srx-languagetool 0.9.0 → 0.10.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 15d222238ca97d49bdcfd3ab55fed3ac9556cd1479ede47c6bcc126eca67db91
-  data.tar.gz: c67a1a5931a94b815d45147c091083ee776006e21003c95872b534158bdbd9fa
+  metadata.gz: 638b3fcf903059ad698b170d019a2bf52000e0cf20ee69fbf931e83ad0facb90
+  data.tar.gz: cee145db2f6fdabed2364cb03b66867f25b70ec0534a826dedea2c44943a6a58
 SHA512:
-  metadata.gz: bd4ad4a1740d7c1f880bcfcb32acfd6b636c1491716d59d256d60ae66b0611783f134e116c1c8c16e1acf147aced8484865492fc7293fdd8f0330bf22d622a59
-  data.tar.gz: 6c28a74647adcd1a87e26d6c4cba540fcd38f0f71c817a4ce2a59cbcbbc6e21932aee3b56e2ad316fe11c838f48a70547c518a169495f14409acac8ada194d9b
+  metadata.gz: 6593a82667e0ca2f9695ec4c51e68d4df286d2edadd001f46e2399a8111b2af82f4c7c28588c4040511130cf6853354e096c827ba5f5f7d118fac1a476507f8e
+  data.tar.gz: 775883a45f56a328c31ae2d4d9aa12f72694c62611e4c2b9d93852282715aa537bd624f29d2ffc7408c533411baa5f371058fe06242ffcf9a7c02909db29f8b9

data/.github/workflows/main.yml

@@ -13,7 +13,7 @@ jobs:
         ruby-version: 2.7.7
     - name: Install
       run: |
-        gem install bundler -v 2.4.1
+        gem install bundler -v 2.4.10
         bundle install
     - name: Type check
       run: bundle exec solargraph typecheck --level typed

data/CHANGELOG.md

@@ -1,5 +1,9 @@
 ## [Unreleased]
 
+## [0.10.0] - 2023-03-29
+
+- Update rules to LanguageTool 6.1
+
 ## [0.9.0] - 2022-12-30
 
 - Update rules to LanguageTool 6.0

data/Gemfile

@@ -4,3 +4,10 @@ source 'https://rubygems.org'
 
 # Specify your gem's dependencies in srx-languagetool.gemspec
 gemspec
+
+gem 'byebug'
+gem 'minitest'
+gem 'rake'
+gem 'rspec-expectations'
+gem 'rubocop'
+gem 'solargraph'

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    srx-languagetool (0.9.0)
+    srx-languagetool (0.10.0)
       srx (< 1.0)
 
 GEM
@@ -19,36 +19,36 @@ GEM
       rexml
     kramdown-parser-gfm (1.1.0)
       kramdown (~> 2.0)
-    minitest (5.16.3)
-    nokogiri (1.13.10-x86_64-darwin)
+    minitest (5.18.0)
+    nokogiri (1.14.2-x86_64-darwin)
       racc (~> 1.4)
     parallel (1.22.1)
-    parser (3.1.3.0)
+    parser (3.2.1.1)
       ast (~> 2.4.1)
     racc (1.6.2)
     rainbow (3.1.1)
     rake (13.0.6)
-    regexp_parser (2.6.1)
+    regexp_parser (2.7.0)
     reverse_markdown (2.1.1)
       nokogiri
     rexml (3.2.5)
-    rspec-expectations (3.12.1)
+    rspec-expectations (3.12.2)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.12.0)
     rspec-support (3.12.0)
-    rubocop (1.41.1)
+    rubocop (1.48.1)
       json (~> 2.3)
       parallel (~> 1.10)
-      parser (>= 3.1.2.1)
+      parser (>= 3.2.0.0)
       rainbow (>= 2.2.2, < 4.0)
       regexp_parser (>= 1.8, < 3.0)
       rexml (>= 3.2.5, < 4.0)
-      rubocop-ast (>= 1.23.0, < 2.0)
+      rubocop-ast (>= 1.26.0, < 2.0)
       ruby-progressbar (~> 1.7)
-      unicode-display_width (>= 1.4.0, < 3.0)
-    rubocop-ast (1.24.1)
-      parser (>= 3.1.1.0)
-    ruby-progressbar (1.11.0)
+      unicode-display_width (>= 2.4.0, < 3.0)
+    rubocop-ast (1.28.0)
+      parser (>= 3.2.1.0)
+    ruby-progressbar (1.13.0)
     solargraph (0.48.0)
       backport (~> 1.2)
       benchmark
@@ -67,8 +67,8 @@ GEM
     srx (0.6.0)
       nokogiri (~> 1.11)
     thor (1.2.1)
-    tilt (2.0.11)
-    unicode-display_width (2.3.0)
+    tilt (2.1.0)
+    unicode-display_width (2.4.2)
     webrick (1.7.0)
     yard (0.9.28)
       webrick (~> 1.7.0)
@@ -88,4 +88,4 @@ DEPENDENCIES
   srx-languagetool!
 
 BUNDLED WITH
-   2.4.1
+   2.4.10

data/lib/srx/languagetool/version.rb

@@ -2,6 +2,6 @@
 
 module Srx
   module Languagetool
-    VERSION = '0.9.0'
+    VERSION = '0.10.0'
   end
 end

data/lib/srx/segment.srx

@@ -4,10 +4,8 @@
 <formathandle type="start" include="no"></formathandle>
 <formathandle type="end" include="yes"></formathandle>
 <formathandle type="isolated" include="no"></formathandle>
-<okpsrx:options oneSegmentIncludesAll="no" trimLeadingWhitespaces="no" trimTrailingWhitespaces="no" useJavaRegex="yes"></okpsrx:options>
-<okpsrx:sample language="pl" useMappedRules="yes">Als een hoogleraar met emeritaat ('pensioen') is, mag hij de functieaanduiding prof. blijven gebruiken, maar hij heeft tevens het recht gekregen om het bijvoeglijk naamwoord emeritus (Latijn voor 'uitgediend') aan zijn functietitel toe te voegen: em. prof. dr.
-Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg waar hij met behulp van een oudere, rijke vriendin (met wie hij later trouwde) zijn eerste vliegtuigen bouwde, het waren passagierstoestellen.
-250 p. n.e.</okpsrx:sample>
+<okpsrx:options oneSegmentIncludesAll="no" trimLeadingWhitespaces="no" trimTrailingWhitespaces="no" useJavaRegex="yes" useIcu4JBreakRules="no" treatIsolatedCodesAsWhitespace="no"></okpsrx:options>
+<okpsrx:sample language="nl" useMappedRules="yes">De organisatie Doe! is een rare.</okpsrx:sample>
 <okpsrx:rangeRule></okpsrx:rangeRule>
 </header>
 <body>
@@ -1186,6 +1184,10 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <afterbreak>[^\p{P}]</afterbreak>
 </rule>
 <rule break="no">
+<beforebreak>[?!.…]["”]\)[\s\u00A0]</beforebreak>
+<afterbreak>[a-z].*</afterbreak>
+</rule>
+<rule break="no">
 <beforebreak>(…|\.\.\.)[\s\u00A0]?\?\)[\s\u00A0]</beforebreak>
 <afterbreak>[^\p{P}]</afterbreak>
 </rule>
@@ -1567,6 +1569,10 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <beforebreak>\b(enz|etc|zat|ambt|al|ver|art|wed|lab|bv|Bros)\.\s</beforebreak>
 <afterbreak>\p{Ll}</afterbreak>
 </rule>
+<rule break="yes">
+<beforebreak>\s(la|do|del)\sMar\.\s</beforebreak>
+<afterbreak></afterbreak>
+</rule>
 <rule break="no">
 <beforebreak>\b(Ge?n|Ex|Le?v|Nu?m|D(eu)?t|Jo?z|Ri|R[ei]cht|Sa?m|Ko?n|Kr[on]{0,2}|Neh?|Est?|Jb|Ps|Spr?|Pr[ed]{0,2}|H(oog)?l|Je?s|Je?r|Kl(aagl)?|Ez(ech)?|Da?n|Ho?s|Jl|Am|Ob|Mc|Mi[ch]{0,2}|Nah?|Hk|Hab|Zf|[SZ]ef|Ha?g|Zc|Zach|Ma?l|Ma?t|Mk|Mar|Lk|Jh|H(an)?d|Ro?m|Kor|Ga?l|Ef|Fp|Fil|Ko|[CK]ol|Th|Th?e[s]{1,2}|Tm|Ti?t|Fm|Fil(em)?|Hb|Hebr?|Jk|Ja[ck]|Pe?tr?|Joh|Jud|Op(enb)?|Wijsh|Tob|Sir|Bar|Makk)\.\s</beforebreak>
 <afterbreak></afterbreak>
@@ -1576,7 +1582,11 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <afterbreak>\p{Lu}</afterbreak>
 </rule>
 <rule break="no">
-<beforebreak>\b([Mm]ej|[Mm]evr|[Mm]rs|[Mm]s|[Mm]gr|[Mm]w|Ndl|Ned|Nl|No|Prof|[Ss]ecr|Chr|Jac|[Ww]ed)\.\s</beforebreak>
+<beforebreak>\b(Stb)\.\s</beforebreak>
+<afterbreak>[0-9]</afterbreak>
+</rule>
+<rule break="no">
+<beforebreak>\b([Mm]ej|[Mm]evr|[Mm]rs|[Mm]s|[Mm]gr|[Mm]w|Ndl|Ned|Nl|No|Prof|[Ss]ecr|Chr|Jac|[Ww]ed|Zr)\.\s</beforebreak>
 <afterbreak></afterbreak>
 </rule>
 <rule break="no">
@@ -1616,7 +1626,7 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <afterbreak></afterbreak>
 </rule>
 <rule break="no">
-<beforebreak>\b(geb|[Gg]em|get|gld|id|[Ii]ncl|ind|inf|ing|intern|inz|ir|jhr|jkvr)\.\s</beforebreak>
+<beforebreak>\b(geb|[Gg]em|get|gld|id|[Ii]ncl|ind|inf|ing|intern|[Ss]ec|inz|ir|jhr|jkvr)\.\s</beforebreak>
 <afterbreak></afterbreak>
 </rule>
 <rule break="no">
@@ -1729,6 +1739,14 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <afterbreak>[a-z]</afterbreak>
 </rule>
 <rule break="yes">
+<beforebreak>\s'[2-9][.]\s</beforebreak>
+<afterbreak></afterbreak>
+</rule>
+<rule break="no">
+<beforebreak>\s[A-Z].+!\s</beforebreak>
+<afterbreak>[a-z]</afterbreak>
+</rule>
+<rule break="yes">
 <beforebreak>[.!?…][’'"\u00BB\u2019\u201D\u203A\u00AB\p{Pe}\u0002¹²³]*\s</beforebreak>
 <afterbreak></afterbreak>
 </rule>
@@ -4972,7 +4990,7 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 </rule>
 <!-- don't split at cases like "Friedrich II. wird auch..." -->
 <rule break="no">
-<beforebreak>[\u00A0\s ][IVX]+\.[\u00A0\s]{1,2}</beforebreak>
+<beforebreak>[\u00A0\s ][IVX]+\.[\u00A0\s]{1,2}</beforebreak>
 <afterbreak>[^\p{Lu}]+</afterbreak>
 </rule>
 <!-- don't split at cases like "im 13. oder 14. Jahrhundert" -->
@@ -5012,7 +5030,7 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 </rule>
 <!-- German abbreviations -->
 <rule break="no">
-<beforebreak>\b(betr|Geb|Stk|ggü|Mag|mtl|[Pp]arl|versch|[Ss]tellv|d|Übers|usw|[Bb]zw|Ab[hkst]|[Aa]bzü?gl|[Ll]tda|[Ee]inschl|[Vv]mtl|Ev|bezgl|Abzw|[Vv]sl|ahd|Akk|aktual|allg|alltagsspr|altdt|alttest|amerikan|Anh|Ank|Anm|Art|[Aa]utom|Auftragsnr|Az|Bat|bayr|Bde?|bearb|Bed|Bem|bes|bez|Bez|Bhf|bspw|btto|bw|Dtl|Dez)\.[\u00A0\s]{1,2}</beforebreak>
+<beforebreak>\b(betr|Geb|Stk|ggü|Mag|mtl|[Pp]arl|Bsp|versch|[Ss]tellv|d|Übers|usw|[Bb]zw|Ab[hkst]|[Aa]bzü?gl|[Ll]tda|[Ee]inschl|[Vv]mtl|Ev|bezgl|Abzw|[Vv]sl|ahd|Akk|aktual|[Öö]ffentl|prof|allg|alltagsspr|altdt|alttest|amerikan|Anh|Ank|Anm|Art|[Aa]utom|Auftragsnr|Az|Bat|bayr|Bde?|bearb|Bed|Bem|bes|bez|Bez|Bhf|bspw|btto|bw|Dtl|Dez)\.[\u00A0\s]{1,2}</beforebreak>
 <afterbreak></afterbreak>
 </rule>
 <rule break="no">
@@ -5052,7 +5070,7 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <afterbreak>[\-–][\u00A0\s]\d+</afterbreak>
 </rule>
 <rule break="no">
-<beforebreak>\b(Tel|teilw|Temp|trans|Tsd|übertr|übl|ff|überarb|ugs|univ|unveränd|urspr|USt|UST|USt\-IdNr|[Aa][bn]schl|sw|kl|[Gg]r|vgl|vll|Vll|vlt|Vlt|vllt|Vllt|Vgl|Vol|vollst|vorm|Vp|Vs|vs|wesentl|[Rr]echts?staatl|[Ss]taatl|wg|Whg|Hd|Ztr|zus|Zus|zzt?|zzgl|zB|zb|Zz|Zt|zw|Min|Bzgl|bzgl|bezügl|Frhr|ggfs|insb|autom|Mw[sS]t)\.[\u00A0\s]{1,2}</beforebreak>
+<beforebreak>\b(Tel|teilw|Temp|trans|Tsd|übertr|übl|ff|überarb|ugs|univ|unveränd|urspr|USt|UST|USt\-IdNr|[Aa][bn]schl|sw|kl|[Gg]r|vgl|vll|Vll|vlt|Vlt|vllt|Vllt|Vgl|Vol|vollst|vorm|Vp|Vs|vs|wesentl|voraussichtl|[Rr]echts?staatl|[Ss]taatl|wg|Whg|Hd|Ztr|zus|Zus|zzt?|zzgl|zB|zb|Zz|Zt|zw|Min|Bzgl|bzgl|bezügl|Frhr|ggfs|insb|autom|Mw[sS]t)\.[\u00A0\s]{1,2}</beforebreak>
 <afterbreak></afterbreak>
 </rule>
 <!-- Break rules -->
@@ -5644,6 +5662,10 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <afterbreak></afterbreak>
 </rule>
 <rule break="no">
+<beforebreak>\b(кін)\.[\h\v]*</beforebreak>
+<afterbreak>[а-яіїєґ0-9IXV]|[ІХ]+\b</afterbreak>
+</rule>
+<rule break="no">
 <beforebreak>\b[сС]т\.[\h\v]</beforebreak>
 <afterbreak>[\h]*(?!([АВУОІЄ]|На|Але|Так?)[\h\v])</afterbreak>
 </rule>
@@ -5713,12 +5735,12 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <!-- але розбиваємо «всього 20 м. Почалося» -->
 <rule break="no">
 <beforebreak>(?&lt;!\d[\h\v]*)\bм\.[\h\v]*</beforebreak>
-<afterbreak>[А-ЯІЇЄҐ][а-яіїєґ]</afterbreak>
+<afterbreak>[А-ЯІЇЄҐ][а-яіїєґ']</afterbreak>
 </rule>
 <!-- село/сторінка/місто, але щоб не збігалося з секундами/метрами -->
 <rule break="no">
 <beforebreak>([\h\v][«(][см]|[^0-9/. ][\h\v]+[см])\.[\h\v]+</beforebreak>
-<afterbreak>[А-ЯІЇЄҐ]</afterbreak>
+<afterbreak>[А-ЯІЇЄҐ][а-яіїєґ']</afterbreak>
 </rule>
 <!-- (реж. Емманюель -->
 <rule break="no">
@@ -5735,6 +5757,11 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 <beforebreak></beforebreak>
 <afterbreak>[\h\v]*[‐-―-][\h\v]*([Рр]ед|[Аа]вт)[\h\v]*\.[\)\]]</afterbreak>
 </rule>
+<!-- Цензор.НЕТ -->
+<rule break="no">
+<beforebreak>[а-яіїєґ]\.</beforebreak>
+<afterbreak>НЕТ|Інфо|Info|City|Life|UA|Ру</afterbreak>
+</rule>
 <!-- але лишаємо ініціали: С.\u202F Шелухин -->
 <rule break="yes">
 <beforebreak>(?&lt;!\h[А-ЯІЇЄҐ])[.!?…]{1,3}\u202F[\h\v]+</beforebreak>
@@ -6520,7 +6547,7 @@ Tussen de twee wereldoorlogen vestigde prof. ir. Messerschmitt zich in Augsburg
 </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"-->

data/srx-languagetool.gemspec

@@ -28,11 +28,4 @@ Gem::Specification.new do |spec|
   spec.require_paths = ['lib']
 
   spec.add_dependency 'srx', '<1.0'
-
-  spec.add_development_dependency 'byebug'
-  spec.add_development_dependency 'minitest'
-  spec.add_development_dependency 'rake'
-  spec.add_development_dependency 'rspec-expectations'
-  spec.add_development_dependency 'rubocop'
-  spec.add_development_dependency 'solargraph'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: srx-languagetool
 version: !ruby/object:Gem::Version
-  version: 0.9.0
+  version: 0.10.0
 platform: ruby
 authors:
 - Aaron Madlon-Kay
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2022-12-30 00:00:00.000000000 Z
+date: 2023-03-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: srx
@@ -24,90 +24,6 @@ dependencies:
     - - "<"
       - !ruby/object:Gem::Version
         version: '1.0'
-- !ruby/object:Gem::Dependency
-  name: byebug
-  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'
-- !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: rspec-expectations
-  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: rubocop
-  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: solargraph
-  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: 
 email:
 - aaron@madlon-kay.com