tc211-termbase 0.1.8 → 0.1.9
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/Gemfile.lock +1 -1
- data/lib/tc211/termbase/term.rb +89 -2
- data/lib/tc211/termbase/version.rb +1 -1
- metadata +1 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 76f28b674292d0635730a89b8b2ce2d474d7c89025461e6617a06bcdb456c5d0
|
|
4
|
+
data.tar.gz: 22b29329630f7c9c33f4ac0488ce449665950b2ed988a628c427265986bdccbe
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 4108f4b967b3d95aebd2c93e20b57a45f9af634b3851a1c052ddb66e3fe34c925db6a257249fae2b18351c5b03c1222ab5ff712ff5404e01c2a5f790de1f1a87
|
|
7
|
+
data.tar.gz: e24326cb32164e923f19db96d4cb4fbb516e02f1509cda8da52e04dd4004385c7327fae9a8a275c926e32ad4a7f8479d351d4a2d1f5c0a0f6e0bbf7249d7c6b1
|
data/Gemfile.lock
CHANGED
data/lib/tc211/termbase/term.rb
CHANGED
|
@@ -39,9 +39,9 @@ class Term
|
|
|
39
39
|
next unless v
|
|
40
40
|
case k
|
|
41
41
|
when /^example/
|
|
42
|
-
|
|
42
|
+
add_example(v)
|
|
43
43
|
when /^note/
|
|
44
|
-
|
|
44
|
+
add_note(v)
|
|
45
45
|
else
|
|
46
46
|
# puts"Key #{k}"
|
|
47
47
|
key = k.gsub("-", "_")
|
|
@@ -51,6 +51,93 @@ class Term
|
|
|
51
51
|
self
|
|
52
52
|
end
|
|
53
53
|
|
|
54
|
+
STRIP_PUNCTUATION = [
|
|
55
|
+
"：",
|
|
56
|
+
":",
|
|
57
|
+
".",
|
|
58
|
+
"–",
|
|
59
|
+
" "
|
|
60
|
+
]
|
|
61
|
+
|
|
62
|
+
# WARNING
|
|
63
|
+
# Always put the longer Regexp match in front!
|
|
64
|
+
EXAMPLE_PREFIXES = {
|
|
65
|
+
# TODO: fix this, we should not have "EXAMPLES"
|
|
66
|
+
eng: ["EXAMPLES", "EXAMPLE"],
|
|
67
|
+
ara: "مثال",
|
|
68
|
+
chi: "示例",
|
|
69
|
+
dan: "EKSEMPEL",
|
|
70
|
+
dut: "VOORBEELD",
|
|
71
|
+
fin: "ESIM",
|
|
72
|
+
fre: "Exemple",
|
|
73
|
+
# ger: "",
|
|
74
|
+
jpn: "例",
|
|
75
|
+
kor: "보기",
|
|
76
|
+
pol: "PRZYKŁAD",
|
|
77
|
+
may: "Contoh",
|
|
78
|
+
rus: "Пример",
|
|
79
|
+
spa: "Ejemplo",
|
|
80
|
+
swe: "Exempel"
|
|
81
|
+
}
|
|
82
|
+
|
|
83
|
+
# WARNING
|
|
84
|
+
# Always put the longer Regexp match in front!
|
|
85
|
+
NOTE_PREFIXES = {
|
|
86
|
+
eng: ["Note \\d to entry", "NOTE"],
|
|
87
|
+
ara: "ملاحظة",
|
|
88
|
+
chi: "注",
|
|
89
|
+
dan: "Note",
|
|
90
|
+
dut: "OPMERKING",
|
|
91
|
+
fin: "HUOM\\.?", # Matches "HUOM", "HUOM.", "HUOM 1." and "HUOM. 1." (numeral added by the method)
|
|
92
|
+
fre: "A noter",
|
|
93
|
+
# ger: "",
|
|
94
|
+
jpn: "備考",
|
|
95
|
+
kor: "비고",
|
|
96
|
+
pol: "UWAGA",
|
|
97
|
+
may: "catatan",
|
|
98
|
+
rus: "нота",
|
|
99
|
+
spa: "Nota",
|
|
100
|
+
swe: ["Anm. \\d till termpost", "Anm. \\d till terpost", "Anm."]
|
|
101
|
+
}
|
|
102
|
+
|
|
103
|
+
# To match Chinese and Japanese numerals
|
|
104
|
+
ALL_FULL_HALF_WIDTH_NUMBERS = "[0-9０-９]"
|
|
105
|
+
|
|
106
|
+
def add_example(example)
|
|
107
|
+
c = clean_prefixed_string(example, EXAMPLE_PREFIXES)
|
|
108
|
+
@examples << c
|
|
109
|
+
end
|
|
110
|
+
|
|
111
|
+
def add_note(note)
|
|
112
|
+
c = clean_prefixed_string(note, NOTE_PREFIXES)
|
|
113
|
+
@notes << c
|
|
114
|
+
end
|
|
115
|
+
|
|
116
|
+
def clean_prefixed_string(string, criterion_map)
|
|
117
|
+
carry = string.strip
|
|
118
|
+
criterion_map.values.flatten.each do |mat|
|
|
119
|
+
# puts "example string: #{carry}, mat: #{mat}"
|
|
120
|
+
|
|
121
|
+
# puts "note string: #{carry}, mat: #{mat}"
|
|
122
|
+
# if @id == 318 and mat == "Nota" and string == "NOTA 1 Una operación tiene un nombre y una lista de parámetros."
|
|
123
|
+
# require "pry"
|
|
124
|
+
# binding.pry
|
|
125
|
+
# end
|
|
126
|
+
|
|
127
|
+
carry = carry.sub(
|
|
128
|
+
Regexp.new(
|
|
129
|
+
"^#{mat}\s*[#{STRIP_PUNCTUATION.join('')}]?" +
|
|
130
|
+
"\s*#{ALL_FULL_HALF_WIDTH_NUMBERS}*\s*"+
|
|
131
|
+
"[#{STRIP_PUNCTUATION.join('')}]?\s*",
|
|
132
|
+
Regexp::IGNORECASE
|
|
133
|
+
),
|
|
134
|
+
'')
|
|
135
|
+
end
|
|
136
|
+
|
|
137
|
+
carry
|
|
138
|
+
end
|
|
139
|
+
|
|
140
|
+
|
|
54
141
|
# The termid should ALWAYS be an integer.
|
|
55
142
|
# https://github.com/riboseinc/tc211-termbase/issues/1
|
|
56
143
|
def id=(newid)
|