text_alignment 0.3.20 → 0.3.21
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
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 65e1d9b45ff59ac0a233b7656d2aca99d7e4e1051b1a03a0c7726521d4f2b280
|
|
4
|
+
data.tar.gz: 710a3b68c5263f26572727e6e9591ebd5fdb095af4633bd5037c61eae0bb5cb6
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 598df22e41bbbe0a84b6e1a6a4e631ab0d8166810afd652086595feecbf0808a886685f42e5466626cbb1d6950dd9f1181be776b9938d6174dc7735c3ace24cd
|
|
7
|
+
data.tar.gz: f7dedfb7e64919129f816fbba24dbd1c2e2a056c242a0865915b8a611f594399b17d051d004a846796bba1c2e89c6fb2f17116cd118ca6217cf1a5dff4f6d4d8
|
|
@@ -3,7 +3,7 @@ require 'string-similarity'
|
|
|
3
3
|
|
|
4
4
|
module TextAlignment; end unless defined? TextAlignment
|
|
5
5
|
|
|
6
|
-
TextAlignment::SIZE_NGRAM =
|
|
6
|
+
TextAlignment::SIZE_NGRAM = 8 unless defined? TextAlignment::SIZE_NGRAM
|
|
7
7
|
TextAlignment::SIZE_WINDOW = 40 unless defined? TextAlignment::SIZE_WINDOW
|
|
8
8
|
TextAlignment::TEXT_SIMILARITY_TRESHOLD = 0.8 unless defined? TextAlignment::TEXT_SIMILARITY_TRESHOLD
|
|
9
9
|
|
|
@@ -36,7 +36,7 @@ class TextAlignment::AnchorFinder
|
|
|
36
36
|
search_position = @end_s2_prev
|
|
37
37
|
while @beg_s2 = @s2.index(anchor, search_position)
|
|
38
38
|
# if both the begining points are sufficiantly close to the end points of the last match
|
|
39
|
-
break if @
|
|
39
|
+
break if @beg_s1 > 0 && @beg_s2 > 0 && (@beg_s1 - @end_s1_prev < 5) && (@beg_s2 >= @end_s2_prev) && (@beg_s2 - @end_s2_prev < 5)
|
|
40
40
|
|
|
41
41
|
left_window_s1, left_window_s2 = get_left_windows
|
|
42
42
|
break if left_window_s1 && (text_similarity(left_window_s1, left_window_s2) > TextAlignment::TEXT_SIMILARITY_TRESHOLD)
|
|
@@ -121,8 +121,8 @@ class TextAlignment::TextAlignment
|
|
|
121
121
|
if mblocks[-1][:source][:end] < str1.length && mblocks[-1][:target][:end] < str2.length
|
|
122
122
|
b1 = mblocks[-1][:source][:end]
|
|
123
123
|
b2 = mblocks[-1][:target][:end]
|
|
124
|
-
_str1 = str1[b1 ...
|
|
125
|
-
_str2 = str2[b2 ...
|
|
124
|
+
_str1 = str1[b1 ... str1.length]
|
|
125
|
+
_str2 = str2[b2 ... str2.length]
|
|
126
126
|
|
|
127
127
|
unless _str1.strip.empty?
|
|
128
128
|
if _str2.strip.empty?
|