deepl_diff 1.0.2
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/.gitignore +13 -0
- data/.rspec +2 -0
- data/.rubocop.yml +15 -0
- data/.travis.yml +12 -0
- data/Gemfile +5 -0
- data/README.md +117 -0
- data/Rakefile +8 -0
- data/deepl_diff.gemspec +55 -0
- data/lib/deepl_diff/cache.rb +38 -0
- data/lib/deepl_diff/chunker.rb +56 -0
- data/lib/deepl_diff/linearizer.rb +29 -0
- data/lib/deepl_diff/redis_cache_store.rb +26 -0
- data/lib/deepl_diff/redis_rate_limiter.rb +22 -0
- data/lib/deepl_diff/request.rb +157 -0
- data/lib/deepl_diff/spacing.rb +31 -0
- data/lib/deepl_diff/tokenizer.rb +159 -0
- data/lib/deepl_diff/version.rb +5 -0
- data/lib/deepl_diff.rb +28 -0
- metadata +259 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA256:
|
3
|
+
metadata.gz: 601d9219d37903d18d03a30e85fee9ce783522fbfb75d616f2873ef4ca6da950
|
4
|
+
data.tar.gz: a0b23c609b7f6f48b74a83a818d96d3bb13665568f3563a6e67fe2c16fb42f44
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: a739f33576f117fde32e6c6fd9527835a2f6aeb3fbdd4b3265aa4e0c913e519bc0f8bd536ed2abdcee50721823e75286aed909cbfc59fc889cea40538c83847c
|
7
|
+
data.tar.gz: 87beae80de1cf85db6308cc613e71b428b0008a4b797ef276ef4d16abd65f162aea2cfd09b454ddd6f2fb438ae3939af7c48c42e850f3f4772c2184be1c35839
|
data/.gitignore
ADDED
data/.rspec
ADDED
data/.rubocop.yml
ADDED
data/.travis.yml
ADDED
data/Gemfile
ADDED
data/README.md
ADDED
@@ -0,0 +1,117 @@
|
|
1
|
+
# DeepLDiff
|
2
|
+
|
3
|
+
DeepL API wrapper helps to translate only changes between revisions of long texts.
|
4
|
+
|
5
|
+
|
6
|
+
|
7
|
+
|
8
|
+
**DeepLDiff** based on [GoogleTranslateDiff](https://github.com/gzigzigzeo/google_translate_diff)
|
9
|
+
## Use case
|
10
|
+
|
11
|
+
Assume your project contains a significant amount of products descriptions which:
|
12
|
+
- Require retranslation each time user edits them.
|
13
|
+
- Have a lot of equal parts (like return policy).
|
14
|
+
- Change frequently.
|
15
|
+
|
16
|
+
If your user changes a single word within the long description, you will be charged for the retranslation of the whole text.
|
17
|
+
|
18
|
+
Much better approach is to try to translate every repeated structural element (sentence) in your texts array just once to save money. This gem helps to make it done.
|
19
|
+
|
20
|
+
## Installation
|
21
|
+
|
22
|
+
Add this line to your application's Gemfile:
|
23
|
+
|
24
|
+
```ruby
|
25
|
+
gem 'deepl_diff'
|
26
|
+
```
|
27
|
+
|
28
|
+
And then execute:
|
29
|
+
|
30
|
+
$ bundle
|
31
|
+
|
32
|
+
Or install it yourself as:
|
33
|
+
|
34
|
+
$ gem install deepl_diff
|
35
|
+
|
36
|
+
## Usage
|
37
|
+
|
38
|
+
```ruby
|
39
|
+
require "deepl_diff"
|
40
|
+
|
41
|
+
# This dependencies are not included, as you might need to roll your own cache based on different store
|
42
|
+
require "redis"
|
43
|
+
require "connection_pool"
|
44
|
+
require "redis-namespace"
|
45
|
+
require "ratelimit" # Optional, if you will use
|
46
|
+
|
47
|
+
# Setup https://github.com/wikiti/deepl-rb
|
48
|
+
DeepL.configure do |config|
|
49
|
+
config.auth_key = 'your-api-token'
|
50
|
+
config.host = 'https://api-free.deepl.com' # Default value is 'https://api.deepl.com'
|
51
|
+
config.version = 'v1' # Default value is 'v2'
|
52
|
+
end
|
53
|
+
|
54
|
+
# I always use pool for redis
|
55
|
+
pool = ConnectionPool.new(size: 10, timeout: 5) { Redis.new }
|
56
|
+
|
57
|
+
# Pass DeepL for DeepLDiff
|
58
|
+
DeepLDiff.api = DeepL
|
59
|
+
|
60
|
+
DeepLDiff.cache_store =
|
61
|
+
DeepLDiff::RedisCacheStore.new(pool, timeout: 7.days, namespace: "t")
|
62
|
+
|
63
|
+
# Optional
|
64
|
+
DeepLDiff.rate_limiter =
|
65
|
+
DeepLDiff::RedisRateLimiter.new(
|
66
|
+
pool, threshold: 8000, interval: 60, namespace: "t"
|
67
|
+
)
|
68
|
+
|
69
|
+
DeepLDiff.translate("test translations", from: "en", to: "es")
|
70
|
+
```
|
71
|
+
|
72
|
+
## How it works
|
73
|
+
|
74
|
+
- Text nodes are extracted from HTML.
|
75
|
+
- Every text node is split into sentences (using `punkt-segmenter` gem).
|
76
|
+
- Cache is checked for the presence of each sentence (using language couple and a hash of string).
|
77
|
+
- Missing sentences are translated via API and cached.
|
78
|
+
- Original HTML is recombined from translations and cache data.
|
79
|
+
|
80
|
+
*NOTE:* if `:from` is not specified or equal to nil, then the DeepL API will be called twice, the first time a sample of text up to 100 characters long will be transferred to determine the language, and the second time the entire text will be transferred.
|
81
|
+
Try to specify `:from` explicitly
|
82
|
+
|
83
|
+
## Input
|
84
|
+
|
85
|
+
`::translate` can receive string, array or deep hash and will return the same, but translated.
|
86
|
+
|
87
|
+
```ruby
|
88
|
+
DeepLDiff.translate("test", from: "en", to: "es")
|
89
|
+
DeepLDiff.translate(%w[test language], from: "en", to: "es")
|
90
|
+
DeepLDiff.translate(
|
91
|
+
{ title: "test", values: { type: "frequent" } }, from: "en", to: "es"
|
92
|
+
)
|
93
|
+
```
|
94
|
+
|
95
|
+
See `DeepLDiff::Linearizer` for details.
|
96
|
+
|
97
|
+
## HTML
|
98
|
+
|
99
|
+
You can pass HTML as like as plain text:
|
100
|
+
|
101
|
+
```ruby
|
102
|
+
DeepLDiff.translate("<b>Black</b>", from: "en", to: "es")
|
103
|
+
```
|
104
|
+
|
105
|
+
## Very long texts
|
106
|
+
|
107
|
+
DeepL API has a limitation: query can not be longer than approximately 128 KB. If your text is really that long, multiple queries will be used to translate it automatically.
|
108
|
+
|
109
|
+
## Development
|
110
|
+
|
111
|
+
After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake spec` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
|
112
|
+
|
113
|
+
To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
|
114
|
+
|
115
|
+
## Contributing
|
116
|
+
|
117
|
+
Bug reports and pull requests are welcome on GitHub at https://github.com/wikiti/deepl-rb.
|
data/Rakefile
ADDED
data/deepl_diff.gemspec
ADDED
@@ -0,0 +1,55 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
lib = File.expand_path("lib", __dir__)
|
4
|
+
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
5
|
+
require "deepl_diff/version"
|
6
|
+
|
7
|
+
# rubocop:disable Metrics/BlockLength
|
8
|
+
Gem::Specification.new do |spec|
|
9
|
+
spec.name = "deepl_diff"
|
10
|
+
spec.version = DeepLDiff::VERSION
|
11
|
+
spec.authors = ["Islam Gagiev"]
|
12
|
+
spec.email = ["omniacinis@gmail.com"]
|
13
|
+
|
14
|
+
spec.summary = %(
|
15
|
+
DeepL API wrapper for Ruby which helps to translate only changes
|
16
|
+
between revisions of long texts.
|
17
|
+
|
18
|
+
)
|
19
|
+
spec.description = %(
|
20
|
+
DeepL API wrapper for Ruby which helps to translate only changes
|
21
|
+
between revisions of long texts.
|
22
|
+
)
|
23
|
+
spec.homepage = "https://github.com/Halvanhelv/deepl_diff"
|
24
|
+
|
25
|
+
if spec.respond_to?(:metadata)
|
26
|
+
spec.metadata["allowed_push_host"] = "https://rubygems.org"
|
27
|
+
else
|
28
|
+
raise "RubyGems 2.0 or newer is required to protect against " \
|
29
|
+
"public gem pushes."
|
30
|
+
end
|
31
|
+
|
32
|
+
spec.files = `git ls-files -z`.split("\x0").reject do |f|
|
33
|
+
f.match(%r{^(test|spec|features)/})
|
34
|
+
end
|
35
|
+
spec.bindir = "exe"
|
36
|
+
spec.executables = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
|
37
|
+
spec.require_paths = ["lib"]
|
38
|
+
|
39
|
+
spec.add_development_dependency "bundler", "~> 1.14"
|
40
|
+
spec.add_development_dependency "codeclimate-test-reporter", "~> 1.0.0"
|
41
|
+
spec.add_development_dependency "rake", "~> 10.0"
|
42
|
+
spec.add_development_dependency "rspec", "~> 3.0"
|
43
|
+
spec.add_development_dependency "rubocop"
|
44
|
+
spec.add_development_dependency "simplecov"
|
45
|
+
|
46
|
+
spec.add_dependency "connection_pool"
|
47
|
+
spec.add_dependency "deepl-rb"
|
48
|
+
spec.add_dependency "dry-initializer"
|
49
|
+
spec.add_dependency "ox"
|
50
|
+
spec.add_dependency "punkt-segmenter"
|
51
|
+
spec.add_dependency "ratelimit"
|
52
|
+
spec.add_dependency "redis"
|
53
|
+
spec.add_dependency "redis-namespace"
|
54
|
+
end
|
55
|
+
# rubocop:enable Metrics/BlockLength
|
@@ -0,0 +1,38 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
class DeepLDiff::Cache
|
4
|
+
extend Dry::Initializer
|
5
|
+
|
6
|
+
param :from
|
7
|
+
param :to
|
8
|
+
|
9
|
+
def cached_and_missing(values)
|
10
|
+
keys = values.map { |v| key(v) }
|
11
|
+
cached = cache_store.read_multi(keys)
|
12
|
+
missing = values.map.with_index { |v, i| v if cached[i].nil? }.compact
|
13
|
+
|
14
|
+
[cached, missing]
|
15
|
+
end
|
16
|
+
|
17
|
+
def store(values, cached, updates)
|
18
|
+
cached.map.with_index do |value, index|
|
19
|
+
value || store_value(values[index], updates.shift)
|
20
|
+
end
|
21
|
+
end
|
22
|
+
|
23
|
+
private
|
24
|
+
|
25
|
+
def store_value(value, translation)
|
26
|
+
cache_store.write(key(value), translation)
|
27
|
+
translation
|
28
|
+
end
|
29
|
+
|
30
|
+
def key(value)
|
31
|
+
hash = Digest::MD5.hexdigest(value.strip) # No matter how much spaces
|
32
|
+
"#{from}:#{to}:#{hash}"
|
33
|
+
end
|
34
|
+
|
35
|
+
def cache_store
|
36
|
+
DeepLDiff.cache_store
|
37
|
+
end
|
38
|
+
end
|
@@ -0,0 +1,56 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
class DeepLDiff::Chunker
|
4
|
+
extend ::Dry::Initializer
|
5
|
+
|
6
|
+
class Error < StandardError; end
|
7
|
+
|
8
|
+
Chunk = Struct.new(:values, :size)
|
9
|
+
|
10
|
+
param :values
|
11
|
+
option :limit, default: proc { MAX_CHUNK_SIZE }
|
12
|
+
option :count_limit, default: proc { COUNT_LIMIT }
|
13
|
+
|
14
|
+
def call
|
15
|
+
chunks.map(&:values)
|
16
|
+
end
|
17
|
+
|
18
|
+
def chunks
|
19
|
+
values.each_with_object([]) do |value, chunks|
|
20
|
+
validate_value_size(value)
|
21
|
+
|
22
|
+
tail = chunks.last
|
23
|
+
|
24
|
+
if next_chunk?(tail, value)
|
25
|
+
chunks << Chunk.new([], 0)
|
26
|
+
tail = chunks.last
|
27
|
+
end
|
28
|
+
|
29
|
+
update_chunk(tail, value)
|
30
|
+
end
|
31
|
+
end
|
32
|
+
|
33
|
+
private
|
34
|
+
|
35
|
+
def next_chunk?(tail, value)
|
36
|
+
tail.nil? ||
|
37
|
+
(size(value) + tail.size > limit) ||
|
38
|
+
tail.values.size > count_limit
|
39
|
+
end
|
40
|
+
|
41
|
+
def size(text)
|
42
|
+
CGI.escape(text).size
|
43
|
+
end
|
44
|
+
|
45
|
+
def update_chunk(chunk, value)
|
46
|
+
chunk.values << value
|
47
|
+
chunk.size = chunk.size + value.size
|
48
|
+
end
|
49
|
+
|
50
|
+
def validate_value_size(value)
|
51
|
+
raise Error, "Too long part #{value.size} > #{limit}" if value.size > limit
|
52
|
+
end
|
53
|
+
|
54
|
+
MAX_CHUNK_SIZE = 1700
|
55
|
+
COUNT_LIMIT = 300
|
56
|
+
end
|
@@ -0,0 +1,29 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
class DeepLDiff::Linearizer
|
4
|
+
class << self
|
5
|
+
def linearize(struct, array = [])
|
6
|
+
case struct
|
7
|
+
when Hash
|
8
|
+
struct.each { |_k, v| linearize(v, array) }
|
9
|
+
when Array
|
10
|
+
struct.each { |v| linearize(v, array) }
|
11
|
+
else
|
12
|
+
array << struct
|
13
|
+
end
|
14
|
+
|
15
|
+
array
|
16
|
+
end
|
17
|
+
|
18
|
+
def restore(struct, array)
|
19
|
+
case struct
|
20
|
+
when Hash
|
21
|
+
struct.transform_values { |v| restore(v, array) }
|
22
|
+
when Array
|
23
|
+
struct.map { |v| restore(v, array) }
|
24
|
+
else
|
25
|
+
array.shift
|
26
|
+
end
|
27
|
+
end
|
28
|
+
end
|
29
|
+
end
|
@@ -0,0 +1,26 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
class DeepLDiff::RedisCacheStore
|
4
|
+
extend Dry::Initializer
|
5
|
+
|
6
|
+
param :connection_pool
|
7
|
+
|
8
|
+
option :timeout, default: proc { 60 * 60 * 24 * 7 }
|
9
|
+
option :namespace, default: proc { DeepLDiff::CACHE_NAMESPACE }
|
10
|
+
|
11
|
+
def read_multi(keys)
|
12
|
+
redis { |redis| redis.mget(*keys) }
|
13
|
+
end
|
14
|
+
|
15
|
+
def write(key, value)
|
16
|
+
redis { |redis| redis.setex(key, timeout, value) }
|
17
|
+
end
|
18
|
+
|
19
|
+
private
|
20
|
+
|
21
|
+
def redis
|
22
|
+
connection_pool.with do |redis|
|
23
|
+
yield Redis::Namespace.new(namespace, redis: redis)
|
24
|
+
end
|
25
|
+
end
|
26
|
+
end
|
@@ -0,0 +1,22 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
class DeepLDiff::RedisRateLimiter
|
4
|
+
extend Dry::Initializer
|
5
|
+
|
6
|
+
class RateLimitExceeded < StandardError; end
|
7
|
+
|
8
|
+
param :connection_pool
|
9
|
+
param :threshold, default: proc { 8000 }
|
10
|
+
param :interval, default: proc { 60 }
|
11
|
+
|
12
|
+
option :namespace, default: proc { DeepLDiff::CACHE_NAMESPACE }
|
13
|
+
|
14
|
+
def check(size)
|
15
|
+
connection_pool.with do |redis|
|
16
|
+
rate_limit = Ratelimit.new(namespace, redis: redis)
|
17
|
+
raise RateLimitExceeded if rate_limit.exceeded?("call", threshold: threshold, interval: interval)
|
18
|
+
|
19
|
+
rate_limit.add size
|
20
|
+
end
|
21
|
+
end
|
22
|
+
end
|
@@ -0,0 +1,157 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
class DeepLDiff::Request
|
4
|
+
extend Dry::Initializer
|
5
|
+
extend Forwardable
|
6
|
+
|
7
|
+
param :values
|
8
|
+
param :options
|
9
|
+
|
10
|
+
def_delegators :DeepLDiff, :api, :cache_store, :rate_limiter
|
11
|
+
def_delegators :"DeepLDiff::Linearizer", :linearize, :restore
|
12
|
+
|
13
|
+
def call
|
14
|
+
validate_globals
|
15
|
+
|
16
|
+
return values if from == to || values.empty?
|
17
|
+
|
18
|
+
translation
|
19
|
+
end
|
20
|
+
|
21
|
+
private
|
22
|
+
|
23
|
+
def from
|
24
|
+
@from ||= options.delete(:from) || detect_language
|
25
|
+
end
|
26
|
+
|
27
|
+
def to
|
28
|
+
@to ||= options.delete(:to) { nil }
|
29
|
+
end
|
30
|
+
|
31
|
+
def detect_language
|
32
|
+
api.translate(text_tokens_texts.join(" ")[0..100], nil, to)
|
33
|
+
.detected_source_language.downcase
|
34
|
+
end
|
35
|
+
|
36
|
+
def validate_globals
|
37
|
+
raise "Set DeepLDiff.api before calling ::translate" unless api
|
38
|
+
return if cache_store
|
39
|
+
|
40
|
+
raise "Set DeepLDiff.cache_store before calling ::translate"
|
41
|
+
end
|
42
|
+
|
43
|
+
# Extracts flat text array
|
44
|
+
# => "Name", "<b>Good</b> boy"
|
45
|
+
#
|
46
|
+
# #values might be something like { name: "Name", bio: "<b>Good</b> boy" }
|
47
|
+
def texts
|
48
|
+
@texts ||= linearize(values)
|
49
|
+
end
|
50
|
+
|
51
|
+
# Converts each array item to token list
|
52
|
+
# => [..., [["<b>", :markup], ["Good", :text], ...]]
|
53
|
+
def tokens
|
54
|
+
@tokens ||= texts.map do |value|
|
55
|
+
DeepLDiff::Tokenizer.tokenize(value)
|
56
|
+
end
|
57
|
+
end
|
58
|
+
|
59
|
+
# Extracts text tokens from token list
|
60
|
+
# => { ..., "1_1" => "Good", 1_3 => "Boy", ... }
|
61
|
+
def text_tokens
|
62
|
+
@text_tokens ||= extract_text_tokens.to_h
|
63
|
+
end
|
64
|
+
|
65
|
+
def extract_text_tokens
|
66
|
+
tokens.each_with_object([]).with_index do |(group, result), group_index|
|
67
|
+
group.each_with_index do |(value, type), index|
|
68
|
+
result << ["#{group_index}_#{index}", value] if type == :text
|
69
|
+
end
|
70
|
+
end
|
71
|
+
end
|
72
|
+
|
73
|
+
# Extracts values from text tokens
|
74
|
+
# => [ ..., "Good", "Boy", ... ]
|
75
|
+
def text_tokens_texts
|
76
|
+
@text_tokens_texts ||= linearize(text_tokens).map(&:to_s).map(&:strip)
|
77
|
+
end
|
78
|
+
|
79
|
+
# Splits things requires translations to per-request chunks
|
80
|
+
# (groups less 2k sym)
|
81
|
+
# => [[ ..., "Good", "Boy", ... ]]
|
82
|
+
def chunks
|
83
|
+
@chunks ||= DeepLDiff::Chunker.new(text_tokens_texts).call
|
84
|
+
end
|
85
|
+
|
86
|
+
# Translates/loads from cache values from each chunk
|
87
|
+
# => [[ ..., "Horoshiy", "Malchik", ... ]]
|
88
|
+
def chunks_translated
|
89
|
+
@chunks_translated ||= chunks.map do |chunk|
|
90
|
+
cached, missing = cache.cached_and_missing(chunk)
|
91
|
+
if missing.empty?
|
92
|
+
cached
|
93
|
+
else
|
94
|
+
cache.store(chunk, cached, call_api(missing))
|
95
|
+
end
|
96
|
+
end
|
97
|
+
end
|
98
|
+
|
99
|
+
# Restores indexes for translated tokens
|
100
|
+
# => { ..., "1_1" => "Horoshiy", 1_3 => "Malchik", ... }
|
101
|
+
def text_tokens_translated
|
102
|
+
@text_tokens_texts_translated ||=
|
103
|
+
restore(text_tokens, chunks_translated.flatten)
|
104
|
+
end
|
105
|
+
|
106
|
+
# Restores tokens translated + adds same spacing as in source token
|
107
|
+
# => [[..., [ "Horoshiy", :text ], ...]]
|
108
|
+
# rubocop:disable Metrics/AbcSize
|
109
|
+
def tokens_translated
|
110
|
+
@tokens_translated ||= tokens.dup.tap do |tokens|
|
111
|
+
text_tokens_translated.each do |index, value|
|
112
|
+
group_index, index = index.split("_")
|
113
|
+
tokens[group_index.to_i][index.to_i][0] =
|
114
|
+
restore_spacing(tokens[group_index.to_i][index.to_i][0], value)
|
115
|
+
end
|
116
|
+
end
|
117
|
+
end
|
118
|
+
# rubocop:enable Metrics/AbcSize
|
119
|
+
|
120
|
+
def restore_spacing(source_value, value)
|
121
|
+
DeepLDiff::Spacing.restore(source_value, value)
|
122
|
+
end
|
123
|
+
|
124
|
+
# Restores texts from tokens
|
125
|
+
# [..., "<b>Horoshiy</b> Malchik", ...]
|
126
|
+
def texts_translated
|
127
|
+
@texts_translated ||= tokens_translated.map do |group|
|
128
|
+
group.map { |value, type| type == :text ? value : fix_ascii(value) }.join
|
129
|
+
end
|
130
|
+
end
|
131
|
+
|
132
|
+
# Final result
|
133
|
+
def translation
|
134
|
+
@translation ||= restore(values, texts_translated)
|
135
|
+
end
|
136
|
+
|
137
|
+
def call_api(values)
|
138
|
+
check_rate_limit(values)
|
139
|
+
[api.translate(values, from, to, **options)].flatten.map(&:text)
|
140
|
+
end
|
141
|
+
|
142
|
+
def cache
|
143
|
+
@cache ||= DeepLDiff::Cache.new(from, to)
|
144
|
+
end
|
145
|
+
|
146
|
+
def check_rate_limit(values)
|
147
|
+
return if rate_limiter.nil?
|
148
|
+
|
149
|
+
size = values.map(&:size).inject(0) { |sum, x| sum + x }
|
150
|
+
rate_limiter.check(size)
|
151
|
+
end
|
152
|
+
|
153
|
+
# Markup should not contain control characters
|
154
|
+
def fix_ascii(value)
|
155
|
+
value.gsub(/[\u0000-\u001F]/, " ")
|
156
|
+
end
|
157
|
+
end
|
@@ -0,0 +1,31 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
# Adds same count leading-trailing spaces left has to the right
|
4
|
+
class DeepLDiff::Spacing
|
5
|
+
class << self
|
6
|
+
# DeepLDiff::Spacing.restore(" a ", "Z") # => " Z "
|
7
|
+
def restore(left, right)
|
8
|
+
leading(left) + right.strip + trailing(left)
|
9
|
+
end
|
10
|
+
|
11
|
+
private
|
12
|
+
|
13
|
+
def spaces(count)
|
14
|
+
([" "] * count).join
|
15
|
+
end
|
16
|
+
|
17
|
+
def leading(value)
|
18
|
+
pos = value =~ /[^[:space:]]+/ui
|
19
|
+
return "" if pos.nil? || pos.zero?
|
20
|
+
|
21
|
+
value[0..(pos - 1)]
|
22
|
+
end
|
23
|
+
|
24
|
+
def trailing(value)
|
25
|
+
pos = value =~ /[[:space:]]+\z/ui
|
26
|
+
return "" if pos.nil?
|
27
|
+
|
28
|
+
value[pos..]
|
29
|
+
end
|
30
|
+
end
|
31
|
+
end
|
@@ -0,0 +1,159 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
class DeepLDiff::Tokenizer < ::Ox::Sax
|
4
|
+
def initialize(source)
|
5
|
+
@pos = nil
|
6
|
+
@source = source
|
7
|
+
@tokens = nil
|
8
|
+
@context = []
|
9
|
+
@sequence = []
|
10
|
+
@indicies = []
|
11
|
+
end
|
12
|
+
|
13
|
+
def instruct(target)
|
14
|
+
start_markup(target)
|
15
|
+
end
|
16
|
+
|
17
|
+
def end_instruct(target)
|
18
|
+
end_markup(target)
|
19
|
+
end
|
20
|
+
|
21
|
+
def start_element(name)
|
22
|
+
start_markup(name)
|
23
|
+
end
|
24
|
+
|
25
|
+
def end_element(name)
|
26
|
+
end_markup(name)
|
27
|
+
end
|
28
|
+
|
29
|
+
def attr(name, value)
|
30
|
+
return unless @context.last == :span
|
31
|
+
return unless name == :class && value == "notranslate"
|
32
|
+
return if notranslate?
|
33
|
+
|
34
|
+
@sequence[-1] = :notranslate
|
35
|
+
end
|
36
|
+
|
37
|
+
def text(value)
|
38
|
+
return if value == ""
|
39
|
+
|
40
|
+
@sequence << (SKIP.include?(@context.last) ? :markup : :text)
|
41
|
+
@indicies << @pos - 1
|
42
|
+
end
|
43
|
+
|
44
|
+
def tokens
|
45
|
+
@tokens ||= token_sequences_joined
|
46
|
+
.tap { |tokens| make_sentences_from_last_token(tokens) }
|
47
|
+
end
|
48
|
+
|
49
|
+
private
|
50
|
+
|
51
|
+
def token_sequences_joined
|
52
|
+
raw_tokens.each_with_object([]) do |token, tokens|
|
53
|
+
if tokens.empty? # Initial state
|
54
|
+
tokens << token
|
55
|
+
elsif tokens.last[1] == token[1]
|
56
|
+
# Join series of tokens of the same type into one
|
57
|
+
tokens.last[0].concat(token[0])
|
58
|
+
else
|
59
|
+
# If token before :markup is :text we need to split it into sentences
|
60
|
+
make_sentences_from_last_token(tokens)
|
61
|
+
tokens << token
|
62
|
+
end
|
63
|
+
end
|
64
|
+
end
|
65
|
+
|
66
|
+
def make_sentences_from_last_token(tokens)
|
67
|
+
return if tokens.empty?
|
68
|
+
|
69
|
+
tokens.concat(sentences(tokens.pop[0])) if tokens.last[1] == :text
|
70
|
+
end
|
71
|
+
|
72
|
+
# rubocop: disable Metrics/MethodLength
|
73
|
+
def sentences(value)
|
74
|
+
return [] if value.strip.empty?
|
75
|
+
|
76
|
+
boundaries =
|
77
|
+
Punkt::SentenceTokenizer
|
78
|
+
.new(value)
|
79
|
+
.sentences_from_text(value)
|
80
|
+
|
81
|
+
return [[value, :text]] if boundaries.size == 1
|
82
|
+
|
83
|
+
boundaries.map.with_index do |(left, right), index|
|
84
|
+
next_boundary = boundaries[index + 1]
|
85
|
+
right = next_boundary[0] - 1 if next_boundary
|
86
|
+
|
87
|
+
[value[left..right], :text]
|
88
|
+
end
|
89
|
+
end
|
90
|
+
# rubocop:enable Metrics/MethodLength
|
91
|
+
|
92
|
+
# Whether the sequence is between `:notranslate` and `:end_notranslate`
|
93
|
+
def notranslate?
|
94
|
+
@sequence.select { |item| item[/notranslate/] }.last == :notranslate
|
95
|
+
end
|
96
|
+
|
97
|
+
# Returns the item for last opened span
|
98
|
+
def end_span
|
99
|
+
return :markup unless notranslate?
|
100
|
+
|
101
|
+
opened_spans = @sequence
|
102
|
+
.reverse
|
103
|
+
.take_while { |item| item != :notranslate }
|
104
|
+
.map { |item| { span: 1, end_span: -1 }.fetch(item, 0) }
|
105
|
+
.reduce(0, :+)
|
106
|
+
|
107
|
+
opened_spans.positive? ? :end_span : :end_notranslate
|
108
|
+
end
|
109
|
+
|
110
|
+
def raw_tokens
|
111
|
+
@raw_tokens ||= @indicies.map.with_index do |i, n|
|
112
|
+
first = i
|
113
|
+
last = (@indicies[n + 1] || 0) - 1
|
114
|
+
value = fix_utf(@source.byteslice(first..last))
|
115
|
+
type = @sequence[n]
|
116
|
+
type = :text if INNER_SPANS.include?(type)
|
117
|
+
[value, type]
|
118
|
+
end
|
119
|
+
end
|
120
|
+
|
121
|
+
def fix_utf(value)
|
122
|
+
value.encode("UTF-8", undef: :replace, invalid: :replace, replace: " ")
|
123
|
+
end
|
124
|
+
|
125
|
+
def start_markup(name)
|
126
|
+
@context << name
|
127
|
+
@sequence << (if notranslate?
|
128
|
+
name == :span ? :span : :text
|
129
|
+
else
|
130
|
+
:markup
|
131
|
+
end)
|
132
|
+
@indicies << @pos - 1
|
133
|
+
end
|
134
|
+
|
135
|
+
def end_markup(name)
|
136
|
+
@context.pop
|
137
|
+
@sequence << (if notranslate?
|
138
|
+
name == :span ? end_span : :text
|
139
|
+
else
|
140
|
+
:markup
|
141
|
+
end)
|
142
|
+
@indicies << @pos - 1 unless @pos == @source.bytesize
|
143
|
+
end
|
144
|
+
|
145
|
+
class << self
|
146
|
+
def tokenize(value)
|
147
|
+
return [] if value.nil?
|
148
|
+
|
149
|
+
tokenizer = new(value).tap do |h|
|
150
|
+
Ox.sax_parse(h, StringIO.new(value), HTML_OPTIONS)
|
151
|
+
end
|
152
|
+
tokenizer.tokens
|
153
|
+
end
|
154
|
+
end
|
155
|
+
|
156
|
+
SKIP = %i[script style].freeze
|
157
|
+
INNER_SPANS = %i[notranslate span end_span end_notranslate].freeze
|
158
|
+
HTML_OPTIONS = { smart: true, skip: :skip_none }.freeze
|
159
|
+
end
|
data/lib/deepl_diff.rb
ADDED
@@ -0,0 +1,28 @@
|
|
1
|
+
# frozen_string_literal: true
|
2
|
+
|
3
|
+
require "ox"
|
4
|
+
require "punkt-segmenter"
|
5
|
+
require "dry/initializer"
|
6
|
+
require "deepl"
|
7
|
+
|
8
|
+
require "deepl_diff/version"
|
9
|
+
require "deepl_diff/tokenizer"
|
10
|
+
require "deepl_diff/linearizer"
|
11
|
+
require "deepl_diff/chunker"
|
12
|
+
require "deepl_diff/spacing"
|
13
|
+
require "deepl_diff/cache"
|
14
|
+
require "deepl_diff/redis_cache_store"
|
15
|
+
require "deepl_diff/redis_rate_limiter"
|
16
|
+
require "deepl_diff/request"
|
17
|
+
|
18
|
+
module DeepLDiff
|
19
|
+
class << self
|
20
|
+
attr_accessor :api, :cache_store, :rate_limiter
|
21
|
+
|
22
|
+
def translate(*args)
|
23
|
+
Request.new(*args).call
|
24
|
+
end
|
25
|
+
end
|
26
|
+
|
27
|
+
CACHE_NAMESPACE = "deepl-diff"
|
28
|
+
end
|
metadata
ADDED
@@ -0,0 +1,259 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: deepl_diff
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 1.0.2
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Islam Gagiev
|
8
|
+
autorequire:
|
9
|
+
bindir: exe
|
10
|
+
cert_chain: []
|
11
|
+
date: 2023-02-15 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: bundler
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - "~>"
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: '1.14'
|
20
|
+
type: :development
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - "~>"
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '1.14'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: codeclimate-test-reporter
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: 1.0.0
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: 1.0.0
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: rake
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - "~>"
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '10.0'
|
48
|
+
type: :development
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - "~>"
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '10.0'
|
55
|
+
- !ruby/object:Gem::Dependency
|
56
|
+
name: rspec
|
57
|
+
requirement: !ruby/object:Gem::Requirement
|
58
|
+
requirements:
|
59
|
+
- - "~>"
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '3.0'
|
62
|
+
type: :development
|
63
|
+
prerelease: false
|
64
|
+
version_requirements: !ruby/object:Gem::Requirement
|
65
|
+
requirements:
|
66
|
+
- - "~>"
|
67
|
+
- !ruby/object:Gem::Version
|
68
|
+
version: '3.0'
|
69
|
+
- !ruby/object:Gem::Dependency
|
70
|
+
name: rubocop
|
71
|
+
requirement: !ruby/object:Gem::Requirement
|
72
|
+
requirements:
|
73
|
+
- - ">="
|
74
|
+
- !ruby/object:Gem::Version
|
75
|
+
version: '0'
|
76
|
+
type: :development
|
77
|
+
prerelease: false
|
78
|
+
version_requirements: !ruby/object:Gem::Requirement
|
79
|
+
requirements:
|
80
|
+
- - ">="
|
81
|
+
- !ruby/object:Gem::Version
|
82
|
+
version: '0'
|
83
|
+
- !ruby/object:Gem::Dependency
|
84
|
+
name: simplecov
|
85
|
+
requirement: !ruby/object:Gem::Requirement
|
86
|
+
requirements:
|
87
|
+
- - ">="
|
88
|
+
- !ruby/object:Gem::Version
|
89
|
+
version: '0'
|
90
|
+
type: :development
|
91
|
+
prerelease: false
|
92
|
+
version_requirements: !ruby/object:Gem::Requirement
|
93
|
+
requirements:
|
94
|
+
- - ">="
|
95
|
+
- !ruby/object:Gem::Version
|
96
|
+
version: '0'
|
97
|
+
- !ruby/object:Gem::Dependency
|
98
|
+
name: connection_pool
|
99
|
+
requirement: !ruby/object:Gem::Requirement
|
100
|
+
requirements:
|
101
|
+
- - ">="
|
102
|
+
- !ruby/object:Gem::Version
|
103
|
+
version: '0'
|
104
|
+
type: :runtime
|
105
|
+
prerelease: false
|
106
|
+
version_requirements: !ruby/object:Gem::Requirement
|
107
|
+
requirements:
|
108
|
+
- - ">="
|
109
|
+
- !ruby/object:Gem::Version
|
110
|
+
version: '0'
|
111
|
+
- !ruby/object:Gem::Dependency
|
112
|
+
name: deepl-rb
|
113
|
+
requirement: !ruby/object:Gem::Requirement
|
114
|
+
requirements:
|
115
|
+
- - ">="
|
116
|
+
- !ruby/object:Gem::Version
|
117
|
+
version: '0'
|
118
|
+
type: :runtime
|
119
|
+
prerelease: false
|
120
|
+
version_requirements: !ruby/object:Gem::Requirement
|
121
|
+
requirements:
|
122
|
+
- - ">="
|
123
|
+
- !ruby/object:Gem::Version
|
124
|
+
version: '0'
|
125
|
+
- !ruby/object:Gem::Dependency
|
126
|
+
name: dry-initializer
|
127
|
+
requirement: !ruby/object:Gem::Requirement
|
128
|
+
requirements:
|
129
|
+
- - ">="
|
130
|
+
- !ruby/object:Gem::Version
|
131
|
+
version: '0'
|
132
|
+
type: :runtime
|
133
|
+
prerelease: false
|
134
|
+
version_requirements: !ruby/object:Gem::Requirement
|
135
|
+
requirements:
|
136
|
+
- - ">="
|
137
|
+
- !ruby/object:Gem::Version
|
138
|
+
version: '0'
|
139
|
+
- !ruby/object:Gem::Dependency
|
140
|
+
name: ox
|
141
|
+
requirement: !ruby/object:Gem::Requirement
|
142
|
+
requirements:
|
143
|
+
- - ">="
|
144
|
+
- !ruby/object:Gem::Version
|
145
|
+
version: '0'
|
146
|
+
type: :runtime
|
147
|
+
prerelease: false
|
148
|
+
version_requirements: !ruby/object:Gem::Requirement
|
149
|
+
requirements:
|
150
|
+
- - ">="
|
151
|
+
- !ruby/object:Gem::Version
|
152
|
+
version: '0'
|
153
|
+
- !ruby/object:Gem::Dependency
|
154
|
+
name: punkt-segmenter
|
155
|
+
requirement: !ruby/object:Gem::Requirement
|
156
|
+
requirements:
|
157
|
+
- - ">="
|
158
|
+
- !ruby/object:Gem::Version
|
159
|
+
version: '0'
|
160
|
+
type: :runtime
|
161
|
+
prerelease: false
|
162
|
+
version_requirements: !ruby/object:Gem::Requirement
|
163
|
+
requirements:
|
164
|
+
- - ">="
|
165
|
+
- !ruby/object:Gem::Version
|
166
|
+
version: '0'
|
167
|
+
- !ruby/object:Gem::Dependency
|
168
|
+
name: ratelimit
|
169
|
+
requirement: !ruby/object:Gem::Requirement
|
170
|
+
requirements:
|
171
|
+
- - ">="
|
172
|
+
- !ruby/object:Gem::Version
|
173
|
+
version: '0'
|
174
|
+
type: :runtime
|
175
|
+
prerelease: false
|
176
|
+
version_requirements: !ruby/object:Gem::Requirement
|
177
|
+
requirements:
|
178
|
+
- - ">="
|
179
|
+
- !ruby/object:Gem::Version
|
180
|
+
version: '0'
|
181
|
+
- !ruby/object:Gem::Dependency
|
182
|
+
name: redis
|
183
|
+
requirement: !ruby/object:Gem::Requirement
|
184
|
+
requirements:
|
185
|
+
- - ">="
|
186
|
+
- !ruby/object:Gem::Version
|
187
|
+
version: '0'
|
188
|
+
type: :runtime
|
189
|
+
prerelease: false
|
190
|
+
version_requirements: !ruby/object:Gem::Requirement
|
191
|
+
requirements:
|
192
|
+
- - ">="
|
193
|
+
- !ruby/object:Gem::Version
|
194
|
+
version: '0'
|
195
|
+
- !ruby/object:Gem::Dependency
|
196
|
+
name: redis-namespace
|
197
|
+
requirement: !ruby/object:Gem::Requirement
|
198
|
+
requirements:
|
199
|
+
- - ">="
|
200
|
+
- !ruby/object:Gem::Version
|
201
|
+
version: '0'
|
202
|
+
type: :runtime
|
203
|
+
prerelease: false
|
204
|
+
version_requirements: !ruby/object:Gem::Requirement
|
205
|
+
requirements:
|
206
|
+
- - ">="
|
207
|
+
- !ruby/object:Gem::Version
|
208
|
+
version: '0'
|
209
|
+
description: "\nDeepL API wrapper for Ruby which helps to translate only changes\nbetween
|
210
|
+
revisions of long texts.\n "
|
211
|
+
email:
|
212
|
+
- omniacinis@gmail.com
|
213
|
+
executables: []
|
214
|
+
extensions: []
|
215
|
+
extra_rdoc_files: []
|
216
|
+
files:
|
217
|
+
- ".gitignore"
|
218
|
+
- ".rspec"
|
219
|
+
- ".rubocop.yml"
|
220
|
+
- ".travis.yml"
|
221
|
+
- Gemfile
|
222
|
+
- README.md
|
223
|
+
- Rakefile
|
224
|
+
- deepl_diff.gemspec
|
225
|
+
- lib/deepl_diff.rb
|
226
|
+
- lib/deepl_diff/cache.rb
|
227
|
+
- lib/deepl_diff/chunker.rb
|
228
|
+
- lib/deepl_diff/linearizer.rb
|
229
|
+
- lib/deepl_diff/redis_cache_store.rb
|
230
|
+
- lib/deepl_diff/redis_rate_limiter.rb
|
231
|
+
- lib/deepl_diff/request.rb
|
232
|
+
- lib/deepl_diff/spacing.rb
|
233
|
+
- lib/deepl_diff/tokenizer.rb
|
234
|
+
- lib/deepl_diff/version.rb
|
235
|
+
homepage: https://github.com/Halvanhelv/deepl_diff
|
236
|
+
licenses: []
|
237
|
+
metadata:
|
238
|
+
allowed_push_host: https://rubygems.org
|
239
|
+
post_install_message:
|
240
|
+
rdoc_options: []
|
241
|
+
require_paths:
|
242
|
+
- lib
|
243
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
244
|
+
requirements:
|
245
|
+
- - ">="
|
246
|
+
- !ruby/object:Gem::Version
|
247
|
+
version: '0'
|
248
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
249
|
+
requirements:
|
250
|
+
- - ">="
|
251
|
+
- !ruby/object:Gem::Version
|
252
|
+
version: '0'
|
253
|
+
requirements: []
|
254
|
+
rubygems_version: 3.1.6
|
255
|
+
signing_key:
|
256
|
+
specification_version: 4
|
257
|
+
summary: DeepL API wrapper for Ruby which helps to translate only changes between
|
258
|
+
revisions of long texts.
|
259
|
+
test_files: []
|