt)
-      end
-      return output
-    end
-
-    def parse
+    def parse(content)
       # remove bom first
-      @content.gsub!("\uFEFF", '')
+      content.gsub!("\uFEFF", '')
+
+      cues = content.split(/\n\n+/)
 
-      cues = @content.split("\n\n")
       @header = cues.shift
       header_lines = @header.split("\n").map(&:strip)
       if (header_lines[0] =~ /^WEBVTT/).nil?
@@ -81,6 +80,29 @@ module WebVTT
     end
   end
 
+  class File < Blob
+    attr_reader :path, :filename
+
+    def initialize(webvtt_file)
+      if !::File.exist?(webvtt_file)
+        raise InputError, "WebVTT file not found"
+      end
+
+      @path = webvtt_file
+      @filename = ::File.basename(@path)
+      super(::File.read(webvtt_file))
+    end
+
+    def save(output=nil)
+      output ||= @path.gsub(".srt", ".vtt")
+
+      ::File.open(output, "w") do |f|
+        f.write(to_webvtt)
+      end
+      return output
+    end
+  end
+
   class Cue
     attr_accessor :identifier, :start, :end, :style, :text
 
@@ -146,10 +168,12 @@ module WebVTT
         return
       end
 
-      if lines[0].match(/([0-9]{2}:[0-9]{2}:[0-9]{2}\.[0-9]{3}) -+> ([0-9]{2}:[0-9]{2}:[0-9]{2}\.[0-9]{3})(.*)/)
+      if lines[0].match(/(([0-9]{2}:)?[0-9]{2}:[0-9]{2}\.[0-9]{3}) -+> (([0-9]{2}:)?[0-9]{2}:[0-9]{2}\.[0-9]{3})(.*)/)
         @start = Timestamp.new $1
-        @end = Timestamp.new $2
-        @style = Hash[$3.strip.split(" ").map{|s| s.split(":").map(&:strip) }]
+        @end = Timestamp.new $3
+        @style = Hash[$5.strip.split(" ").map{|s| s.split(":").map(&:strip) }]
+      else
+        raise WebVTT::MalformedFile
       end
       @text = lines[1..-1].join("\n")
     end

data/lib/webvtt/segmenter.rb

@@ -1,3 +1,5 @@
+require 'fileutils'
+
 module WebVTT
 
   def self.segment(input, options={})
@@ -25,6 +27,25 @@ module WebVTT
       @options[:length] ||= 10
       @options[:output] ||= "fileSequence-%05d.vtt"
       @options[:playlist] ||= "prog_index.m3u8"
+
+      # a dirty hack to check if output and playlist are ending with / (indicating a directory) and if so use the default filename and playlist name and create directory which is required.
+
+      if @options[:output].end_with?('/')
+        @options[:output] = "#{@options[:output]}fileSequence-%05d.vtt"
+      end
+      if @options[:playlist].end_with?('/')
+        @options[:playlist] = "#{@options[:playlist]}prog_index.m3u8"
+      end
+
+      output_dirname = ::File.dirname(@options[:output])
+      playlist_dirname = ::File.dirname(@options[:playlist])
+
+      unless ::File.directory?(output_dirname)
+        ::FileUtils.mkdir_p(output_dirname)
+      end
+      unless ::File.directory?(playlist_dirname)
+        ::FileUtils.mkdir_p(playlist_dirname)
+      end
     end
 
     def find_segment_files(cue)
@@ -123,4 +144,4 @@ module WebVTT
       return filenames.map{|f| File.new(f) }
     end
   end
-end
+end

data/tests/parser.rb

@@ -20,6 +20,12 @@ class ParserTest < Minitest::Test
     }
   end
 
+  def test_can_create_empty_webvtt
+    webvtt = WebVTT::Blob.new
+    assert_equal 'WEBVTT', webvtt.header
+    assert_equal [], webvtt.cues
+  end
+
   def test_list_cues
     webvtt = WebVTT.read("tests/subtitles/test.vtt")
     assert_instance_of Array, webvtt.cues
@@ -43,6 +49,16 @@ class ParserTest < Minitest::Test
     assert_equal "English subtitle 15 -Forced- (00:00:27.000)\nline:75%", cue.text
   end
 
+  def test_cue_non_hours
+    webvtt = WebVTT.read("tests/subtitles/test_mmss_format.vtt")
+    cue = webvtt.cues[0]
+    assert_equal "00:00:29.000", cue.start.to_s
+    assert_equal "00:00:31.000", cue.end.to_s
+    assert_instance_of Hash, cue.style
+    assert_equal "75%", cue.style["line"]
+    assert_equal "English subtitle 15 -Forced- (00:00:27.000)\nline:75%", cue.text
+  end
+
   def test_cue_identifier
     webvtt = WebVTT.read("tests/subtitles/test.vtt")
     cue = webvtt.cues[1]
@@ -54,6 +70,11 @@ class ParserTest < Minitest::Test
     assert_equal "English subtitle 16 -Unforced- (00:00:31.000)\nalign:start line:0%", cue.text
   end
 
+  def test_multiple_line_separators
+    webvtt = WebVTT.read("tests/subtitles/test_multiple_line_separators.vtt")
+    assert_equal 2, webvtt.cues.length
+  end
+
   def test_ignore_if_note
     webvtt = WebVTT.read("tests/subtitles/withnote.vtt")
     assert_equal 3, webvtt.cues.size
@@ -129,6 +150,17 @@ The text should change)
     assert_equal 2, webvtt.cues.size
   end
 
+  def test_convert_weird_format_srt_to_webvtt
+    webvtt = WebVTT.convert_from_srt("tests/subtitles/weird_format.srt")
+    correct_webvtt = WebVTT.read("tests/subtitles/weird_format_corrected.vtt")
+
+    converted_cues = webvtt.cues.map { |cue| cue.start.to_s }
+    correct_cues = correct_webvtt.cues.map { |cue| cue.start.to_s }
+
+    assert_instance_of WebVTT::File, webvtt
+    assert_equal correct_cues, converted_cues
+  end
+
   def test_parse_big_file
     return
     webvtt = WebVTT.read("tests/subtitles/big_srt.vtt")
@@ -154,7 +186,7 @@ The text should change)
     assert_equal "00:09:02.373", webvtt.cues[1].end.to_s
     assert_equal "", webvtt.cues[1].text
   end
-  
+
   def test_cue_offset_by
     cue = WebVTT::Cue.parse <<-CUE
     00:00:01.000 --> 00:00:25.432
@@ -210,4 +242,18 @@ The text should change)
     webvtt = WebVTT.convert_from_srt("tests/subtitles/invalid_cue.srt")
     assert_equal 1, webvtt.cues.size
   end
+
+  def test_can_validate_webvtt_with_carriage_returns
+    webvtt = WebVTT::File.new("tests/subtitles/test_carriage_returns.vtt")
+    assert_instance_of Array, webvtt.cues
+    assert !webvtt.cues.empty?, "Cues should not be empty"
+    assert_instance_of WebVTT::Cue, webvtt.cues[0]
+    assert_equal 15, webvtt.cues.size
+  end
+
+  def test_invalid_vtt_without_milliseconds
+    assert_raises WebVTT::MalformedFile do
+      vtt = WebVTT::File.new('tests/subtitles/no_milliseconds.vtt')
+    end
+  end
 end

data/tests/segmenter.rb

@@ -43,7 +43,7 @@ class ParserTest < Minitest::Test
     subs = segmenter.split_to_files
     segmenter.generate_playlist(subs)
 
-    assert File.exists?("test.m3u8")
+    assert File.exist?("test.m3u8")
     # clean up
     subs.each {|f| FileUtils.rm(f.filename)}
     FileUtils.rm("test.m3u8")

data/tests/subtitles/test_carriage_returns.vtt

@@ -0,0 +1 @@
+WEBVTT

data/tests/subtitles/test_mmss_format.vtt

@@ -0,0 +1,9 @@
+WEBVTT
+
+00:29.000 --> 00:31.000 line:75%
+English subtitle 15 -Forced- (00:00:27.000)
+line:75%
+
+2
+00:31.000 --> 00:33.000 align:start line:0%
+ 

data/tests/subtitles/test_multiple_line_separators.vtt

@@ -0,0 +1,15 @@
+WEBVTT X-TIMESTAMP-MAP=MPEGTS:900000,LOCAL:00:00:00.000
+
+
+
+
+00:00:29.000 --> 00:00:31.000 line:75%
+English subtitle 15 -Forced- (00:00:27.000)
+line:75%
+
+
+
+
+00:00:31.000 --> 00:00:33.000 align:start line:0%
+English subtitle 16 -Unforced- (00:00:31.000)
+align:start line:0%

data/tests/subtitles/weird_format.srt

@@ -0,0 +1,15 @@
+1
+00:00:00,970 --> 00:00:03,000
+Normal formatting
+
+2
+00:00:4,080 --> 00:00:06,080
+Third digit
+
+3
+00:0:09,350 --> 00:00:13,350
+Second digit
+
+4
+0:00:15,000 --> 00:00:16,000
+First digit

data/tests/subtitles/weird_format.vtt

@@ -0,0 +1,17 @@
+WEBVTT
+
+1
+00:00:00.970 --> 00:00:03.000
+Normal formatting
+
+2
+00:00:04.080 --> 00:00:06.080
+Third digit
+
+3
+00:00:09.350 --> 00:00:13.350
+Second digit
+
+4
+00:00:15.000 --> 00:00:16.000
+First digit

data/tests/subtitles/weird_format_corrected.vtt

@@ -0,0 +1,17 @@
+WEBVTT
+
+1
+00:00:00.970 --> 00:00:03.000
+Normal formatting
+
+2
+00:00:04.080 --> 00:00:06.080
+Third digit
+
+3
+00:00:09.350 --> 00:00:13.350
+Second digit
+
+4
+00:00:15.000 --> 00:00:16.000
+First digit

data/webvtt-ruby.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name          = 'webvtt-ruby'
-  s.version       = '0.3.2'
+  s.version       = '0.4.2'
   s.summary       = "WebVTT parser and segmenter in ruby"
   s.description   = "WebVTT parser and segmenter in ruby for HTML5 and HTTP Live Streaming (HLS)."
   s.authors       = ["Bruno Celeste"]

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: webvtt-ruby
 version: !ruby/object:Gem::Version
-  version: 0.3.2
+  version: 0.4.2
 platform: ruby
 authors:
 - Bruno Celeste
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-11-09 00:00:00.000000000 Z
+date: 2024-05-17 00:00:00.000000000 Z
 dependencies: []
 description: WebVTT parser and segmenter in ruby for HTML5 and HTTP Live Streaming
   (HLS).
@@ -36,8 +36,14 @@ files:
 - tests/subtitles/no_text.vtt
 - tests/subtitles/notvalid.vtt
 - tests/subtitles/test.vtt
+- tests/subtitles/test_carriage_returns.vtt
 - tests/subtitles/test_from_srt.srt
 - tests/subtitles/test_from_srt.vtt
+- tests/subtitles/test_mmss_format.vtt
+- tests/subtitles/test_multiple_line_separators.vtt
+- tests/subtitles/weird_format.srt
+- tests/subtitles/weird_format.vtt
+- tests/subtitles/weird_format_corrected.vtt
 - tests/subtitles/withnote.vtt
 - webvtt-ruby.gemspec
 homepage: https://github.com/HeyWatch/webvtt-ruby
@@ -59,8 +65,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.4.3
+rubygems_version: 3.1.4
 signing_key: 
 specification_version: 4
 summary: WebVTT parser and segmenter in ruby