jieba-rb 5.0.0

data/lib/jieba_rb/version.rb

@@ -0,0 +1,3 @@
+module JiebaRb
+  VERSION = "5.0.0"
+end

data/test/test_keyword.rb

@@ -0,0 +1,17 @@
+# coding: utf-8
+require 'minitest/autorun'
+require 'jieba_rb'
+class JiebaTest < Minitest::Test
+  def test_keywords
+    keyword = JiebaRb::Keyword.new
+    keywords_weights = keyword.extract "我是拖拉机学院手扶拖拉机专业的。不用多久，我就会升职加薪，当上CEO，走上人生巅峰。", 5
+
+    assert_equal [["CEO",
+                   11.739204307083542],
+                  ["升职", 10.8561552143],
+                  ["加薪", 10.642581114],
+                  ["手扶拖拉机", 10.0088573539],
+                  ["巅峰", 9.49395840471]], keywords_weights
+
+  end
+end

data/test/test_segment.rb

@@ -0,0 +1,32 @@
+# coding: utf-8
+require 'minitest/autorun'
+require 'jieba_rb'
+class JiebaTest < Minitest::Test
+  def test_mix_segment
+    seg = JiebaRb::Segment.new
+    words = seg.cut "我来到南京市长江大桥"
+    assert_equal %w(我 来到 南京市 长江大桥), words
+
+    words = seg.cut "令狐冲是云计算行业的专家"
+    assert_equal %w(令狐冲 是 云 计算 行业 的 专家), words
+  end
+
+  def test_mix_segment_with_user_dict
+    seg = JiebaRb::Segment.new user_dict: :default
+    words = seg.cut "令狐冲是云计算行业的专家"
+    assert_equal %w(令狐冲 是 云计算 行业 的 专家), words
+  end
+
+  def test_hmm_segment
+    seg = JiebaRb::Segment.new mode: :hmm
+    words = seg.cut "令狐冲是云计算行业的专家"
+    assert_equal %w(令狐冲 是 云计算 行业 的 专家), words
+  end
+
+  def test_max_prob_segment
+    seg = JiebaRb::Segment.new mode: :mp
+    words = seg.cut "令狐冲是云计算行业的专家"
+    assert_equal %w(令狐冲 是 云 计算 行业 的 专家), words
+  end
+
+end

data/test/test_tagging.rb

@@ -0,0 +1,22 @@
+# coding: utf-8
+require 'minitest/autorun'
+require 'jieba_rb'
+class JiebaTest < Minitest::Test
+  def test_tagging
+    tagging = JiebaRb::Tagging.new
+    pairs = tagging.tag "我是蓝翔技工拖拉机学院手扶拖拉机专业的。"
+    assert_equal [{"我"=>"r"}, {"是"=>"v"}, {"蓝翔"=>"x"}, {"技工"=>"n"}, {"拖拉机"=>"n"}, {"学院"=>"n"}, {"手扶拖拉机"=>"n"}, {"专业"=>"n"}, {"的"=>"uj"}, {"。"=>"x"}], pairs
+  end
+
+  def test_tagging_with_user_dict
+    tagging = JiebaRb::Tagging.new user_dict: :default
+    pairs = tagging.tag "我是蓝翔技工拖拉机学院手扶拖拉机专业的。"
+    assert_equal [{"我"=>"r"}, {"是"=>"v"}, {"蓝翔"=>"nz"}, {"技工"=>"n"}, {"拖拉机"=>"n"}, {"学院"=>"n"}, {"手扶拖拉机"=>"n"}, {"专业"=>"n"}, {"的"=>"uj"}, {"。"=>"x"}], pairs
+  end
+
+  def test_tagging_with_real_user_dict
+    tagging = JiebaRb::Tagging.new user_dict: "#{File.dirname(__FILE__)}/user.dict.utf8"
+    pairs = tagging.tag "分包合同"
+    assert_equal [{"分包合同"=>"x"}], pairs
+  end
+end

data/test/user.dict.utf8

@@ -0,0 +1,23 @@
+费用中心
+增值税专用发票
+应付单
+二级策划
+一级策划
+产值结算
+审批人
+专业负责人
+合同归档
+分包合同
+在职证明
+收入证明
+小孩随员
+采购发票
+报销单
+年假
+社保公积金
+积分办理
+怎么发
+落户申请
+软件仓库
+软件库
+海归落户

metadata

@@ -0,0 +1,219 @@
+--- !ruby/object:Gem::Specification
+name: jieba-rb
+version: !ruby/object:Gem::Version
+  version: 5.0.0
+platform: ruby
+authors:
+- Chris Li
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2019-08-08 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rake-compiler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: minitest
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+description: cppjieba binding for ruby
+email:
+- liqi8822@gmail.com
+executables: []
+extensions:
+- ext/jieba/extconf.rb
+extra_rdoc_files: []
+files:
+- ".gitignore"
+- ".gitmodules"
+- ".travis.yml"
+- Gemfile
+- LICENSE.txt
+- README.md
+- Rakefile
+- ext/cppjieba/.gitignore
+- ext/cppjieba/.travis.yml
+- ext/cppjieba/CMakeLists.txt
+- ext/cppjieba/ChangeLog.md
+- ext/cppjieba/README.md
+- ext/cppjieba/README_EN.md
+- ext/cppjieba/appveyor.yml
+- ext/cppjieba/deps/CMakeLists.txt
+- ext/cppjieba/deps/gtest/CMakeLists.txt
+- ext/cppjieba/deps/gtest/include/gtest/gtest-death-test.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest-message.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest-param-test.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest-param-test.h.pump
+- ext/cppjieba/deps/gtest/include/gtest/gtest-printers.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest-spi.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest-test-part.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest-typed-test.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest_pred_impl.h
+- ext/cppjieba/deps/gtest/include/gtest/gtest_prod.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-death-test-internal.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-filepath.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-internal.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-linked_ptr.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-param-util-generated.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-param-util-generated.h.pump
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-param-util.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-port.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-string.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-tuple.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-tuple.h.pump
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-type-util.h
+- ext/cppjieba/deps/gtest/include/gtest/internal/gtest-type-util.h.pump
+- ext/cppjieba/deps/gtest/src/.deps/.dirstamp
+- ext/cppjieba/deps/gtest/src/.deps/gtest-all.Plo
+- ext/cppjieba/deps/gtest/src/.deps/gtest_main.Plo
+- ext/cppjieba/deps/gtest/src/.dirstamp
+- ext/cppjieba/deps/gtest/src/gtest-all.cc
+- ext/cppjieba/deps/gtest/src/gtest-death-test.cc
+- ext/cppjieba/deps/gtest/src/gtest-filepath.cc
+- ext/cppjieba/deps/gtest/src/gtest-internal-inl.h
+- ext/cppjieba/deps/gtest/src/gtest-port.cc
+- ext/cppjieba/deps/gtest/src/gtest-printers.cc
+- ext/cppjieba/deps/gtest/src/gtest-test-part.cc
+- ext/cppjieba/deps/gtest/src/gtest-typed-test.cc
+- ext/cppjieba/deps/gtest/src/gtest.cc
+- ext/cppjieba/deps/gtest/src/gtest_main.cc
+- ext/cppjieba/deps/limonp/ArgvContext.hpp
+- ext/cppjieba/deps/limonp/BlockingQueue.hpp
+- ext/cppjieba/deps/limonp/BoundedBlockingQueue.hpp
+- ext/cppjieba/deps/limonp/BoundedQueue.hpp
+- ext/cppjieba/deps/limonp/Closure.hpp
+- ext/cppjieba/deps/limonp/Colors.hpp
+- ext/cppjieba/deps/limonp/Condition.hpp
+- ext/cppjieba/deps/limonp/Config.hpp
+- ext/cppjieba/deps/limonp/FileLock.hpp
+- ext/cppjieba/deps/limonp/ForcePublic.hpp
+- ext/cppjieba/deps/limonp/LocalVector.hpp
+- ext/cppjieba/deps/limonp/Logging.hpp
+- ext/cppjieba/deps/limonp/Md5.hpp
+- ext/cppjieba/deps/limonp/MutexLock.hpp
+- ext/cppjieba/deps/limonp/NonCopyable.hpp
+- ext/cppjieba/deps/limonp/StdExtension.hpp
+- ext/cppjieba/deps/limonp/StringUtil.hpp
+- ext/cppjieba/deps/limonp/Thread.hpp
+- ext/cppjieba/deps/limonp/ThreadPool.hpp
+- ext/cppjieba/dict/README.md
+- ext/cppjieba/dict/hmm_model.utf8
+- ext/cppjieba/dict/idf.utf8
+- ext/cppjieba/dict/jieba.dict.utf8
+- ext/cppjieba/dict/pos_dict/char_state_tab.utf8
+- ext/cppjieba/dict/pos_dict/prob_emit.utf8
+- ext/cppjieba/dict/pos_dict/prob_start.utf8
+- ext/cppjieba/dict/pos_dict/prob_trans.utf8
+- ext/cppjieba/dict/stop_words.utf8
+- ext/cppjieba/dict/user.dict.utf8
+- ext/cppjieba/include/cppjieba/DictTrie.hpp
+- ext/cppjieba/include/cppjieba/FullSegment.hpp
+- ext/cppjieba/include/cppjieba/HMMModel.hpp
+- ext/cppjieba/include/cppjieba/HMMSegment.hpp
+- ext/cppjieba/include/cppjieba/Jieba.hpp
+- ext/cppjieba/include/cppjieba/KeywordExtractor.hpp
+- ext/cppjieba/include/cppjieba/MPSegment.hpp
+- ext/cppjieba/include/cppjieba/MixSegment.hpp
+- ext/cppjieba/include/cppjieba/PosTagger.hpp
+- ext/cppjieba/include/cppjieba/PreFilter.hpp
+- ext/cppjieba/include/cppjieba/QuerySegment.hpp
+- ext/cppjieba/include/cppjieba/SegmentBase.hpp
+- ext/cppjieba/include/cppjieba/SegmentTagged.hpp
+- ext/cppjieba/include/cppjieba/TextRankExtractor.hpp
+- ext/cppjieba/include/cppjieba/Trie.hpp
+- ext/cppjieba/include/cppjieba/Unicode.hpp
+- ext/jieba/extconf.rb
+- ext/jieba/jieba.c
+- ext/jieba/jieba.h
+- ext/jieba/keyword.cc
+- ext/jieba/keyword.h
+- ext/jieba/segment.cc
+- ext/jieba/segment.h
+- ext/jieba/tagging.cc
+- ext/jieba/tagging.h
+- jieba_rb.gemspec
+- lib/jieba-rb.rb
+- lib/jieba_rb/version.rb
+- test/test_keyword.rb
+- test/test_segment.rb
+- test/test_tagging.rb
+- test/user.dict.utf8
+homepage: https://github.com/Xu-Zhiqing/jieba_rb
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: 1.9.2
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.0.3
+signing_key: 
+specification_version: 4
+summary: cppjieba binding for ruby
+test_files:
+- test/test_keyword.rb
+- test/test_segment.rb
+- test/test_tagging.rb
+- test/user.dict.utf8