tmail_es 1.2.7.2
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/CHANGES +83 -0
- data/LICENSE +21 -0
- data/NOTES +7 -0
- data/README +182 -0
- data/Rakefile +2 -0
- data/ext/Makefile +20 -0
- data/ext/tmailscanner/tmail/MANIFEST +4 -0
- data/ext/tmailscanner/tmail/depend +1 -0
- data/ext/tmailscanner/tmail/extconf.rb +33 -0
- data/ext/tmailscanner/tmail/tmailscanner.c +614 -0
- data/lib/tmail/Makefile +18 -0
- data/lib/tmail/address.rb +392 -0
- data/lib/tmail/attachments.rb +65 -0
- data/lib/tmail/base64.rb +46 -0
- data/lib/tmail/compat.rb +41 -0
- data/lib/tmail/config.rb +67 -0
- data/lib/tmail/core_extensions.rb +63 -0
- data/lib/tmail/encode.rb +590 -0
- data/lib/tmail/header.rb +962 -0
- data/lib/tmail/index.rb +9 -0
- data/lib/tmail/interface.rb +1162 -0
- data/lib/tmail/loader.rb +3 -0
- data/lib/tmail/mail.rb +578 -0
- data/lib/tmail/mailbox.rb +496 -0
- data/lib/tmail/main.rb +6 -0
- data/lib/tmail/mbox.rb +3 -0
- data/lib/tmail/net.rb +250 -0
- data/lib/tmail/obsolete.rb +132 -0
- data/lib/tmail/parser.rb +1060 -0
- data/lib/tmail/parser.y +416 -0
- data/lib/tmail/port.rb +379 -0
- data/lib/tmail/quoting.rb +164 -0
- data/lib/tmail/require_arch.rb +58 -0
- data/lib/tmail/scanner.rb +49 -0
- data/lib/tmail/scanner_r.rb +261 -0
- data/lib/tmail/stringio.rb +280 -0
- data/lib/tmail/utils.rb +361 -0
- data/lib/tmail/vendor/rchardet-1.3/COPYING +504 -0
- data/lib/tmail/vendor/rchardet-1.3/README +12 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/big5freq.rb +927 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/big5prober.rb +42 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/chardistribution.rb +238 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/charsetgroupprober.rb +112 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/charsetprober.rb +75 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/codingstatemachine.rb +64 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/constants.rb +42 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/escprober.rb +89 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/escsm.rb +244 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/eucjpprober.rb +88 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/euckrfreq.rb +596 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/euckrprober.rb +42 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/euctwfreq.rb +430 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/euctwprober.rb +42 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/gb2312freq.rb +474 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/gb2312prober.rb +42 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/hebrewprober.rb +289 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/jisfreq.rb +570 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/jpcntx.rb +229 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/langbulgarianmodel.rb +229 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/langcyrillicmodel.rb +330 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/langgreekmodel.rb +227 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/langhebrewmodel.rb +202 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/langhungarianmodel.rb +226 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/langthaimodel.rb +201 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/latin1prober.rb +147 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/mbcharsetprober.rb +89 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/mbcsgroupprober.rb +45 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/mbcssm.rb +542 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/sbcharsetprober.rb +124 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/sbcsgroupprober.rb +56 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/sjisprober.rb +88 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/universaldetector.rb +167 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet/utf8prober.rb +87 -0
- data/lib/tmail/vendor/rchardet-1.3/lib/rchardet.rb +67 -0
- data/lib/tmail/version.rb +40 -0
- data/lib/tmail.rb +6 -0
- data/setup.rb +1482 -0
- data/test/extctrl.rb +6 -0
- data/test/fixtures/apple_unquoted_content_type +44 -0
- data/test/fixtures/inline_attachment.txt +2095 -0
- data/test/fixtures/iso_8859_1_email_without_encoding_and_message_id.txt +16 -0
- data/test/fixtures/mailbox +414 -0
- data/test/fixtures/mailbox.zip +0 -0
- data/test/fixtures/mailbox_without_any_from_or_sender +10 -0
- data/test/fixtures/mailbox_without_from +11 -0
- data/test/fixtures/mailbox_without_return_path +12 -0
- data/test/fixtures/marked_as_iso_8859_1_but_it_is_utf_8.txt +33 -0
- data/test/fixtures/marked_as_utf_8_but_it_is_iso_8859_1.txt +56 -0
- data/test/fixtures/raw_attack_email_with_zero_length_whitespace +29 -0
- data/test/fixtures/raw_base64_decoded_string +0 -0
- data/test/fixtures/raw_base64_email +83 -0
- data/test/fixtures/raw_base64_encoded_string +1 -0
- data/test/fixtures/raw_email +14 -0
- data/test/fixtures/raw_email10 +20 -0
- data/test/fixtures/raw_email11 +34 -0
- data/test/fixtures/raw_email12 +32 -0
- data/test/fixtures/raw_email13 +29 -0
- data/test/fixtures/raw_email2 +114 -0
- data/test/fixtures/raw_email3 +70 -0
- data/test/fixtures/raw_email4 +59 -0
- data/test/fixtures/raw_email5 +19 -0
- data/test/fixtures/raw_email6 +20 -0
- data/test/fixtures/raw_email7 +66 -0
- data/test/fixtures/raw_email8 +47 -0
- data/test/fixtures/raw_email9 +28 -0
- data/test/fixtures/raw_email_bad_time +62 -0
- data/test/fixtures/raw_email_double_at_in_header +14 -0
- data/test/fixtures/raw_email_multiple_from +30 -0
- data/test/fixtures/raw_email_only_attachment +17 -0
- data/test/fixtures/raw_email_quoted_with_0d0a +14 -0
- data/test/fixtures/raw_email_reply +32 -0
- data/test/fixtures/raw_email_simple +11 -0
- data/test/fixtures/raw_email_string_in_date_field +17 -0
- data/test/fixtures/raw_email_trailing_dot +21 -0
- data/test/fixtures/raw_email_with_bad_date +48 -0
- data/test/fixtures/raw_email_with_illegal_boundary +58 -0
- data/test/fixtures/raw_email_with_mimepart_without_content_type +94 -0
- data/test/fixtures/raw_email_with_multipart_mixed_quoted_boundary +50 -0
- data/test/fixtures/raw_email_with_nested_attachment +100 -0
- data/test/fixtures/raw_email_with_partially_quoted_subject +14 -0
- data/test/fixtures/raw_email_with_quoted_attachment_filename +60 -0
- data/test/fixtures/raw_email_with_quoted_illegal_boundary +58 -0
- data/test/fixtures/raw_email_with_wrong_splitted_multibyte_encoded_word_subject +15 -0
- data/test/fixtures/the_only_part_is_a_word_document.txt +425 -0
- data/test/fixtures/unquoted_filename_in_attachment +177 -0
- data/test/kcode.rb +14 -0
- data/test/temp_test_one.rb +46 -0
- data/test/test_address.rb +1216 -0
- data/test/test_attachments.rb +133 -0
- data/test/test_base64.rb +64 -0
- data/test/test_encode.rb +139 -0
- data/test/test_header.rb +1021 -0
- data/test/test_helper.rb +9 -0
- data/test/test_mail.rb +756 -0
- data/test/test_mbox.rb +184 -0
- data/test/test_port.rb +440 -0
- data/test/test_quote.rb +107 -0
- data/test/test_scanner.rb +209 -0
- data/test/test_utils.rb +36 -0
- data/tmail_es.gemspec +35 -0
- metadata +257 -0
@@ -0,0 +1,56 @@
|
|
1
|
+
######################## BEGIN LICENSE BLOCK ########################
|
2
|
+
# The Original Code is Mozilla Universal charset detector code.
|
3
|
+
#
|
4
|
+
# The Initial Developer of the Original Code is
|
5
|
+
# Netscape Communications Corporation.
|
6
|
+
# Portions created by the Initial Developer are Copyright (C) 2001
|
7
|
+
# the Initial Developer. All Rights Reserved.
|
8
|
+
#
|
9
|
+
# Contributor(s):
|
10
|
+
# Jeff Hodges - port to Ruby
|
11
|
+
# Mark Pilgrim - port to Python
|
12
|
+
# Shy Shalom - original C code
|
13
|
+
#
|
14
|
+
# This library is free software; you can redistribute it and/or
|
15
|
+
# modify it under the terms of the GNU Lesser General Public
|
16
|
+
# License as published by the Free Software Foundation; either
|
17
|
+
# version 2.1 of the License, or (at your option) any later version.
|
18
|
+
#
|
19
|
+
# This library is distributed in the hope that it will be useful,
|
20
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
21
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
22
|
+
# Lesser General Public License for more details.
|
23
|
+
#
|
24
|
+
# You should have received a copy of the GNU Lesser General Public
|
25
|
+
# License along with this library; if not, write to the Free Software
|
26
|
+
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
|
27
|
+
# 02110-1301 USA
|
28
|
+
######################### END LICENSE BLOCK #########################
|
29
|
+
|
30
|
+
module CharDet
|
31
|
+
class SBCSGroupProber < CharSetGroupProber
|
32
|
+
def initialize
|
33
|
+
super
|
34
|
+
@_mProbers = [ SingleByteCharSetProber.new(Win1251CyrillicModel),
|
35
|
+
SingleByteCharSetProber.new(Koi8rModel),
|
36
|
+
SingleByteCharSetProber.new(Latin5CyrillicModel),
|
37
|
+
SingleByteCharSetProber.new(MacCyrillicModel),
|
38
|
+
SingleByteCharSetProber.new(Ibm866Model),
|
39
|
+
SingleByteCharSetProber.new(Ibm855Model),
|
40
|
+
SingleByteCharSetProber.new(Latin7GreekModel),
|
41
|
+
SingleByteCharSetProber.new(Win1253GreekModel),
|
42
|
+
SingleByteCharSetProber.new(Latin5BulgarianModel),
|
43
|
+
SingleByteCharSetProber.new(Win1251BulgarianModel),
|
44
|
+
SingleByteCharSetProber.new(Latin2HungarianModel),
|
45
|
+
SingleByteCharSetProber.new(Win1250HungarianModel),
|
46
|
+
SingleByteCharSetProber.new(TIS620ThaiModel) ]
|
47
|
+
hebrewProber = HebrewProber.new()
|
48
|
+
logicalHebrewProber = SingleByteCharSetProber.new(Win1255HebrewModel, false, hebrewProber)
|
49
|
+
visualHebrewProber = SingleByteCharSetProber.new(Win1255HebrewModel, true, hebrewProber)
|
50
|
+
hebrewProber.set_model_probers(logicalHebrewProber, visualHebrewProber)
|
51
|
+
@_mProbers += [hebrewProber, logicalHebrewProber, visualHebrewProber]
|
52
|
+
|
53
|
+
reset()
|
54
|
+
end
|
55
|
+
end
|
56
|
+
end
|
@@ -0,0 +1,88 @@
|
|
1
|
+
######################## BEGIN LICENSE BLOCK ########################
|
2
|
+
# The Original Code is mozilla.org code.
|
3
|
+
#
|
4
|
+
# The Initial Developer of the Original Code is
|
5
|
+
# Netscape Communications Corporation.
|
6
|
+
# Portions created by the Initial Developer are Copyright (C) 1998
|
7
|
+
# the Initial Developer. All Rights Reserved.
|
8
|
+
#
|
9
|
+
# Contributor(s):
|
10
|
+
# Jeff Hodges - port to Ruby
|
11
|
+
# Mark Pilgrim - port to Python
|
12
|
+
#
|
13
|
+
# This library is free software; you can redistribute it and/or
|
14
|
+
# modify it under the terms of the GNU Lesser General Public
|
15
|
+
# License as published by the Free Software Foundation; either
|
16
|
+
# version 2.1 of the License, or (at your option) any later version.
|
17
|
+
#
|
18
|
+
# This library is distributed in the hope that it will be useful,
|
19
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
20
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
21
|
+
# Lesser General Public License for more details.
|
22
|
+
#
|
23
|
+
# You should have received a copy of the GNU Lesser General Public
|
24
|
+
# License along with this library; if not, write to the Free Software
|
25
|
+
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
|
26
|
+
# 02110-1301 USA
|
27
|
+
######################### END LICENSE BLOCK #########################
|
28
|
+
|
29
|
+
module CharDet
|
30
|
+
class SJISProber < MultiByteCharSetProber
|
31
|
+
def initialize
|
32
|
+
super()
|
33
|
+
@_mCodingSM = CodingStateMachine.new(SJISSMModel)
|
34
|
+
@_mDistributionAnalyzer = SJISDistributionAnalysis.new()
|
35
|
+
@_mContextAnalyzer = SJISContextAnalysis.new()
|
36
|
+
reset()
|
37
|
+
end
|
38
|
+
|
39
|
+
def reset
|
40
|
+
super()
|
41
|
+
@_mContextAnalyzer.reset()
|
42
|
+
end
|
43
|
+
|
44
|
+
def get_charset_name
|
45
|
+
return "SHIFT_JIS"
|
46
|
+
end
|
47
|
+
|
48
|
+
def feed(aBuf)
|
49
|
+
aLen = aBuf.length
|
50
|
+
for i in (0...aLen)
|
51
|
+
codingState = @_mCodingSM.next_state(aBuf[i..i])
|
52
|
+
if codingState == EError
|
53
|
+
$stderr << "#{get_charset_name} prober hit error at byte #{i}\n" if $debug
|
54
|
+
@_mState = ENotMe
|
55
|
+
break
|
56
|
+
elsif codingState == EItsMe
|
57
|
+
@_mState = EFoundIt
|
58
|
+
break
|
59
|
+
elsif codingState == EStart
|
60
|
+
charLen = @_mCodingSM.get_current_charlen()
|
61
|
+
if i == 0
|
62
|
+
@_mLastChar[1] = aBuf[0..0]
|
63
|
+
@_mContextAnalyzer.feed(@_mLastChar[2 - charLen..-1], charLen)
|
64
|
+
@_mDistributionAnalyzer.feed(@_mLastChar, charLen)
|
65
|
+
else
|
66
|
+
@_mContextAnalyzer.feed(aBuf[i + 1 - charLen ... i + 3 - charLen], charLen)
|
67
|
+
@_mDistributionAnalyzer.feed(aBuf[i - 1 ... i + 1], charLen)
|
68
|
+
end
|
69
|
+
end
|
70
|
+
end
|
71
|
+
|
72
|
+
@_mLastChar[0] = aBuf[aLen - 1.. aLen-1]
|
73
|
+
|
74
|
+
if get_state() == EDetecting
|
75
|
+
if @_mContextAnalyzer.got_enough_data() and (get_confidence() > SHORTCUT_THRESHOLD)
|
76
|
+
@_mState = EFoundIt
|
77
|
+
end
|
78
|
+
end
|
79
|
+
|
80
|
+
return get_state()
|
81
|
+
end
|
82
|
+
|
83
|
+
def get_confidence
|
84
|
+
l = [@_mContextAnalyzer.get_confidence(), @_mDistributionAnalyzer.get_confidence()]
|
85
|
+
return l.max
|
86
|
+
end
|
87
|
+
end
|
88
|
+
end
|
@@ -0,0 +1,167 @@
|
|
1
|
+
######################## BEGIN LICENSE BLOCK ########################
|
2
|
+
# The Original Code is Mozilla Universal charset detector code.
|
3
|
+
#
|
4
|
+
# The Initial Developer of the Original Code is
|
5
|
+
# Netscape Communications Corporation.
|
6
|
+
# Portions created by the Initial Developer are Copyright (C) 2001
|
7
|
+
# the Initial Developer. All Rights Reserved.
|
8
|
+
#
|
9
|
+
# Contributor(s):
|
10
|
+
# Jeff Hodges - port to Ruby
|
11
|
+
# Mark Pilgrim - port to Python
|
12
|
+
# Shy Shalom - original C code
|
13
|
+
#
|
14
|
+
# This library is free software; you can redistribute it and/or
|
15
|
+
# modify it under the terms of the GNU Lesser General Public
|
16
|
+
# License as published by the Free Software Foundation; either
|
17
|
+
# version 2.1 of the License, or (at your option) any later version.
|
18
|
+
#
|
19
|
+
# This library is distributed in the hope that it will be useful,
|
20
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
21
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
22
|
+
# Lesser General Public License for more details.
|
23
|
+
#
|
24
|
+
# You should have received a copy of the GNU Lesser General Public
|
25
|
+
# License along with this library; if not, write to the Free Software
|
26
|
+
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
|
27
|
+
# 02110-1301 USA
|
28
|
+
######################### END LICENSE BLOCK #########################
|
29
|
+
|
30
|
+
module CharDet
|
31
|
+
MINIMUM_THRESHOLD = 0.20
|
32
|
+
EPureAscii = 0
|
33
|
+
EEscAscii = 1
|
34
|
+
EHighbyte = 2
|
35
|
+
|
36
|
+
class UniversalDetector
|
37
|
+
attr_accessor :result
|
38
|
+
def initialize
|
39
|
+
@_highBitDetector = /[\x80-\xFF]/
|
40
|
+
@_escDetector = /(\033|\~\{)/
|
41
|
+
@_mEscCharSetProber = nil
|
42
|
+
@_mCharSetProbers = []
|
43
|
+
reset()
|
44
|
+
end
|
45
|
+
|
46
|
+
def reset
|
47
|
+
@result = {'encoding' => nil, 'confidence' => 0.0}
|
48
|
+
@done = false
|
49
|
+
@_mStart = true
|
50
|
+
@_mGotData = false
|
51
|
+
@_mInputState = EPureAscii
|
52
|
+
@_mLastChar = ''
|
53
|
+
if @_mEscCharSetProber
|
54
|
+
@_mEscCharSetProber.reset()
|
55
|
+
end
|
56
|
+
for prober in @_mCharSetProbers
|
57
|
+
prober.reset()
|
58
|
+
end
|
59
|
+
end
|
60
|
+
|
61
|
+
def feed(aBuf)
|
62
|
+
return if @done
|
63
|
+
|
64
|
+
aLen = aBuf.length
|
65
|
+
return if not aLen
|
66
|
+
|
67
|
+
if not @_mGotData
|
68
|
+
# If the data starts with BOM, we know it is UTF
|
69
|
+
if aBuf[0...3] == "\xEF\xBB\xBF"
|
70
|
+
# EF BB BF UTF-8 with BOM
|
71
|
+
@result = {'encoding' => "UTF-8", 'confidence' => 1.0}
|
72
|
+
elsif aBuf[0...4] == "\xFF\xFE\x00\x00"
|
73
|
+
# FF FE 00 00 UTF-32, little-endian BOM
|
74
|
+
@result = {'encoding' => "UTF-32LE", 'confidence' => 1.0}
|
75
|
+
elsif aBuf[0...4] == "\x00\x00\xFE\xFF"
|
76
|
+
# 00 00 FE FF UTF-32, big-endian BOM
|
77
|
+
@result = {'encoding' => "UTF-32BE", 'confidence' => 1.0}
|
78
|
+
elsif aBuf[0...4] == "\xFE\xFF\x00\x00"
|
79
|
+
# FE FF 00 00 UCS-4, unusual octet order BOM (3412)
|
80
|
+
@result = {'encoding' => "X-ISO-10646-UCS-4-3412", 'confidence' => 1.0}
|
81
|
+
elsif aBuf[0...4] == "\x00\x00\xFF\xFE"
|
82
|
+
# 00 00 FF FE UCS-4, unusual octet order BOM (2143)
|
83
|
+
@result = {'encoding' => "X-ISO-10646-UCS-4-2143", 'confidence' => 1.0}
|
84
|
+
elsif aBuf[0...2] == "\xFF\xFE"
|
85
|
+
# FF FE UTF-16, little endian BOM
|
86
|
+
@result = {'encoding' => "UTF-16LE", 'confidence' => 1.0}
|
87
|
+
elsif aBuf[0...2] == "\xFE\xFF"
|
88
|
+
# FE FF UTF-16, big endian BOM
|
89
|
+
@result = {'encoding' => "UTF-16BE", 'confidence' => 1.0}
|
90
|
+
end
|
91
|
+
end
|
92
|
+
|
93
|
+
@_mGotData = true
|
94
|
+
if @result['encoding'] and (@result['confidence'] > 0.0)
|
95
|
+
@done = true
|
96
|
+
return
|
97
|
+
end
|
98
|
+
|
99
|
+
if @_mInputState == EPureAscii
|
100
|
+
if @_highBitDetector =~ (aBuf)
|
101
|
+
@_mInputState = EHighbyte
|
102
|
+
elsif (@_mInputState == EPureAscii) and @_escDetector =~ (@_mLastChar + aBuf)
|
103
|
+
@_mInputState = EEscAscii
|
104
|
+
end
|
105
|
+
end
|
106
|
+
|
107
|
+
@_mLastChar = aBuf[-1..-1]
|
108
|
+
if @_mInputState == EEscAscii
|
109
|
+
if not @_mEscCharSetProber
|
110
|
+
@_mEscCharSetProber = EscCharSetProber.new()
|
111
|
+
end
|
112
|
+
if @_mEscCharSetProber.feed(aBuf) == EFoundIt
|
113
|
+
@result = {'encoding' => self._mEscCharSetProber.get_charset_name(),
|
114
|
+
'confidence' => @_mEscCharSetProber.get_confidence()
|
115
|
+
}
|
116
|
+
@done = true
|
117
|
+
end
|
118
|
+
elsif @_mInputState == EHighbyte
|
119
|
+
if not @_mCharSetProbers or @_mCharSetProbers.empty?
|
120
|
+
@_mCharSetProbers = [MBCSGroupProber.new(), SBCSGroupProber.new(), Latin1Prober.new()]
|
121
|
+
end
|
122
|
+
for prober in @_mCharSetProbers
|
123
|
+
if prober.feed(aBuf) == EFoundIt
|
124
|
+
@result = {'encoding' => prober.get_charset_name(),
|
125
|
+
'confidence' => prober.get_confidence()}
|
126
|
+
@done = true
|
127
|
+
break
|
128
|
+
end
|
129
|
+
end
|
130
|
+
end
|
131
|
+
|
132
|
+
end
|
133
|
+
|
134
|
+
def close
|
135
|
+
return if @done
|
136
|
+
if not @_mGotData
|
137
|
+
$stderr << "no data received!\n" if $debug
|
138
|
+
return
|
139
|
+
end
|
140
|
+
@done = true
|
141
|
+
|
142
|
+
if @_mInputState == EPureAscii
|
143
|
+
@result = {'encoding' => 'ascii', 'confidence' => 1.0}
|
144
|
+
return @result
|
145
|
+
end
|
146
|
+
|
147
|
+
if @_mInputState == EHighbyte
|
148
|
+
confidences = {}
|
149
|
+
@_mCharSetProbers.each{ |prober| confidences[prober] = prober.get_confidence }
|
150
|
+
maxProber = @_mCharSetProbers.max{ |a,b| confidences[a] <=> confidences[b] }
|
151
|
+
if maxProber and maxProber.get_confidence > MINIMUM_THRESHOLD
|
152
|
+
@result = {'encoding' => maxProber.get_charset_name(),
|
153
|
+
'confidence' => maxProber.get_confidence()}
|
154
|
+
return @result
|
155
|
+
end
|
156
|
+
end
|
157
|
+
|
158
|
+
if $debug
|
159
|
+
$stderr << "no probers hit minimum threshhold\n" if $debug
|
160
|
+
for prober in @_mCharSetProbers[0]._mProbers
|
161
|
+
next if not prober
|
162
|
+
$stderr << "#{prober.get_charset_name} confidence = #{prober.get_confidence}\n" if $debug
|
163
|
+
end
|
164
|
+
end
|
165
|
+
end
|
166
|
+
end
|
167
|
+
end
|
@@ -0,0 +1,87 @@
|
|
1
|
+
######################## BEGIN LICENSE BLOCK ########################
|
2
|
+
# The Original Code is mozilla.org code.
|
3
|
+
#
|
4
|
+
# The Initial Developer of the Original Code is
|
5
|
+
# Netscape Communications Corporation.
|
6
|
+
# Portions created by the Initial Developer are Copyright (C) 1998
|
7
|
+
# the Initial Developer. All Rights Reserved.
|
8
|
+
#
|
9
|
+
# Contributor(s):
|
10
|
+
# Jeff Hodges - port to Ruby
|
11
|
+
# Mark Pilgrim - port to Python
|
12
|
+
#
|
13
|
+
# This library is free software; you can redistribute it and/or
|
14
|
+
# modify it under the terms of the GNU Lesser General Public
|
15
|
+
# License as published by the Free Software Foundation; either
|
16
|
+
# version 2.1 of the License, or (at your option) any later version.
|
17
|
+
#
|
18
|
+
# This library is distributed in the hope that it will be useful,
|
19
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
20
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
21
|
+
# Lesser General Public License for more details.
|
22
|
+
#
|
23
|
+
# You should have received a copy of the GNU Lesser General Public
|
24
|
+
# License along with this library; if not, write to the Free Software
|
25
|
+
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
|
26
|
+
# 02110-1301 USA
|
27
|
+
######################### END LICENSE BLOCK #########################
|
28
|
+
|
29
|
+
module CharDet
|
30
|
+
ONE_CHAR_PROB = 0.5
|
31
|
+
|
32
|
+
class UTF8Prober < CharSetProber
|
33
|
+
def initialize
|
34
|
+
super()
|
35
|
+
@_mCodingSM = CodingStateMachine.new(UTF8SMModel)
|
36
|
+
reset()
|
37
|
+
end
|
38
|
+
|
39
|
+
def reset
|
40
|
+
super()
|
41
|
+
@_mCodingSM.reset()
|
42
|
+
@_mNumOfMBChar = 0
|
43
|
+
end
|
44
|
+
|
45
|
+
def get_charset_name
|
46
|
+
return "utf-8"
|
47
|
+
end
|
48
|
+
|
49
|
+
def feed(aBuf)
|
50
|
+
aBuf.each_byte do |b|
|
51
|
+
c = b.chr
|
52
|
+
codingState = @_mCodingSM.next_state(c)
|
53
|
+
if codingState == EError
|
54
|
+
@_mState = ENotMe
|
55
|
+
break
|
56
|
+
elsif codingState == EItsMe
|
57
|
+
@_mState = EFoundIt
|
58
|
+
break
|
59
|
+
elsif codingState == EStart
|
60
|
+
if @_mCodingSM.get_current_charlen() >= 2
|
61
|
+
@_mNumOfMBChar += 1
|
62
|
+
end
|
63
|
+
end
|
64
|
+
end
|
65
|
+
|
66
|
+
if get_state() == EDetecting
|
67
|
+
if get_confidence() > SHORTCUT_THRESHOLD
|
68
|
+
@_mState = EFoundIt
|
69
|
+
end
|
70
|
+
end
|
71
|
+
|
72
|
+
return get_state()
|
73
|
+
end
|
74
|
+
|
75
|
+
def get_confidence
|
76
|
+
unlike = 0.99
|
77
|
+
if @_mNumOfMBChar < 6
|
78
|
+
for i in (0...@_mNumOfMBChar)
|
79
|
+
unlike = unlike * ONE_CHAR_PROB
|
80
|
+
end
|
81
|
+
return 1.0 - unlike
|
82
|
+
else
|
83
|
+
return unlike
|
84
|
+
end
|
85
|
+
end
|
86
|
+
end
|
87
|
+
end
|
@@ -0,0 +1,67 @@
|
|
1
|
+
######################## BEGIN LICENSE BLOCK ########################
|
2
|
+
# This library is free software; you can redistribute it and/or
|
3
|
+
# modify it under the terms of the GNU Lesser General Public
|
4
|
+
# License as published by the Free Software Foundation; either
|
5
|
+
# version 2.1 of the License, or (at your option) any later version.
|
6
|
+
#
|
7
|
+
# This library is distributed in the hope that it will be useful,
|
8
|
+
# but WITHOUT ANY WARRANTY; without even the implied warranty of
|
9
|
+
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
10
|
+
# Lesser General Public License for more details.
|
11
|
+
#
|
12
|
+
# You should have received a copy of the GNU Lesser General Public
|
13
|
+
# License along with this library; if not, write to the Free Software
|
14
|
+
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA
|
15
|
+
# 02110-1301 USA
|
16
|
+
######################### END LICENSE BLOCK #########################
|
17
|
+
|
18
|
+
$LOAD_PATH.unshift File.expand_path(File.dirname(__FILE__))
|
19
|
+
|
20
|
+
require 'rchardet/charsetprober'
|
21
|
+
require 'rchardet/mbcharsetprober'
|
22
|
+
|
23
|
+
require 'rchardet/big5freq'
|
24
|
+
require 'rchardet/big5prober'
|
25
|
+
require 'rchardet/chardistribution'
|
26
|
+
require 'rchardet/charsetgroupprober'
|
27
|
+
|
28
|
+
require 'rchardet/codingstatemachine'
|
29
|
+
require 'rchardet/constants'
|
30
|
+
require 'rchardet/escprober'
|
31
|
+
require 'rchardet/escsm'
|
32
|
+
require 'rchardet/eucjpprober'
|
33
|
+
require 'rchardet/euckrfreq'
|
34
|
+
require 'rchardet/euckrprober'
|
35
|
+
require 'rchardet/euctwfreq'
|
36
|
+
require 'rchardet/euctwprober'
|
37
|
+
require 'rchardet/gb2312freq'
|
38
|
+
require 'rchardet/gb2312prober'
|
39
|
+
require 'rchardet/hebrewprober'
|
40
|
+
require 'rchardet/jisfreq'
|
41
|
+
require 'rchardet/jpcntx'
|
42
|
+
require 'rchardet/langbulgarianmodel'
|
43
|
+
require 'rchardet/langcyrillicmodel'
|
44
|
+
require 'rchardet/langgreekmodel'
|
45
|
+
require 'rchardet/langhebrewmodel'
|
46
|
+
require 'rchardet/langhungarianmodel'
|
47
|
+
require 'rchardet/langthaimodel'
|
48
|
+
require 'rchardet/latin1prober'
|
49
|
+
|
50
|
+
require 'rchardet/mbcsgroupprober'
|
51
|
+
require 'rchardet/mbcssm'
|
52
|
+
require 'rchardet/sbcharsetprober'
|
53
|
+
require 'rchardet/sbcsgroupprober'
|
54
|
+
require 'rchardet/sjisprober'
|
55
|
+
require 'rchardet/universaldetector'
|
56
|
+
require 'rchardet/utf8prober'
|
57
|
+
|
58
|
+
module CharDet
|
59
|
+
VERSION = "1.3"
|
60
|
+
def CharDet.detect(aBuf)
|
61
|
+
u = UniversalDetector.new
|
62
|
+
u.reset
|
63
|
+
u.feed(aBuf)
|
64
|
+
u.close
|
65
|
+
u.result
|
66
|
+
end
|
67
|
+
end
|
@@ -0,0 +1,40 @@
|
|
1
|
+
#
|
2
|
+
# version.rb
|
3
|
+
#
|
4
|
+
#--
|
5
|
+
# Copyright (c) 1998-2003 Minero Aoki <aamine@loveruby.net>
|
6
|
+
#
|
7
|
+
# Permission is hereby granted, free of charge, to any person obtaining
|
8
|
+
# a copy of this software and associated documentation files (the
|
9
|
+
# "Software"), to deal in the Software without restriction, including
|
10
|
+
# without limitation the rights to use, copy, modify, merge, publish,
|
11
|
+
# distribute, sublicense, and/or sell copies of the Software, and to
|
12
|
+
# permit persons to whom the Software is furnished to do so, subject to
|
13
|
+
# the following conditions:
|
14
|
+
#
|
15
|
+
# The above copyright notice and this permission notice shall be
|
16
|
+
# included in all copies or substantial portions of the Software.
|
17
|
+
#
|
18
|
+
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
|
19
|
+
# EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
|
20
|
+
# MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
|
21
|
+
# NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
|
22
|
+
# LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
|
23
|
+
# OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
|
24
|
+
# WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
|
25
|
+
#
|
26
|
+
# Note: Originally licensed under LGPL v2+. Using MIT license for Rails
|
27
|
+
# with permission of Minero Aoki.
|
28
|
+
#++
|
29
|
+
|
30
|
+
#:stopdoc:
|
31
|
+
module TMail
|
32
|
+
module VERSION
|
33
|
+
MAJOR = 1
|
34
|
+
MINOR = 2
|
35
|
+
TINY = 7
|
36
|
+
MICRO = 1
|
37
|
+
|
38
|
+
STRING = [MAJOR, MINOR, TINY, MICRO].join('.')
|
39
|
+
end
|
40
|
+
end
|