entityextractor 0.0.1 → 0.0.2
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.
- data/lib/entityextractor.rb +7 -5
- metadata +2 -2
data/lib/entityextractor.rb
CHANGED
|
@@ -37,16 +37,17 @@ class EntityExtractor
|
|
|
37
37
|
def extractALLCAPS(minchar, ignoreterms)
|
|
38
38
|
@input.each do |i|
|
|
39
39
|
addlist = Array.new
|
|
40
|
-
|
|
40
|
+
savefield = i[@extractfield].to_s + " "
|
|
41
|
+
parseALLCAPS(i[@extractfield].to_s, i, minchar, addlist, ignoreterms, savefield)
|
|
41
42
|
end
|
|
42
43
|
end
|
|
43
44
|
|
|
44
45
|
# Parses terms in all caps
|
|
45
|
-
def parseALLCAPS(toParse, i, minchar, addlist, ignoreterms)
|
|
46
|
+
def parseALLCAPS(toParse, i, minchar, addlist, ignoreterms, savefield)
|
|
46
47
|
if toParse =~ (/[A-Z]{#{minchar}}/)
|
|
47
48
|
index = toParse =~ (/[A-Z]{#{minchar}}/)
|
|
48
49
|
charnum = 0
|
|
49
|
-
|
|
50
|
+
|
|
50
51
|
# Find word in all caps
|
|
51
52
|
toParse.each_char do |c|
|
|
52
53
|
if charnum >= index
|
|
@@ -71,14 +72,15 @@ class EntityExtractor
|
|
|
71
72
|
if !(ignoreterms.include? toParse[index..charnum])
|
|
72
73
|
addlist.push(toParse[index..charnum])
|
|
73
74
|
end
|
|
74
|
-
|
|
75
|
+
|
|
75
76
|
parsedstring = toParse[0..charnum]
|
|
76
77
|
toParse.slice! parsedstring
|
|
77
|
-
parseALLCAPS(toParse, i, minchar, addlist, ignoreterms)
|
|
78
|
+
parseALLCAPS(toParse, i, minchar, addlist, ignoreterms, savefield)
|
|
78
79
|
|
|
79
80
|
# If there are no (more) results, append addlist to JSON
|
|
80
81
|
else
|
|
81
82
|
i["extract"] = addlist
|
|
83
|
+
i[@extractfield] = savefield
|
|
82
84
|
@output.push(i)
|
|
83
85
|
end
|
|
84
86
|
end
|
metadata
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: entityextractor
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.0.
|
|
4
|
+
version: 0.0.2
|
|
5
5
|
prerelease:
|
|
6
6
|
platform: ruby
|
|
7
7
|
authors:
|
|
@@ -9,7 +9,7 @@ authors:
|
|
|
9
9
|
autorequire:
|
|
10
10
|
bindir: bin
|
|
11
11
|
cert_chain: []
|
|
12
|
-
date: 2014-02-
|
|
12
|
+
date: 2014-02-26 00:00:00.000000000 Z
|
|
13
13
|
dependencies: []
|
|
14
14
|
description: Extracts entities and terms from any JSON.
|
|
15
15
|
email: shidash@shidash.com
|