konjak 0.0.6 → 0.0.7
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.
- checksums.yaml +4 -4
- data/lib/konjak/tmx_segmentor/gtt_html_strategy.rb +1 -1
- data/lib/konjak/tmx_segmentor/strategy.rb +12 -1
- data/lib/konjak/tmx_segmentor/text_strategy.rb +1 -1
- data/lib/konjak/translator.rb +14 -2
- data/lib/konjak/translator/translated_string.rb +6 -0
- data/lib/konjak/version.rb +1 -1
- metadata +2 -1
checksums.yaml
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
---
|
2
2
|
SHA1:
|
3
|
-
metadata.gz:
|
4
|
-
data.tar.gz:
|
3
|
+
metadata.gz: c2fc4ee1a4ae07170e1d4d0af6fbed84b6ab60e5
|
4
|
+
data.tar.gz: 93ca5aa335e0b012a9f9f727926ed4918099fa5e
|
5
5
|
SHA512:
|
6
|
-
metadata.gz:
|
7
|
-
data.tar.gz:
|
6
|
+
metadata.gz: 9ebaa19e5a5598a4053de05f718bf643776c2611bf2d4006ddae8195806e593f8f35b43471f943e1d46ee8f4dea4f617ccc607f621f4ec5351c30f4110725fe2
|
7
|
+
data.tar.gz: 30be6bc899940236afe204122fcc916e4441eb2125db0e31a2d0698ec70f2b1fe5aad70a8fa226caa348faf410de2bcef5ea6a26e37f8232233a978c6cc4a0c2
|
@@ -7,9 +7,10 @@ module Konjak
|
|
7
7
|
class Strategy
|
8
8
|
include Mem
|
9
9
|
|
10
|
-
def initialize(tmx, lang)
|
10
|
+
def initialize(tmx, lang, options = {})
|
11
11
|
@tmx = tmx
|
12
12
|
@lang = lang
|
13
|
+
@options = default_options.merge(options)
|
13
14
|
end
|
14
15
|
|
15
16
|
def segmentize(text)
|
@@ -26,9 +27,19 @@ module Konjak
|
|
26
27
|
|
27
28
|
private
|
28
29
|
|
30
|
+
def default_options
|
31
|
+
{min_segment_length: 10}
|
32
|
+
end
|
33
|
+
|
34
|
+
def min_segment_length
|
35
|
+
@options[:min_segment_length]
|
36
|
+
end
|
37
|
+
|
29
38
|
def translation_units
|
30
39
|
@tmx.body.translation_units.sort_by {|tu|
|
31
40
|
-tu.variant(@lang).segment.text.length
|
41
|
+
}.reject {|tu|
|
42
|
+
tu.variant(@lang).segment.text.length < min_segment_length
|
32
43
|
}
|
33
44
|
end
|
34
45
|
memoize :translation_units
|
@@ -14,7 +14,7 @@ module Konjak
|
|
14
14
|
texts = []
|
15
15
|
while true
|
16
16
|
head, match, tail = text.partition(segment_text)
|
17
|
-
break if match.empty?
|
17
|
+
break if match.empty? || text.length < min_segment_length
|
18
18
|
texts << head unless head.empty?
|
19
19
|
|
20
20
|
texts << SegmentString.new(match, segment)
|
data/lib/konjak/translator.rb
CHANGED
@@ -1,5 +1,6 @@
|
|
1
1
|
require 'mem'
|
2
2
|
require 'konjak/tmx_segmentor'
|
3
|
+
require 'konjak/translator/translated_string'
|
3
4
|
|
4
5
|
module Konjak
|
5
6
|
class Translator
|
@@ -20,7 +21,14 @@ module Konjak
|
|
20
21
|
next text unless text.is_a?(TmxSegmentor::SegmentString)
|
21
22
|
source_segment = text.segment
|
22
23
|
target_segment = source_segment.translation_unit.variant(target_lang).segment
|
23
|
-
|
24
|
+
|
25
|
+
if format == :gtt_html
|
26
|
+
gtt_tags = source_segment.extract_gtt_tags_from(text)
|
27
|
+
translated_string = target_segment.interpolate_gtt_tags(gtt_tags)
|
28
|
+
TranslatedString.new(translated_string)
|
29
|
+
else
|
30
|
+
TranslatedString.new(target_segment.text)
|
31
|
+
end
|
24
32
|
end
|
25
33
|
end
|
26
34
|
|
@@ -31,8 +39,12 @@ module Konjak
|
|
31
39
|
content,
|
32
40
|
tmx: tmx,
|
33
41
|
lang: src_lang,
|
34
|
-
format:
|
42
|
+
format: format
|
35
43
|
)
|
36
44
|
end
|
45
|
+
|
46
|
+
def format
|
47
|
+
options[:format]
|
48
|
+
end
|
37
49
|
end
|
38
50
|
end
|
data/lib/konjak/version.rb
CHANGED
metadata
CHANGED
@@ -1,7 +1,7 @@
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
2
2
|
name: konjak
|
3
3
|
version: !ruby/object:Gem::Version
|
4
|
-
version: 0.0.
|
4
|
+
version: 0.0.7
|
5
5
|
platform: ruby
|
6
6
|
authors:
|
7
7
|
- Seiei Higa
|
@@ -171,6 +171,7 @@ files:
|
|
171
171
|
- lib/konjak/translation_unit.rb
|
172
172
|
- lib/konjak/translation_unit_variant.rb
|
173
173
|
- lib/konjak/translator.rb
|
174
|
+
- lib/konjak/translator/translated_string.rb
|
174
175
|
- lib/konjak/unknown_tag.rb
|
175
176
|
- lib/konjak/user_defined_encoding.rb
|
176
177
|
- lib/konjak/version.rb
|