dwc_agent 1.4.4 → 1.4.5
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +4 -4
- data/lib/dwc_agent/constants.rb +7 -6
- data/lib/dwc_agent/parser.rb +4 -4
- data/lib/dwc_agent/version.rb +1 -1
- metadata +2 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA256:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: 7b8890828a04d30f7989f3545df032be216d1bab086f1ec6576cbe93569231bd
|
4
|
+
data.tar.gz: d94e120615540e65f29a13132d46b34639e3380e2265439fd05cf5febc580ff6
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: f2e30d42a8c2d743791c2322a51d3fe022e5ef458f68ebd30763cfc74e5ec9a1914faca1ff90cfb3364bead66cde22bccbc0ab47aef6a7d17a78c6954c8cbeff
|
7
|
+
data.tar.gz: d2de0c26c64f77fd273db67465d8772aef5480eb0c4dd1b48446dbb6a6495164bc9e2fbd133280c18ef7e975f82bf9f46dcd65e87727531fa7c514ebc09e0be7
|
data/lib/dwc_agent/constants.rb
CHANGED
@@ -15,7 +15,6 @@ module DwcAgent
|
|
15
15
|
\b[,;]?\s*(?i:ann?onymous)\b|
|
16
16
|
\b[,;]?\s*\(?(?i:undetermined|indeterminable|dummy|interim|accession|ill(eg|is)ible|scripsit)\)?\b|
|
17
17
|
\b[,;]?\s*(?i:importer|gift)\:?\b|
|
18
|
-
\b[,;]?\s*(?i:frère|frere|père|pere|soeur|sister|bro)\.?(\b|\z)|
|
19
18
|
\b[,;]?\s*(?i:string)\b|
|
20
19
|
\b[,;]?\s*(?i:person\s*string)\b|
|
21
20
|
\b[,;]?\s*(?i:colls)\.(\b|\z)|
|
@@ -154,10 +153,12 @@ module DwcAgent
|
|
154
153
|
}
|
155
154
|
|
156
155
|
PHRASE_SUBS = {
|
157
|
-
'
|
158
|
-
'
|
159
|
-
'
|
160
|
-
'
|
156
|
+
'dr\.' => 'Dr. ',
|
157
|
+
'mr\.' => 'Mr. ',
|
158
|
+
'mrs\.' => 'Mrs. ',
|
159
|
+
'prof\.' => 'Prof. ',
|
160
|
+
'\, ph\.d\.' => ' Ph.D.',
|
161
|
+
'\, bro\.' => ' Bro.'
|
161
162
|
}
|
162
163
|
|
163
164
|
COMPLEX_SEPARATORS = %r{
|
@@ -258,6 +259,6 @@ module DwcAgent
|
|
258
259
|
"has not"
|
259
260
|
]
|
260
261
|
|
261
|
-
TITLE = /\s*\b(sir|count(ess)?|(gen|adm|col|maj|capt|cmdr|lt|sgt|cpl|pvt|prof|dr|md|ph\.?d|rev|
|
262
|
+
TITLE = /\s*\b(sir|count(ess)?|(gen|adm|col|maj|capt|cmdr|lt|sgt|cpl|pvt|prof|dr|md|ph\.?d|rev|mme|abbé|ptre|bro|esq)\.?|docteur|father|cantor|vicar|père|pastor|rabbi|reverend|pere|soeur|sister|professor)(\s+|$)/i
|
262
263
|
|
263
264
|
end
|
data/lib/dwc_agent/parser.rb
CHANGED
@@ -8,7 +8,7 @@ module DwcAgent
|
|
8
8
|
end
|
9
9
|
|
10
10
|
def initialize
|
11
|
-
options = {
|
11
|
+
options = {
|
12
12
|
prefer_comma_as_separator: true,
|
13
13
|
separator: SPLIT_BY,
|
14
14
|
title: TITLE
|
@@ -18,11 +18,11 @@ module DwcAgent
|
|
18
18
|
@strip_out_regex = Regexp.new STRIP_OUT.to_s
|
19
19
|
@residual_terminators_regex = Regexp.new SPLIT_BY.to_s + %r{\s*\z}.to_s
|
20
20
|
@char_subs_regex = Regexp.new [CHAR_SUBS.keys.join].to_s
|
21
|
-
@phrase_subs_regex = Regexp.new
|
21
|
+
@phrase_subs_regex = Regexp.new((PHRASE_SUBS.keys.join('|')).to_s, Regexp::IGNORECASE)
|
22
22
|
@complex_separators_regex = Regexp.new COMPLEX_SEPARATORS.to_s
|
23
23
|
@add_separators_regex = Regexp.new %r{(\S{1}\.)([[:alpha:]]{2,})}.to_s
|
24
24
|
end
|
25
|
-
|
25
|
+
|
26
26
|
# Parses the passed-in string and returns a list of names.
|
27
27
|
#
|
28
28
|
# @param names [String] the name or names to be parsed
|
@@ -42,4 +42,4 @@ module DwcAgent
|
|
42
42
|
end
|
43
43
|
|
44
44
|
end
|
45
|
-
end
|
45
|
+
end
|
data/lib/dwc_agent/version.rb
CHANGED
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: dwc_agent
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 1.4.
|
4
|
+
version: 1.4.5
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- David P. Shorthouse
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2020-01-
|
11
|
+
date: 2020-01-22 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: namae
|