blob: c99c2355aa7554c1fd43f62164938f500c502607 (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
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
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
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
|
#!/usr/bin/perl -w
$|++;
use strict;
use utf8;
binmode(STDIN,"utf8");
binmode(STDOUT,"utf8");
while(<STDIN>) {
chomp;
$_ = " $_ ";
# Regularlize spaces:
s/\x{a0}/ /g; # non-breaking space
s/\x{2009}/ /g; # thin space
s/\x{2028}/ /g; # "line separator"
s/\x{2029}/ /g; # "paragraph separator"
s/\x{202a}/ /g; # "left-to-right embedding"
s/\x{202b}/ /g; # "right-to-left embedding"
s/\x{202c}/ /g; # "pop directional formatting"
s/\x{202d}/ /g; # "left-to-right override"
s/\x{202e}/ /g; # "right-to-left override"
s/\x{85}/ /g; # "next line"
s/\x{fffd}/ /g; # "replacement character"
s/\x{feff}/ /g; # byte-order mark
s/\x{fdd3}/ /g; # "unicode non-character"
# Regularize named HTML/XML escapes:
s/&\s*lt\s*;/</gi; # HTML opening angle bracket
s/&\s*gt\s*;/>/gi; # HTML closing angle bracket
s/&\s*squot\s*;/'/gi; # HTML single quote
s/&\s*quot\s*;/"/gi; # HTML double quote
s/&\s*nbsp\s*;/ /gi; # HTML non-breaking space
s/'/\'/g; # HTML apostrophe
s/&\s*amp\s*;/&/gi; # HTML ampersand (last)
# Regularize known HTML numeric codes:
s/&\s*#\s*160\s*;/ /gi; # no-break space
s/&\s*#45\s*;\s*&\s*#45\s*;/--/g; # hyphen-minus hyphen-minus
s/&\s*#45\s*;/--/g; # hyphen-minus
# Convert other Windows 1252 characters to UTF-8
s/\x{80}/\x{20ac}/g; # euro sign
s/\x{95}/\x{2022}/g; # bullet
s/\x{99}/\x{2122}/g; # trademark sign
# Currency and measure conversions:
s/ (\d\d): (\d\d)/ $1:$2/g;
s/[\x{20a0}]\x{20ac}]/ EUR /g;
s/[\x{00A3}]/ GBP /g;
s/(\W)([A-Z]+\$?)(\d*\.\d+|\d+)/$1$2 $3/g;
s/(\W)(euro?)(\d*\.\d+|\d+)/$1EUR $3/gi;
# Ridiculous double conversions(?) (news commentary and Giga-FrEn):
s/�c/--/g; # long dash
s/\x{e2}\x{20ac}oe/\"/g; # opening double quote
s/\x{e2}\x{20ac}\x{9c}/\"/g; # opening double quote
s/\x{e2}\x{20ac}\x{9d}/\"/g; # closing double quote
s/\x{e2}\x{20ac}\x{2122}/\'/g; # apostrophe
s/\x{e2}\x{20ac}\x{201c}/ -- /g; # en dash?
s/\x{e2}\x{20ac}\x{201d}/ -- /g; # em dash?
s/â(\x{80}\x{99}|\x{80}\x{98})/'/g; # single quote?
s/â(\x{80}\x{9c}|\x{80}\x{9d})/"/g; # double quote?
# Regularize quotes:
s/ˇ/'/g; # caron
s/´/'/g; # acute accent
s/`/'/g; # grave accent
s/ˉ/'/g; # modified letter macron
s/ ,,/ "/g; # ghetto low-99 quote
s/``/"/g; # latex-style left quote
s/''/"/g; # latex-style right quote
s/\x{300c}/"/g; # left corner bracket
s/\x{300d}/"/g; # right corner bracket
s/\x{3003}/"/g; # ditto mark
s/\x{00a8}/"/g; # diaeresis
s/\x{92}/\'/g; # curly apostrophe
s/\x{2019}/\'/g; # curly apostrophe
s/\x{f03d}/\'/g; # curly apostrophe
s/\x{b4}/\'/g; # curly apostrophe
s/\x{2018}/\'/g; # curly single open quote
s/\x{201a}/\'/g; # low-9 quote
s/\x{93}/\"/g; # curly left quote
s/\x{201c}/\"/g; # curly left quote
s/\x{94}/\"/g; # curly right quote
s/\x{201d}/\"/g; # curly right quote
s/\x{2033}/\"/g; # curly right quote
s/\x{201e}/\"/g; # low-99 quote
s/\x{84}/\"/g; # low-99 quote (bad enc)
s/\x{201f}/\"/g; # high-rev-99 quote
s/\x{ab}/\"/g; # opening guillemet
s/\x{bb}/\"/g; # closing guillemet
s/\x{0301}/'/g; # combining acute accent
# Space inverted punctuation:
s/¡/ ¡ /g;
s/¿/ ¿ /g;
# Russian abbreviations:
s/ п. п. / п.п. /g;
s/ ст. л. / ст.л. /g;
s/ т. е. / т.е. /g;
s/ т. к. / т.к. /g;
s/ т. ч. / т.ч. /g;
s/ т. д. / т.д. /g;
s/ т. п. / т.п. /g;
s/ и. о. / и.о. /g;
s/ с. г. / с.г. /g;
s/ г. р. / г.р. /g;
s/ т. н. / т.н. /g;
s/ т. ч. / т.ч. /g;
s/ н. э. / н.э. /g;
# Convert foreign numerals into Arabic numerals
tr/०-९/0-9/; # devangari
tr/౦-౯/0-9/; # telugu
tr/೦-೯/0-9/; # kannada
tr/೦-௯/0-9/; # tamil
tr/൦-൯/0-9/; # malayalam
# Random punctuation:
tr/！-～/!-~/;
s/、/,/g;
# s/。/./g;
s/\x{85}/.../g;
s/…/.../g;
s/―/--/g;
s/–/--/g;
s/─/--/g;
s/—/--/g;
s/•/ * /g;
s/\*/ * /g;
s/،/,/g;
s/؟/?/g;
s/ـ/ /g;
s/Ã ̄/i/g;
s/’/'/g;
s/â€"/"/g;
s/؛/;/g;
# Regularize ligatures:
s/\x{9c}/oe/g; # "oe" ligature
s/\x{0153}/oe/g; # "oe" ligature
s/\x{8c}/Oe/g; # "OE" ligature
s/\x{0152}/Oe/g; # "OE" ligature
s/\x{fb00}/ff/g; # "ff" ligature
s/\x{fb01}/fi/g; # "fi" ligature
s/\x{fb02}/fl/g; # "fl" ligature
s/\x{fb03}/ffi/g; # "ffi" ligature
s/\x{fb04}/ffi/g; # "ffl" ligature
# Convert arbitrary hex or decimal HTML entities to actual characters:
s/&\#x([0-9A-Fa-f]+);/pack("U", hex($1))/ge;
s/&\#([0-9]+);/pack("U", $1)/ge;
s/β/ß/g; # WMT 2010 error
# Strip extra spaces:
s/\s+/ /g;
s/^\s+//;
s/\s+$//;
# Delete control characters:
s/[\x{00}-\x{1f}]//g;
print "$_\n";
}
|
