pragmatic_segmenter 0.3.2 → 0.3.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 7e52d1869830dfba91d5e5a00f15d3529081691e
-  data.tar.gz: 5754723c0ba657a31e3471d785b034e3c1814e33
+  metadata.gz: 9e26400933e02aca93c63db3221fb897f49cdf78
+  data.tar.gz: a0199ca43d0f8558ba3fe867c6e32c69787210c9
 SHA512:
-  metadata.gz: d3a4700ba0369b60d36f633c4e042a5113197956ad02dff64243c70445bc445435af17689f456327753cead2d9d1cee00db548e2e0dad337341f110194522d2c
-  data.tar.gz: f8a28d6582e846ad2ffff21b87f4888b922cc4224faaf85e2782ca92671506dbf55fab86c0ddf06ae6363a3851f8ec9dd44f7bc9956a86f2ed2aea4bc6d69fb5
+  metadata.gz: 1dca5e20b2b062070b0cd319da6549594add065b670506b7e7dd9cbfe9eb8c83985ef197bf6427ee496d700edab5a526c8165fc1a27567e553f7a5e625edb6fe
+  data.tar.gz: a53a4059b4ba41e7d40bf2a21ca1981ae5ed134a8017771f8ecd337bdc96762e2e78ec7c53c00c3ff2e3ac6179cf1e29f24fee703bc6b19fbbf31bc7ff572894

data/README.md

@@ -816,6 +816,9 @@ To test the relative performance of different segmentation tools and libraries I
 **Version 0.3.2**  
 * Add English abbreviations  
 
+**Version 0.3.3**  
+* Fix cleaner bug    
+
 ## Contributing
 
 If you find a text that is incorrectly segmented using this gem, please submit an issue.

data/lib/pragmatic_segmenter/cleaner.rb

@@ -44,6 +44,10 @@ module PragmaticSegmenter
 
     private
 
+    def abbreviations
+      @language::Abbreviation::ABBREVIATIONS
+    end
+
     def check_for_no_space_in_between_sentences(txt)
       words = txt.split(' ')
       words.each do |word|

data/lib/pragmatic_segmenter/version.rb

@@ -1,3 +1,3 @@
 module PragmaticSegmenter
-  VERSION = "0.3.2"
+  VERSION = "0.3.3"
 end

data/spec/pragmatic_segmenter_spec.rb

@@ -2600,11 +2600,17 @@ RSpec.describe PragmaticSegmenter::Segmenter do
     end
 
     describe '#clean' do
-      it 'cleans the text' do
+      it 'cleans the text #001' do
         ps = PragmaticSegmenter::Cleaner.new(text: "It was a cold \nnight in the city.", language: "en")
         expect(ps.clean).to eq("It was a cold night in the city.")
       end
 
+      it 'cleans the text #002' do
+        text = 'injections made by the Shareholder through the years. 7 (max.) 3. Specifications/4.Design and function The operating instructions are part of the product and must be kept in the immediate vicinity of the instrument and readily accessible to skilled "'
+        ps = PragmaticSegmenter::Cleaner.new(text: text)
+        expect(ps.clean).to eq("injections made by the Shareholder through the years. 7 (max.) 3. Specifications/4.Design and function The operating instructions are part of the product and must be kept in the immediate vicinity of the instrument and readily accessible to skilled \"")
+      end
+
       it 'does not mutate the input string (cleaner)' do
         text = "It was a cold \nnight in the city."
         PragmaticSegmenter::Cleaner.new(text: text, language: "en").clean

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: pragmatic_segmenter
 version: !ruby/object:Gem::Version
-  version: 0.3.2
+  version: 0.3.3
 platform: ruby
 authors:
 - Kevin S. Dias