simple_english 0.10.0-aarch64-linux → 0.11.0-aarch64-linux

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 0ebf2662378b400619f748fc25522c548682a96ebdd3a0b72fea03d3f52a6489
-  data.tar.gz: 5d6b4b81d24f7a4ff5c27393c5cbe5cd4c2339b4aa14c2ab021c6d25060a190b
+  metadata.gz: dd3dbf80f63e4a6586d0c1b9ab24cd843994cbac8e73927cf2a7242d030b9d31
+  data.tar.gz: dab8d658d20ab00c39d44a7610fc20cb020ef86c222ec3d1e05c281a7d569e6d
 SHA512:
-  metadata.gz: b4ab7d8ab283bded33454e049dafba66694c9b67bf014b5288379e6b00b86716326958e280b5172922f4fefc467b2c3021ad1b8ce49a4e9128e0c88a9589beb6
-  data.tar.gz: 1430067f026c0aa5d4e5ba4e5344d86d3ef7f4a9f06be94df8c4722eee44e41d163f33cc7245acbda33c6ee16aedde4d4d520b7bdd77dabb3fed499bce8eb43e
+  metadata.gz: 6671ad97fa056f5a1714ce074a146823a45b0b554ac0252dda25f7d1d037c411a3e7e572ea51303a6345a0d62b80065bc24c855527285ae07cc96b77fd7b1af0
+  data.tar.gz: 06e1f94e2e0c904861713691e0d62d3f21a8d975593244e7a6acd1e19ade0612b0e536436e614ab61aecd431f21e68c1437ba03403eee35d940c912af64d0b74

data/README.md

@@ -133,29 +133,59 @@ se mcp
 
 ### CI
 
-Gate the prose in the pull request that changes it. The plain run
-fails the build on findings, and the SARIF report puts them inline:
+Gate the pull request's prose in GitHub Actions. The lint fails the build on findings:
 
 ```yaml
 name: lint-docs
 on: [pull_request]
-permissions:
-  security-events: write
 jobs:
   lint:
     runs-on: ubuntu-latest
     steps:
       - uses: actions/checkout@v7
-      - uses: ruby/setup-ruby@v1
-      - run: gem install simple_english
-      - run: se --format sarif . > lint.sarif
-      - run: se .
+      - uses: TonyCTHsu/simple-english@v0.11.0
+        with:
+          files: |
+            docs/
+            README.md
+```
+
+To see the findings inline in the pull request, lint with
+`format: sarif` and upload the report. Add
+`permissions: {security-events: write}` to the workflow:
+
+```yaml
+      - uses: TonyCTHsu/simple-english@v0.11.0
+        with:
+          format: sarif
       - uses: github/codeql-action/upload-sarif@v3
         with:
-          sarif_file: lint.sarif
+          sarif_file: simple-english.sarif
         if: always()
 ```
 
+Any CI can run the lint. Install the gem, run `se lint .`, and fail the
+build on exit code 1.
+
+### Git hooks
+
+Install the gem once, then wire your hook runner. The command to run
+is `se lint`.
+
+#### Pre-commit framework
+
+```yaml
+# .pre-commit-config.yaml
+repos:
+  - repo: https://github.com/TonyCTHsu/simple-english
+    rev: v0.11.0
+    hooks:
+      - id: se-lint
+```
+
+For Husky with lint-staged, map every staged file to `se lint`. For
+Lefthook, run `se lint {staged_files}` in a `pre-commit` command.
+
 ## Config and suppressions
 
 ### Config file

data/docs/RULES.md

@@ -50,7 +50,7 @@ Use the simple past. Say when it happened.
 
 ### SE_ING_AFTER_COMMA: No -ing after a comma
 
-Start a new sentence instead of the -ing phrase.
+Write a new sentence, or put the condition first. A trailing -ing phrase hides who does what and when.
 
 - Wrong: The tool runs, making it easy.
 - Right: The tool runs. It is easy.

data/lib/simple_english/client/language_tool.rb

@@ -14,6 +14,7 @@ require "net/http"
 require "uri"
 
 require_relative "../lint/plain_text"
+require_relative "../lint/utf16"
 
 module SimpleEnglish
   module LanguageTool
@@ -71,25 +72,16 @@ module SimpleEnglish
       end
 
       def utf16_slice(text, offset, length)
-        first = utf16_index(text, offset)
-        last = utf16_index(text, offset + length)
+        first = UTF16.index(text, offset)
+        last = UTF16.index(text, offset + length)
         text[first...last]
       end
 
-      def utf16_index(text, offset)
-        units = 0
-        text.each_char.with_index do |char, index|
-          return index if units >= offset
-          units += (char.ord > 0xFFFF) ? 2 : 1
-        end
-        text.length
-      end
-
       def to_payload(payload)
         payload.is_a?(String) ? PlainText.new(payload) : payload
       end
 
-      private_class_method :to_payload, :with_context, :utf16_slice, :utf16_index
+      private_class_method :to_payload, :with_context, :utf16_slice
     end
   end
 end

data/lib/simple_english/lint/annotated_text.rb

@@ -9,6 +9,7 @@
 require_relative "extractor"
 require_relative "segment"
 require_relative "result"
+require_relative "utf16"
 
 module SimpleEnglish
   module AnnotatedText
@@ -21,10 +22,12 @@ module SimpleEnglish
     def build(source, spans)
       segments = []
       cursor = 0 # byte offset into source
+      stream_start = 0 # character offset into the annotation stream
       emit = lambda do |text:, content:, file_char:, interpret_as: nil|
         next if content.empty?
         segments << Segment.new(text: text, content: content, file_char: file_char,
-          stream_start: segments.sum { |s| s.content.length }, interpret_as: interpret_as)
+          stream_start: stream_start, interpret_as: interpret_as)
+        stream_start += content.length
       end
       spans = spans.sort_by(&:start_byte)
       spans.each do |span|
@@ -66,14 +69,14 @@ module SimpleEnglish
 
     # -> [line, column], both 1-based. Column in UTF-16 code units.
     def locate(result, utf16_offset)
-      char = utf16_to_char(result.stream, utf16_offset)
+      char = UTF16.index(result.stream, utf16_offset)
       segment = result.segments.find do |s|
         char >= s.stream_start && char < s.stream_start + s.content.length
       end || result.segments.last
       file_char = segment.file_char + (char - segment.stream_start)
       starts = line_starts(result.source)
       line = starts.rindex { |start| start <= file_char } + 1
-      [line, utf16_column(result.source, starts[line - 1], file_char)]
+      [line, UTF16.column(result.source, starts[line - 1], file_char)]
     end
 
     def split(comment)
@@ -96,22 +99,6 @@ module SimpleEnglish
       starts
     end
 
-    def utf16_to_char(stream, offset)
-      units = 0
-      stream.each_char.with_index do |char, index|
-        return index if units >= offset
-        units += (char.ord > 0xFFFF) ? 2 : 1
-      end
-      stream.length
-    end
-
-    def utf16_column(source, from, to)
-      units = 1
-      source[from...to].each_char { |c| units += (c.ord > 0xFFFF) ? 2 : 1 }
-      units
-    end
-
-    private_class_method :split, :char_index, :line_starts,
-      :utf16_to_char, :utf16_column
+    private_class_method :split, :char_index, :line_starts
   end
 end

data/lib/simple_english/lint/markdown.rb

@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 
 require_relative "paragraph"
+require_relative "utf16"
 
 # Markdown stripping and paragraph splitting. Block structure comes
 # from the tree-sitter Markdown grammar shipped with the pack the
@@ -42,16 +43,12 @@ module SimpleEnglish
       # Width-preserving replacements keep LanguageTool positions aligned
       # with the source.
       line
-        .gsub(/`[^`]*`/) { |code| "X" * utf16_length(code) }
+        .gsub(/`[^`]*`/) { |code| "X" * UTF16.length(code) }
         .sub(/\A\#{1,6} /) { |marker| " " * marker.length }
     end
 
     def spaces_for(text)
-      " " * utf16_length(text)
-    end
-
-    def utf16_length(text)
-      text.each_char.sum { |char| (char.ord > 0xFFFF) ? 2 : 1 }
+      " " * UTF16.length(text)
     end
 
     # A vertical list is not one paragraph: each list item is its own.
@@ -115,7 +112,7 @@ module SimpleEnglish
       TreeSitterLanguagePack.get_parser("markdown").parse(text).root_node
     end
 
-    private_class_method :strip_line, :spaces_for, :utf16_length, :node_rows,
+    private_class_method :strip_line, :spaces_for, :node_rows,
       :blank_rows, :each_node, :inside_list_item?, :parse
   end
 end

data/lib/simple_english/lint/utf16.rb

@@ -0,0 +1,33 @@
+# frozen_string_literal: true
+
+# UTF-16 code-unit arithmetic. LanguageTool is a Java service, and
+# Java strings count UTF-16 code units: an astral-plane character
+# counts as two. Every offset crossing the LT wire goes through
+# here.
+
+module SimpleEnglish
+  module UTF16
+    module_function
+
+    # The character index in text where the offsetth code unit
+    # falls.
+    def index(text, offset)
+      units = 0
+      text.each_char.with_index do |char, index|
+        return index if units >= offset
+        units += (char.ord > 0xFFFF) ? 2 : 1
+      end
+      text.length
+    end
+
+    # The code-unit length of text.
+    def length(text)
+      text.each_char.sum { |char| (char.ord > 0xFFFF) ? 2 : 1 }
+    end
+
+    # The 1-based code-unit column of the character at index to.
+    def column(source, from, to)
+      length(source[from...to]) + 1
+    end
+  end
+end

data/lib/simple_english/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module SimpleEnglish
-  VERSION = "0.10.0"
+  VERSION = "0.11.0"
 end

data/lib/simple_english.rb

@@ -8,6 +8,7 @@
 require_relative "simple_english/version"
 require_relative "simple_english/lint/finding"
 require_relative "simple_english/lint/paragraph"
+require_relative "simple_english/lint/utf16"
 require_relative "simple_english/lint/markdown"
 require_relative "simple_english/lint/counts"
 require_relative "simple_english/setup/languagetool"

data/libexec/simple_english/languagetool-server

@@ -1,4 +1,4 @@
 [diffend] Oversized file quarantined before diffing.
 name: data/libexec/simple_english/languagetool-server
 size: 127207544 bytes
-sha256: 6a8731f0c1994d32df7a4e1f48578337fa3eb7469c1d16a1c5d07dd040771269
+sha256: 90096407bc281ca375f61588bba5cb60a0906aac6b0ff37e19822a635e212620

data/libexec/simple_english/languagetool-server.sbom.json

@@ -4,7 +4,7 @@
   "serialNumber" : "urn:uuid:c99881b6-e8ff-383d-92df-6845602c70c5",
   "version" : 1,
   "metadata" : {
-    "timestamp" : "2026-10-15T20:56:12Z",
+    "timestamp" : "2026-10-17T21:16:46Z",
     "lifecycles" : [
       {
         "phase" : "build"

data/rules/simple-english.xml

@@ -2,13 +2,21 @@
 <rules lang="en">
   <category id="SIMPLE_ENGLISH" name="Simple English">
 
+    <!-- LanguageTool splits "you're" into "you" + "'re", so the
+         regex matches the apostrophe-initial tokens: "'re", "'ve",
+         "'ll", "'d", "'m", and the "n't" token of "don't". The
+         apostrophe-s form stays out: it is both the contraction
+         "it is" and the possessive, and the tagger does not tell
+         them apart reliably. -->
     <rule id="SE_NO_CONTRACTIONS" name="No contractions">
       <pattern>
-        <token regexp="yes">[a-z]+['’](s|t|re|ve|ll|d|m)</token>
+        <token regexp="yes">[a-z]*['’](t|re|ve|ll|d|m)</token>
       </pattern>
       <message>Write the words in full. No contractions.</message>
       <example type="incorrect">You <marker>don't</marker> need this file.</example>
+      <example type="incorrect">If <marker>you're</marker> connecting, use TCP.</example>
       <example type="correct">You do not need this file.</example>
+      <example type="correct">sqlpipe's IP is in the config.</example>
     </rule>
 
     <rule id="SE_NO_SEMICOLON" name="No semicolons">
@@ -45,6 +53,11 @@
       <example type="correct">The worker wrote the log.</example>
     </rule>
 
+    <!-- A passive agent changes the finding, not the rule match:
+         "has been read by X" also matches SE_ACTIVE_VOICE, and
+         LanguageTool reports only the first of two overlapping
+         matches, so this rule stays silent there. The active-voice
+         finding names the fix. -->
     <rule id="SE_PRESENT_PERFECT" name="Simple tenses only">
       <pattern>
         <token regexp="yes">have|has|had</token>
@@ -55,11 +68,22 @@
       <example type="correct">The job completed.</example>
     </rule>
 
+    <!-- Spirit: a participle clause after a comma leaves the subject
+         and the time relation implicit, so the reader must guess. Two
+         escapes exist: write a new sentence, or, when the phrase
+         states a condition, put the condition first
+         (SE_CONDITION_FIRST). No word-list carve-outs: this tool ships
+         no approved dictionary, so every comma-plus-ing participle
+         fires, including the preposition-like ones. -->
     <rule id="SE_ING_AFTER_COMMA" name="No -ing after a comma">
       <pattern><token>,</token><token postag="VBG"><exception scope="next" regexp="yes">[,)]|and|or</exception></token></pattern>
-      <message>Start a new sentence instead of the -ing phrase.</message>
+      <message>Write a new sentence, or put the condition first. A trailing -ing phrase hides who does what and when.</message>
       <example type="incorrect">The tool runs<marker>, making</marker> it easy.</example>
+      <example type="incorrect">Open the valve<marker>, depending</marker> on the pressure.</example>
+      <example type="incorrect">Remove the cover<marker>, making</marker> sure not to damage the seal.</example>
       <example type="correct">The tool runs. It is easy.</example>
+      <example type="correct">If the pressure is more than 300 kPa, open the valve.</example>
+      <example type="correct">Remove the cover. Make sure that you do not damage the seal.</example>
       <example type="correct">The walkthrough (accounts, VPC, secrets, troubleshooting) is in the guide.</example>
       <example type="correct">The guide covers billing, troubleshooting, and support for each environment.</example>
       <example type="correct">The guide covers billing, troubleshooting and support for each environment.</example>

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: simple_english
 version: !ruby/object:Gem::Version
-  version: 0.10.0
+  version: 0.11.0
 platform: aarch64-linux
 authors:
 - TonyCTHsu
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2026-10-15 00:00:00.000000000 Z
+date: 2026-10-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: tree_sitter_language_pack
@@ -154,6 +154,7 @@ files:
 - lib/simple_english/lint/segment.rb
 - lib/simple_english/lint/span.rb
 - lib/simple_english/lint/suppressions.rb
+- lib/simple_english/lint/utf16.rb
 - lib/simple_english/model_context_protocol.rb
 - lib/simple_english/setup/config.rb
 - lib/simple_english/setup/fingerprint.rb
@@ -177,7 +178,7 @@ licenses:
 metadata:
   homepage_uri: https://github.com/TonyCTHsu/simple-english
   source_code_uri: https://github.com/TonyCTHsu/simple-english
-  changelog_uri: https://github.com/TonyCTHsu/simple-english/blob/v0.10.0/CHANGELOG.md
+  changelog_uri: https://github.com/TonyCTHsu/simple-english/blob/v0.11.0/CHANGELOG.md
 post_install_message: 
 rdoc_options: []
 require_paths: