srx-languagetool 0.1.0 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 13c9ee40fd02a436a0016cade2b0aeeecb615c20ff912e703558800a9deffbec
-  data.tar.gz: 4c81d6bffbd73860fa60d5e9b65ee0a62428454dd69fe73452d6a103cf65abe1
+  metadata.gz: d929b515d7e87e260c01373e1f5d6c35bf35fa05f8fe29fea39ba18d0078e120
+  data.tar.gz: ee821f5cfb3b4fea3d347c3ef3583459da60d044c458aa5ad907b378d5eef963
 SHA512:
-  metadata.gz: e0dabae1c8f62bcd5133c566c750e99f5a473b9ff7c19d1981e92c4aec99155adf99ddc0158a3bf128869d678e7862addc2344b3b6f42c779af141ae84030068
-  data.tar.gz: f89e5779174040869de0a5f25d4e73380a87a051fc5da969964d544f62a9919e8cc952fcf28a6561aedc89006c38ec4d56ebf8bef0a230f60b2fc89d65f0d841
+  metadata.gz: 2733d28e0ed99d6b5fd0c15b30311aed10059c34c8c411de051901ca545ed6a4eb299e0695190817b45d6691f74b3d78215d6a5ce68c589b0776e25ca3c3dada
+  data.tar.gz: 4344f1b2dc867dcbffb8db8b2db95da458de718f0481814e8dd23965c2b3831199702333637d56017566388de5da0d745995276e0376c9e16e63dc9f15cd8d10

data/.github/workflows/main.yml

@@ -10,10 +10,10 @@ jobs:
     - name: Set up Ruby
       uses: ruby/setup-ruby@v1
       with:
-        ruby-version: 2.7.2
+        ruby-version: 2.7.3
     - name: Install
       run: |
-        gem install bundler -v 2.2.9
+        gem install bundler -v 2.2.16
         bundle install
     - name: Type check
       run: bundle exec solargraph typecheck --level typed

data/.gitignore

@@ -6,3 +6,4 @@
 /pkg/
 /spec/reports/
 /tmp/
+/vendor

data/.ruby-version

@@ -0,0 +1 @@
+2.7.3

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 ## [Unreleased]
 
+## [0.2.0] - 2021-04-15
+
+- Update rules to LanguageTool 5.3
+
 ## [0.1.0] - 2021-02-13
 
 - Initial release

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    srx-languagetool (0.1.0)
+    srx-languagetool (0.2.0)
       srx (< 1.0)
 
 GEM
@@ -14,28 +14,28 @@ GEM
     diff-lcs (1.4.4)
     e2mmap (0.1.0)
     jaro_winkler (1.5.4)
-    kramdown (2.3.0)
+    kramdown (2.3.1)
       rexml
     kramdown-parser-gfm (1.1.0)
       kramdown (~> 2.0)
-    minitest (5.14.3)
-    nokogiri (1.11.1-x86_64-darwin)
+    minitest (5.14.4)
+    nokogiri (1.11.3-x86_64-darwin)
       racc (~> 1.4)
     parallel (1.20.1)
-    parser (3.0.0.0)
+    parser (3.0.1.0)
       ast (~> 2.4.1)
     racc (1.5.2)
     rainbow (3.0.0)
     rake (13.0.3)
-    regexp_parser (2.0.3)
+    regexp_parser (2.1.1)
     reverse_markdown (2.0.0)
       nokogiri
-    rexml (3.2.4)
+    rexml (3.2.5)
     rspec-expectations (3.10.1)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.10.0)
     rspec-support (3.10.2)
-    rubocop (1.9.1)
+    rubocop (1.12.1)
       parallel (~> 1.10)
       parser (>= 3.0.0.0)
       rainbow (>= 2.2.2, < 4.0)
@@ -47,7 +47,7 @@ GEM
     rubocop-ast (1.4.1)
       parser (>= 2.7.1.5)
     ruby-progressbar (1.11.0)
-    solargraph (0.40.3)
+    solargraph (0.40.4)
       backport (~> 1.1)
       benchmark
       bundler (>= 1.17.2)
@@ -61,7 +61,7 @@ GEM
       thor (~> 1.0)
       tilt (~> 2.0)
       yard (~> 0.9, >= 0.9.24)
-    srx (0.1.0)
+    srx (0.6.0)
       nokogiri (~> 1.11)
     thor (1.1.0)
     tilt (2.0.10)
@@ -81,4 +81,4 @@ DEPENDENCIES
   srx-languagetool!
 
 BUNDLED WITH
-   2.2.9
+   2.2.16

data/README.md

@@ -29,7 +29,7 @@ For detailed usage information, please see
 [srx-ruby](https://github.com/amake/srx-ruby).
 
 ```ruby
-require 'srx-languagetool'
+require 'srx/languagetool'
 
 data = Srx::Data.languagetool
 engine = Srx::Engine.new(data)

data/lib/srx/languagetool/version.rb

@@ -2,6 +2,6 @@
 
 module Srx
   module Languagetool
-    VERSION = '0.1.0'
+    VERSION = '0.2.0'
   end
 end

data/lib/srx/segment.srx

@@ -1368,7 +1368,7 @@
 <afterbreak>\p{Ll}+</afterbreak>
 </rule>
 <rule break="no">
-<beforebreak>[\.\s]\p{L}{1,2}\.\s</beforebreak>
+<beforebreak>[\.\s](?!(on|it|of|to|be|by|at|he|we|so|do|if|up|my|me|us|go|am))\p{L}{1,2}\.\s</beforebreak><!-- not 'no'/'in', these could be abbreviations-->
 <afterbreak>[\p{N}\p{Ll}]</afterbreak>
 </rule>
 <rule break="no">
@@ -1705,6 +1705,14 @@
 <beforebreak>\bmax\.\s</beforebreak>
 <afterbreak>\p{Ll}</afterbreak>
 </rule>
+<rule break="yes">
+<beforebreak>[?!.]['"\u00BB\u2019\u201D\u203A\u00AB\p{Pe}\u0002]\s</beforebreak>
+<afterbreak>[A-Z][a-z]</afterbreak>
+</rule>
+<rule break="yes">
+<beforebreak>[?!.]\s</beforebreak>
+<afterbreak>['"\u00BB\u2019\u201D\u203A\u00AB\p{Pe}\u0002][A-Z][a-z]</afterbreak>
+</rule>
 </languagerule>
 <languagerule languagerulename="Slovak">
 <rule break="no">
@@ -4554,11 +4562,11 @@
 </rule>
 <!-- Abbreviations that can finish sentences -->
 <rule break="no">
-<beforebreak>\bs\.\s</beforebreak>
+<beforebreak>\b(s|ca)\.\s</beforebreak>
 <afterbreak>[XIV]+\b</afterbreak>
 </rule>
 <rule break="no">
-<beforebreak>\b(min|m)\.\s</beforebreak>
+<beforebreak>\b(min|m|ca)\.\s</beforebreak>
 <afterbreak>[0-9]+\b</afterbreak>
 </rule>
 <rule break="no">
@@ -4658,11 +4666,11 @@
 </rule>
 <!-- Abbreviations that can finish sentences -->
 <rule break="no">
-<beforebreak>\bs\.\s</beforebreak>
+<beforebreak>\b(s|ca)\.\s</beforebreak>
 <afterbreak>[XIV]+\b</afterbreak>
 </rule>
 <rule break="no">
-<beforebreak>\b(min|m)\.\s</beforebreak>
+<beforebreak>\b(min|m|ca)\.\s</beforebreak>
 <afterbreak>[0-9]+\b</afterbreak>
 </rule>
 <rule break="no">
@@ -4862,7 +4870,7 @@
 </rule>
 <!-- don't split at cases like "Friedrich II. wird auch..." -->
 <rule break="no">
-<beforebreak>[\s ][IVX]+\.\s</beforebreak>
+<beforebreak>[\s ][IVX]+\.\s</beforebreak>
 <afterbreak>[^\p{Lu}]+</afterbreak>
 </rule>
 <!-- don't split at cases like "im 13. oder 14. Jahrhundert" -->
@@ -6279,7 +6287,7 @@
 <afterbreak></afterbreak>
 </rule>
 <rule break="no">
-<beforebreak>[^\.]\s[ضصثقفغعهخحجچشسیبلاتنمکگ\ظطزرذدپوًٌٍَُِّْA-Z]\.\s</beforebreak>
+<beforebreak>[^\.]\s[ضصثقفغعهخحجچشسیبلاتنمکگ\ظطزرذدپوًٌٍَُِّْA-Z]\.\s</beforebreak>
 <afterbreak></afterbreak>
 </rule>
 <rule break="no">
@@ -6359,7 +6367,7 @@
 </rule>
 <!--Не раздвајај у случају као на пр.: "Петар I дошао је ..."-->
 <rule break="no">
-<beforebreak>[\s ][IVX]+\s</beforebreak>
+<beforebreak>[\s ][IVX]+\s</beforebreak>
 <afterbreak>[^\p{Lu}]+</afterbreak>
 </rule>
 <!--Не раздвајај у случају као "од 13. до 14. века"-->

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: srx-languagetool
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors:
 - Aaron Madlon-Kay
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2021-02-13 00:00:00.000000000 Z
+date: 2021-04-15 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: srx
@@ -120,6 +120,7 @@ files:
 - ".gitignore"
 - ".rubocop.yml"
 - ".rubocop_todo.yml"
+- ".ruby-version"
 - ".solargraph.yml"
 - CHANGELOG.md
 - Gemfile
@@ -155,7 +156,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.1.4
+rubygems_version: 3.1.6
 signing_key: 
 specification_version: 4
 summary: SRX segmentation rules from LanguageTool