pragmatic_tokenizer 1.1.0 → 1.1.1
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
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA1:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 9defa80663e9e5955967ba95d1bd346b5e06cca3
|
|
4
|
+
data.tar.gz: 5f3ec1a9392c2664b3ce8a26a7de9ba03e3a22e1
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: ddac620c5e335ee8ec3daa1c78ed96d9637e37d57dbb34886a153d7f292bad173ca823c291c44fc1a6f047276049d2e96562ab0a477c5cc58d6f0385636b3ba3
|
|
7
|
+
data.tar.gz: 9361da2cada5ab2e301601f4ddd50bde3fe7935f1b88447a3c5356b72063357a319471b4f1405f2c7aec3fa07f8c794f7a3aadc24a0e82919d2b829411725505
|
|
@@ -23,6 +23,7 @@ module PragmaticTokenizer
|
|
|
23
23
|
shift_vertical_bar(text)
|
|
24
24
|
convert_dbl_quotes(text)
|
|
25
25
|
convert_sgl_quotes(text)
|
|
26
|
+
convert_apostrophe_s(text)
|
|
26
27
|
shift_beginning_hyphen(text)
|
|
27
28
|
shift_ending_hyphen(text)
|
|
28
29
|
text.squeeze(' ')
|
|
@@ -120,6 +121,11 @@ module PragmaticTokenizer
|
|
|
120
121
|
end
|
|
121
122
|
end
|
|
122
123
|
|
|
124
|
+
def convert_apostrophe_s(text)
|
|
125
|
+
puts "Text: #{text.include?("\u{0301}")}"
|
|
126
|
+
text.gsub!(/\s\u{0301}(?=s(\s|\z))/, PragmaticTokenizer::Languages::Common::PUNCTUATION_MAP['`']) || text
|
|
127
|
+
end
|
|
128
|
+
|
|
123
129
|
def shift_beginning_hyphen(text)
|
|
124
130
|
text.gsub!(/\s+-/, ' - ') || text
|
|
125
131
|
end
|
|
@@ -122,7 +122,9 @@ module PragmaticTokenizer
|
|
|
122
122
|
private
|
|
123
123
|
|
|
124
124
|
def post_process(text)
|
|
125
|
+
puts "Text: #{text}"
|
|
125
126
|
@tokens = PostProcessor.new(text: text, abbreviations: abbreviations).post_process
|
|
127
|
+
puts "Tokens: #{@tokens}"
|
|
126
128
|
downcase! if downcase
|
|
127
129
|
expand_contractions!(contractions) if expand_contractions
|
|
128
130
|
clean! if clean
|