Class: Stevedore::ArchiveSplitter

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

Constant Summary collapse

HANDLED_FORMATS = 
["zip", "mbox", "pst", "eml"]

Class Method Summary collapse

Class Method Details

.get_attachments_from_eml(email_filename) ⇒ Object 



112
113
114
115
116
117
118
119
120
121
122
 
# File 'lib/split_archive.rb', line 112

def self.get_attachments_from_eml(email_filename)
  Enumerator.new do |yielder|
    mail = Mail.new open(email_filename){|f| f.read }
    attachment_results = mail.attachments.map do |attachment|
      [attachment.filename, lambda{|fn| open(fn, 'wb'){|fh| fh << attachment.body.decoded }}, [], File.basename(email_filename)]
    end
    attachment_basenames = attachment_results.map{|a| File.basename(a[0]) }
    yielder << [File.basename(email_filename), lambda{|fn| open(fn, 'wb'){|fh| fh << open(email_filename){|f| f.read } } }, attachment_basenames, nil]
    attachment_results.each{|res| yielder << res }
  end
end

.split(archive_filename) ⇒ Object 



20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
 
# File 'lib/split_archive.rb', line 20

def self.split(archive_filename)
  # if it's a PST use split_pst
  # if it's an mbox, use split_mbox, etc.
  # return a list of files
  Enumerator.new do |yielder|
    Dir.mktmpdir do |tmpdir|
      #TODO should probably do magic byte searching etc.
      extension = archive_filename.split(".")[-1]
      puts "splitting #{archive_filename}"
      constituent_files =  if extension == "mbox"
                      self.split_mbox(archive_filename)
                    elsif extension == "pst"
                      self.split_pst(archive_filename)
                    elsif extension == "zip"
                      self.split_zip(archive_filename)
                    elsif extension == "eml"
                      self.get_attachments_from_eml(archive_filename)                          
                    end
      # should yield a relative filename
      # and a lambda that will write the file contents to the given filename
      FileUtils.mkdir_p(File.join(tmpdir, File.basename(archive_filename)))

      constituent_files.each_with_index do |basename_contents_lambda_attachments_parent, idx|
        basename, contents_lambda, attachments, parent = *basename_contents_lambda_attachments_parent
        tmp_filename = File.join(tmpdir, File.basename(archive_filename), basename )
        FileUtils.mkdir_p(File.dirname(tmp_filename))
        begin
          contents_lambda.call(tmp_filename)
        rescue Errno::ENOENT
          puts "#{tmp_filename} wasn't extracted from #{archive_filename}"
          next
        end
        attachments ||= []
        yielder.yield tmp_filename, File.join(File.basename(archive_filename), basename), attachments, parent
      end
    end
  end
end

.split_mbox(archive_filename) ⇒ Object 



124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
 
# File 'lib/split_archive.rb', line 124

def self.split_mbox(archive_filename)
  # stolen shamelessly from the Ruby Enumerable docs, actually
  # split mails in mbox (slice before Unix From line after an empty line)
  Enumerator.new do |yielder|
    open(archive_filename) do |fh|
      fh.slice_before(empty: true) do |line, h|
        previous_was_empty = h[:empty]
        h[:empty] = line == "\n" || line == "\r\n" || line == "\r"
        previous_was_empty && line.start_with?("From ")
      end.each_with_index do |mail_str, idx| 
        # TODO copy over stuff from get_attachments_from_eml for attachment/parents if 
        mail_str.pop if mail_str.last == "\n" # remove last line if prexent
        yielder << ["#{idx}.eml", lambda{|fn| open(fn, 'wb'){|fh| fh << mail_str.join("") } }]
        mail = Mail.new mail_str.join("")
        mail.attachments.each do |attachment|
          yielder << [attachment.filename, lambda{|fn| open(fn, 'wb'){|fh| fh << attachment.body.decoded }}]
        end
      end
    end
  end
end

.split_pst(archive_filename) ⇒ Object 



59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
 
# File 'lib/split_archive.rb', line 59

def self.split_pst(archive_filename)
  pstfile = Java::ComPFF::PSTFile.new(archive_filename)
  idx = 0
  folders = pstfile.root.sub_folders.inject({}) do |memo,f|
    memo[f.name] = f
    memo
  end
  Enumerator.new do |yielder|
    folders.each do |folder_name, folder|
      while mail = folder.getNextChild

        eml_str = mail.get_transport_message_headers + mail.get_body

        yielder << ["#{idx}.eml", lambda{|fn| open(fn, 'wb'){|fh| fh << eml_str } }]
        attachment_count = mail.get_number_of_attachments
        attachment_count.times do |attachment_idx|
          attachment = mail.get_attachment(attachment_idx)
          attachment_filename = attachment.get_filename
          yielder << ["#{idx}-#{attachment_filename}", lambda {|fn| open(fn, 'wb'){ |fh| fh << attachment.get_file_input_stream.to_io.read }}]
        end
        idx += 1
      end
    end
  end
end

.split_zip(archive_filename) ⇒ Object 



146
147
148
149
150
151
152
153
154
155
156
157
158
 
# File 'lib/split_archive.rb', line 146

def self.split_zip(archive_filename)
  Zip::File.open(archive_filename) do |zip_file|
    Enumerator.new do |yielder|
      zip_file.each do |entry|
        begin
          yielder << [entry.name, lambda{|fn| entry.extract(fn) }]
        rescue
          puts "unable to extract #{entry.name} from #{archive_filename}"
        end
      end
    end
  end
end