Class: EncodedWord

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

Instance Method Summary collapse

Constructor Details

#initialize(inputdir) ⇒ EncodedWord

Returns a new instance of EncodedWord.



7
8
9
10
11
12
13
 
# File 'lib/encoded_word.rb', line 7

def initialize(inputdir)
    if File::ALT_SEPARATOR
        @inputdir = inputdir.gsub(File::ALT_SEPARATOR, File::SEPARATOR)
    else
        @inputdir = inputdir
    end
end

Instance Method Details

#combine ⇒ Object 



20
21
22
 
# File 'lib/encoded_word.rb', line 20

def combine
    combine_all_mlog_plain
end

#combine_all_mlog_plain ⇒ Object 



24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
 
# File 'lib/encoded_word.rb', line 24

def combine_all_mlog_plain
    File.open(File.join(@inputdir, 'all_mlog.csv'), 'w:utf-8') do |out|
        out.puts '"key","","","date","from","to","cc","bcc","subject","attach"'
        Dir.glob(File.join(@inputdir, File.join('**', '*.plain'))).select do |f|
            puts f
            File.open(f, 'r:utf-8').each_line do |line|
                parts = mysplit(line)
                key = parts[3]
                date = parts[0]
                subject = parts[2]
                from = parts[3]
                to = to_csv_column(parts[4])
                cc = to_csv_column(parts[5])
                attach = to_csv_column(parts[6])
                out.puts %Q("#{key}","","","#{date}","#{from}","#{to}","#{cc}","","#{subject}","#{attach}")
            end
        end
    end
end

#concat_one_line(words) ⇒ Object 



88
89
90
91
92
 
# File 'lib/encoded_word.rb', line 88

def concat_one_line(words)
    line = words.join('')
    parts = mysplit(line, "\n")
    parts.join('')
end

#decode ⇒ Object 



15
16
17
18
 
# File 'lib/encoded_word.rb', line 15

def decode
    decode_all_mlog
    combine_all_mlog_plain
end

#decode_all_mlog ⇒ Object 



49
50
51
52
53
54
55
56
57
58
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
84
85
86
 
# File 'lib/encoded_word.rb', line 49

def decode_all_mlog
    Dir.glob(File.join(@inputdir, File.join('**', '*.mlog'))).select do |f|
        puts f
        mids = {}
        File.open(f + '.plain', 'w:utf-8') do |out|
            File.open(f) do |input|
                @input_enc = input.external_encoding
                input.each_line do |line|
                    parts = mysplit(line)
                    next if mids.has_key?(parts[1])
                    mids[parts[1]] = 0

                    newparts = []
                    newparts << format_date(parts[0]) #date
                    newparts << parts[1] #message-id
                    newparts << decode_subject(parts[2]) #subject
                    newparts << trim_emails(parts[3]) #from
                    newparts << trim_emails(parts[4]) #to
                    newparts << trim_emails(parts[5]) #cc
                    newparts << decode_attaches(parts)
                    0.upto(newparts.length-2) do |i|
                        begin
                            out.write newparts[i]
                            out.write "\t"
                        rescue
                            out.write "\t"
                        end
                    end
                    begin
                        out.puts newparts[newparts.length-1]
                    rescue
                        out.puts ''
                    end
                end
            end
        end
    end
end

#decode_attach(attach) ⇒ Object 



174
175
176
177
178
179
180
181
182
183
 
# File 'lib/encoded_word.rb', line 174

def decode_attach(attach)
    return '' unless attach and attach.length > 0
    parts = mysplit(attach, "\a")
    words = []
    parts.each do |p|
        wd = word_decode(p) 
        words << wd
    end
    concat_one_line(words)
end

#decode_attaches(parts) ⇒ Object 



165
166
167
168
169
170
171
172
 
# File 'lib/encoded_word.rb', line 165

def decode_attaches(parts)
    attaches = []
    6.upto(parts.length-1) do |i|
        attaches << decode_attach(parts[i])
    end
    return '' unless attaches.length > 0
    attaches.join("\a")
end

#decode_subject(sub) ⇒ Object 



123
124
125
126
127
128
129
130
131
132
 
# File 'lib/encoded_word.rb', line 123

def decode_subject(sub)
    return '' unless sub and sub.length > 0
    parts = mysplit(sub, "\a")
    words = []
    parts.each do |p|
        wd = word_decode(p) 
        words << wd
    end
    concat_one_line(words)
end

#format_date(engdate) ⇒ Object 



116
117
118
119
120
121
 
# File 'lib/encoded_word.rb', line 116

def format_date(engdate)
    return '' unless engdate and engdate.length > 0
    dt = DateTime.parse(engdate)
    dt = dt.new_offset('+0900')
    dt.strftime("%Y/%m/%d %H:%M:%S")
end

#getmail(line, at) ⇒ Object 



149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
 
# File 'lib/encoded_word.rb', line 149

def getmail(line, at)
    pos1 = 0
    pos2 = line.length - 1
    (at-1).step(0, -1) do |i|
        next if line[i] =~ /[\._a-zA-Z0-9-]/
        pos1 = i + 1
        break
    end
    (at+1).upto(line.length) do |i|
        next if line[i] =~ /[\.a-zA-Z0-9-]/
        pos2 = i - 1
        break
    end
    line[pos1..pos2]
end

#mysplit(line, sep = "\t") ⇒ Object 



94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
 
# File 'lib/encoded_word.rb', line 94

def mysplit(line, sep = "\t")
    return [] unless line
    return [] unless line.length > 0

    parts = []
    last = -1
    pos1 = -1
    while true do
        pos1 += 1
        pos2 = line.index(sep, pos1)
        if pos2
            parts << line[pos1...pos2]
            pos1 = pos2
        else
            last -= 1 if line[last] == "\n"
            parts << line[pos1..last]
            break
        end
    end
    return parts
end

#to_csv_column(str) ⇒ Object 



44
45
46
47
 
# File 'lib/encoded_word.rb', line 44

def to_csv_column(str)
    return '' unless str and str.length > 0
    str.gsub("\a", ';')
end

#trim_emails(emails) ⇒ Object 



134
135
136
137
138
139
140
141
142
143
144
145
146
147
 
# File 'lib/encoded_word.rb', line 134

def trim_emails(emails)
    return '' unless emails and emails.length > 0
    emails = emails.encode('utf-8', @input_enc, :undef=>:replace, :invalid=>:replace)
    pos1 = -1
    newparts = []
    while true do
        pos1 += 1
        pos2 = emails.index('@', pos1)
        break unless pos2
        newparts << getmail(emails, pos2)
        pos1 = pos2
    end
    newparts.join("\a")
end

#word_decode(input, out_charset = 'utf-8') ⇒ Object 



185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
 
# File 'lib/encoded_word.rb', line 185

def word_decode(input, out_charset = 'utf-8')
    u8 = input.encode('utf-8', @u8_enc, :undef=>:replace, :invalid=>:replace)
    parts = u8.scan(/=\?([A-Za-z0-9_-]+)\?([BQbq])\?([^\?]+)\?=/).first
    return input unless parts and parts.length == 3
    charset = parts[0]
    enc = parts[1].upcase
    charset = 'utf-8' if charset.downcase == 'utf8'
    wd = parts[2].unpack({ "B"=>"m*", "Q"=>"M*" }[enc]).first
    begin
        return wd.encode(out_charset, charset, :undef=>:replace, :invalid=>:replace)
    rescue => e
        print "Cannot encode #{input} because #{e}, try iconv..."
        begin
            puts "OK"
            return Iconv.conv(out_charset + "//IGNORE", charset, wd)
        rescue => e
            puts "Iconv failed too because #{e}. Return ''."
            return ''
        end
    end
end