Class: SmTranscript::WrdReader

Inherits:
Object
  • Object
show all
Defined in:
lib/sm_transcript/wrd_reader.rb

Instance Attribute Summary collapse

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(src_file) ⇒ WrdReader

Returns a new instance of WrdReader.



21
22
23
24
25
26
 
# File 'lib/sm_transcript/wrd_reader.rb', line 21

def initialize(src_file)
  @metadata = {}
  @words = []
  parse_metadata()
  parse_words(src_file)
end

Instance Attribute Details

#metadata ⇒ Object (readonly)

Returns the value of attribute metadata.



13
14
15
 
# File 'lib/sm_transcript/wrd_reader.rb', line 13

def metadata
  @metadata
end

#words ⇒ Object (readonly)

Returns the value of attribute words.



14
15
16
 
# File 'lib/sm_transcript/wrd_reader.rb', line 14

def words
  @words
end

Class Method Details

.from_file(file_name) ⇒ Object 



16
17
18
19
 
# File 'lib/sm_transcript/wrd_reader.rb', line 16

def self.from_file(file_name)
  # p File.expand_path(file_name)
  new(File.open(file_name))
end

Instance Method Details

#parse_metadata ⇒ Object 



28
29
30
 
# File 'lib/sm_transcript/wrd_reader.rb', line 28

def parse_metadata()
  # there is currently no metadata in .wrd files
end

#parse_words(src_file) ⇒ Object 



32
33
34
35
36
37
38
39
40
41
42
43
 
# File 'lib/sm_transcript/wrd_reader.rb', line 32

def parse_words(src_file)
  src_file.each do |ln|
    # line is expected to contain two integers separated by a space, 
    # followed by a space and one or more words.  The words may contain
    # characters, an apostrophe, and, in some post-processed transcripts,
    # angle brackets. i.e. "<noise>"
    ln.scan(/^\d* \d* [\w'<>]* *[\w'<>]*/) do |t|
      arr = t.split
      @words << SmTranscript::Word.new(arr[0], arr[1], arr[1].to_i - arr[0].to_i, arr[2])
    end
  end
end