text_parser 0.1.6 → 0.1.7
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/text_parser.rb +2 -1
- data/test/text_parser_test.rb +12 -0
- data/text_parser.gemspec +1 -1
- metadata +2 -2
data/lib/text_parser.rb
CHANGED
|
@@ -16,7 +16,8 @@ module TextParser
|
|
|
16
16
|
options[:dictionary] = text.split(" ") unless options[:dictionary]
|
|
17
17
|
return [] if options[:dictionary].count < 1
|
|
18
18
|
regex = Regexp.new(options[:dictionary].join('\\b|\\b'), Regexp::IGNORECASE)
|
|
19
|
-
match_result = text.scan(regex).map{|i| i.downcase}
|
|
19
|
+
match_result = text.scan(regex).map{|i| i.downcase}
|
|
20
|
+
match_result.select!{|i| i.size >= options[:minimum_length]} if options[:minimum_length]
|
|
20
21
|
match_result.each do |w|
|
|
21
22
|
result << {:hits => match_result.count(w), :word => w} unless result.select{|r| r[:word] == w}.shift || options[:negative_dictionary].map{|i| i.downcase}.include?(w)
|
|
22
23
|
end
|
data/test/text_parser_test.rb
CHANGED
|
@@ -100,6 +100,18 @@ class TextParserTest < Test::Unit::TestCase
|
|
|
100
100
|
{:word => "pão", :hits => 1}], "Pão açúcar".parse
|
|
101
101
|
assert_equal [{:word => "ãéç", :hits => 1}], "ãéç".parse
|
|
102
102
|
end
|
|
103
|
+
|
|
104
|
+
def test_minimum_length
|
|
105
|
+
text = "a ab abc "
|
|
106
|
+
assert_equal [{:word => "a", :hits => 1},
|
|
107
|
+
{:word => "ab", :hits => 1},
|
|
108
|
+
{:word => "abc", :hits => 1}], text.parse(:minimum_length => 1)
|
|
109
|
+
assert_equal [{:word => "ab", :hits => 1},
|
|
110
|
+
{:word => "abc", :hits => 1}], text.parse(:minimum_length => 2)
|
|
111
|
+
assert_equal [{:word => "abc", :hits => 1}], text.parse(:minimum_length => 3)
|
|
112
|
+
assert_equal [{:word => "abc", :hits => 1}], text.parse(:minimum_length => 2, :negative_dictionary => ["ab"])
|
|
113
|
+
assert_equal [], text.parse(:minimum_length => 3, :dictionary => ["a"])
|
|
114
|
+
end
|
|
103
115
|
end
|
|
104
116
|
|
|
105
117
|
|
data/text_parser.gemspec
CHANGED
metadata
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: text_parser
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.1.
|
|
4
|
+
version: 0.1.7
|
|
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: 2011-12-
|
|
12
|
+
date: 2011-12-19 00:00:00.000000000Z
|
|
13
13
|
dependencies: []
|
|
14
14
|
description: Using method parse in the String object you can parse any text.
|
|
15
15
|
email: fpaula@gmail.com
|