ncode-syosetu-polly 0.2.1 → 0.2.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 9e4ff3806326fef794399435bbf177031e40ad78
-  data.tar.gz: e60e6e5cecbf57f22be4d0fc76b2a24f590e7a30
+  metadata.gz: 07b6ac5d9bc4b8df8b12b2511e2f7b0f75abaf1e
+  data.tar.gz: 2bcd10de8e21838143005136d35762b9bd1d039f
 SHA512:
-  metadata.gz: bd35ec825273a73a36399d663d797a54bc2f1060960e9bb8ebe4297f02c5540f8ee8b1b81891ba84c34af2d7c05842c1baee4d7f489079833680d45cd4290c32
-  data.tar.gz: 1a06cf7c207a7e92a0ad84e12047ac70b815403f12693d3bda215e6f2a191300a7016684b2f7ef1bf83b7369c07a5bdf8256a5056601ace8e4e4efae9820a448
+  metadata.gz: 508384495e2635f45a26ed40b8f27d7632b2ed9c040c4f3ba4dfced45a8a950defaf3fb79953a16c82723655c8c1dd014a3e64e38793f8a631c42fc2880b29d7
+  data.tar.gz: c337d14dca9d9adef7c54fc249f0c87d5a784a334efaa09c991f0ac172d0c30bc58a483b7945e0e3e48a4e86f78c12ccdb251be3df06ba8d8069c1bba64875d0

data/lib/ncode_syosetu/builder/polly.rb

@@ -3,11 +3,12 @@ require "sanitize"
 require "nokogiri"
 require "expeditor"
 require "concurrent"
+require "htmlentities"
 
 module NcodeSyosetu
   module Builder
     class Polly
-      POLLY_TEXT_LENGTH_LIMIT = 1500
+      POLLY_TEXT_LENGTH_LIMIT = 1000
 
       attr_reader :sample_rate, :client, :logger, :service
 
@@ -15,11 +16,13 @@ module NcodeSyosetu
         options[:region] ||= "us-west-2"
         @logger = options.delete(:logger)
         @sample_rate = options.delete(:sample_rate) || "16000"
+        @max_threads = options.delete(:max_threads) || 10
         @client = Aws::Polly::Client.new(options)
+        @htmlentities = HTMLEntities.new
         @service = Expeditor::Service.new(
           executor: Concurrent::ThreadPoolExecutor.new(
             min_threads: 0,
-            max_threads: 10,
+            max_threads: @max_threads,
           )
         )
       end
@@ -33,22 +36,28 @@ module NcodeSyosetu
         dirname = File.dirname(path)
         basename = File.basename(path, ".mp3")
         tmpdir = File.join(dirname, basename)
-        FileUtils.mkdir(tmpdir)
+        FileUtils.mkdir_p(tmpdir)
 
         ssmls.each_with_index do |ssml, i|
-
-          File.write(File.join(tmpdir, "#{basename}-#{i}.ssml"), ssml)
+          tmp_ssml_path = File.join(tmpdir, "#{basename}-#{i}.ssml")
+          File.write(tmp_ssml_path, ssml)
           tmp_path = File.join(tmpdir, "#{basename}-#{i}.mp3")
           command = Expeditor::Command.new(service: service) do
             logger.info("#{tmp_path}...") if logger
-            client.synthesize_speech(
-              response_target: tmp_path,
-              output_format: "mp3",
-              sample_rate: sample_rate,
-              text: ssml,
-              text_type: "ssml",
-              voice_id: "Mizuki",
-            )
+            begin
+              client.synthesize_speech(
+                response_target: tmp_path,
+                output_format: "mp3",
+                sample_rate: sample_rate,
+                text: ssml,
+                text_type: "ssml",
+                voice_id: "Mizuki",
+              )
+            rescue => e
+              logger.error("#{e.message}\n#{ssml}")
+              logger.error("#{e.message}: #{tmp_ssml_path}\n#{ssml}")
+              raise e
+            end
           end
           command.start
           commands << command
@@ -86,13 +95,16 @@ module NcodeSyosetu
         while elements.size > 0 do
           element = elements.shift
 
-          unless element.is_a?(Nokogiri::XML::Text) || element.is_a?(String)
-            buffer.print(element.to_s)
-            next
+          case element
+            when Nokogiri::XML::Text
+              text = @htmlentities.encode(element.text)
+            when String
+              text = @htmlentities.encode(element)
+            else
+              buffer.print(element.to_s)
+              next
           end
 
-          text = element.to_s
-
           if text.size > POLLY_TEXT_LENGTH_LIMIT
             elements = text.chars.each_slice(POLLY_TEXT_LENGTH_LIMIT).map(&:join) + elements
             next
@@ -103,7 +115,7 @@ module NcodeSyosetu
             buffer = StringIO.new
             text_count = 0
           end
-          buffer.print(element.to_s)
+          buffer.print(text)
           text_count += text.size
         end
         results << buffer.string if buffer.size > 0

data/ncode-syosetu-polly.gemspec

@@ -24,6 +24,7 @@ Gem::Specification.new do |spec|
   spec.add_dependency "sanitize", "~> 4.4"
   spec.add_dependency "nokogiri", ">= 1.6"
   spec.add_dependency "expeditor", "~> 0.5.0"
+  spec.add_dependency "htmlentities"
 
   spec.add_development_dependency "bundler", "~> 1.13"
   spec.add_development_dependency "rake", "~> 10.0"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: ncode-syosetu-polly
 version: !ruby/object:Gem::Version
-  version: 0.2.1
+  version: 0.2.2
 platform: ruby
 authors:
 - hogelog
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2017-01-08 00:00:00.000000000 Z
+date: 2017-01-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ncode-syosetu-core
@@ -16,28 +16,28 @@ dependencies:
     requirements:
     - - '='
       - !ruby/object:Gem::Version
-        version: 0.2.1
+        version: 0.2.2
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - '='
       - !ruby/object:Gem::Version
-        version: 0.2.1
+        version: 0.2.2
 - !ruby/object:Gem::Dependency
   name: ncode-syosetu-ssml
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - '='
       - !ruby/object:Gem::Version
-        version: 0.2.1
+        version: 0.2.2
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - '='
       - !ruby/object:Gem::Version
-        version: 0.2.1
+        version: 0.2.2
 - !ruby/object:Gem::Dependency
   name: aws-sdk-polly
   requirement: !ruby/object:Gem::Requirement
@@ -94,6 +94,20 @@ dependencies:
     - - "~>"
       - !ruby/object:Gem::Version
         version: 0.5.0
+- !ruby/object:Gem::Dependency
+  name: htmlentities
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement