konjak 0.0.21 → 0.0.22
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 +4 -4
- data/lib/konjak/tmx_segmentor/strategy.rb +32 -14
- data/lib/konjak/version.rb +1 -1
- metadata +2 -2
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: c6d01369d26197b67672a4260930011ef66880bd
|
4
|
+
data.tar.gz: 9e2596af227e206039d32f3107fb3ee318adbb25
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 5248cd97806b0f7ad9b487838dc03ae65d631ec85d69dac8b8c1d5f84cc2265e8b9303ce9be543537939349a94dd13f465fe953cc6fc2ff2de640f26320c7ec6
|
7
|
+
data.tar.gz: 0e37939c78bfd23a9be0f86f89c06a0edfbf83140a0dd0d5d1c9debd4ef0beeb7f59c610c3b278115249bc852492122ea1887cf8ea91e3b5da00fad4300cab43
|
@@ -8,7 +8,15 @@ module Konjak
|
|
8
8
|
class TmxSegmentor < Segmentor
|
9
9
|
class Strategy
|
10
10
|
Edge = Struct.new(:prev, :current)
|
11
|
-
Node = Struct.new(:range, :
|
11
|
+
Node = Struct.new(:range, :segments) do
|
12
|
+
def <=>(other)
|
13
|
+
[range.begin, -max_segment_size] <=> [other.range.begin, -other.max_segment_size]
|
14
|
+
end
|
15
|
+
|
16
|
+
def max_segment_size
|
17
|
+
segments.max_by {|s| s.text.size }.text.size
|
18
|
+
end
|
19
|
+
end
|
12
20
|
Node::None = -1
|
13
21
|
|
14
22
|
include Mem
|
@@ -25,19 +33,21 @@ module Konjak
|
|
25
33
|
return [@text] if nodes.empty?
|
26
34
|
|
27
35
|
segments = []
|
28
|
-
|
36
|
+
prev_text_end = 0
|
37
|
+
prev_segment = nil
|
29
38
|
max_cost_path.each do |node|
|
30
39
|
range = node.range
|
31
|
-
segment = node
|
32
|
-
prev_text = @text[
|
40
|
+
segment = select_next_segment(prev_segment, node)
|
41
|
+
prev_text = @text[prev_text_end...range.begin]
|
33
42
|
|
34
43
|
segments << prev_text unless prev_text.empty?
|
35
44
|
|
36
45
|
segments << SegmentString.new(@text[range.begin, range.size], segment)
|
37
46
|
|
38
|
-
|
47
|
+
prev_segment = segment
|
48
|
+
prev_text_end = range.end
|
39
49
|
end
|
40
|
-
after_text = @text[
|
50
|
+
after_text = @text[prev_text_end..-1]
|
41
51
|
segments << after_text unless after_text.empty?
|
42
52
|
segments
|
43
53
|
end
|
@@ -48,10 +58,11 @@ module Konjak
|
|
48
58
|
costs = nodes.map {|node| [node, node.range.size] }.to_h
|
49
59
|
|
50
60
|
edges.each do |edge|
|
51
|
-
node, node2
|
61
|
+
node, node2 = edge.prev, edge.current
|
62
|
+
node2_cost = costs[node2]
|
52
63
|
new_node2_cost = costs[node] + calc_edge_cost(edge)
|
53
64
|
|
54
|
-
if
|
65
|
+
if node2_cost < new_node2_cost
|
55
66
|
costs[node2] = new_node2_cost
|
56
67
|
prev_nodes[node2] = node
|
57
68
|
end
|
@@ -59,7 +70,7 @@ module Konjak
|
|
59
70
|
|
60
71
|
node, _ = costs.max_by {|_, cost| cost }
|
61
72
|
|
62
|
-
|
73
|
+
Enumerator.new {|y|
|
63
74
|
loop do
|
64
75
|
break if node == Node::None
|
65
76
|
y << node
|
@@ -90,16 +101,18 @@ module Konjak
|
|
90
101
|
def nodes
|
91
102
|
return @nodes if @nodes
|
92
103
|
|
93
|
-
@nodes =
|
104
|
+
@nodes = {}
|
94
105
|
|
95
106
|
translation_units.each {|tu|
|
96
107
|
segment = tu.variant(@lang).segment
|
97
108
|
@text.scan(compile_pattern(segment)) {
|
98
|
-
|
109
|
+
r = $~.begin(0)...$~.end(0)
|
110
|
+
@nodes[r] ||= []
|
111
|
+
@nodes[r] << segment
|
99
112
|
}
|
100
113
|
}
|
101
|
-
|
102
|
-
@nodes.
|
114
|
+
@nodes = @nodes.map {|(r, segments)| Node.new(r, segments) }
|
115
|
+
@nodes.sort!
|
103
116
|
|
104
117
|
@nodes
|
105
118
|
end
|
@@ -107,7 +120,8 @@ module Konjak
|
|
107
120
|
def default_options
|
108
121
|
{
|
109
122
|
translation_unit_filter: -> (tu) { true },
|
110
|
-
calc_edge_cost: -> (edge) { edge.current.
|
123
|
+
calc_edge_cost: -> (edge) { edge.current.max_segment_size },
|
124
|
+
select_next_segment: -> (_, node) { node.segments.first }
|
111
125
|
}
|
112
126
|
end
|
113
127
|
|
@@ -119,6 +133,10 @@ module Konjak
|
|
119
133
|
@options[:translation_unit_filter]
|
120
134
|
end
|
121
135
|
|
136
|
+
def select_next_segment(prev_segment, node)
|
137
|
+
@options[:select_next_segment].call(prev_segment, node)
|
138
|
+
end
|
139
|
+
|
122
140
|
def translation_units
|
123
141
|
@translation_units ||= @tmx.body.translation_units.select(&translation_unit_filter)
|
124
142
|
end
|
data/lib/konjak/version.rb
CHANGED
metadata
CHANGED
@@ -1,14 +1,14 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: konjak
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.0.
|
4
|
+
version: 0.0.22
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Seiei Higa
|
8
8
|
autorequire:
|
9
9
|
bindir: bin
|
10
10
|
cert_chain: []
|
11
|
-
date: 2015-10-
|
11
|
+
date: 2015-10-19 00:00:00.000000000 Z
|
12
12
|
dependencies:
|
13
13
|
- !ruby/object:Gem::Dependency
|
14
14
|
name: activesupport
|