suma 0.1.18 → 0.1.20

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 3fa571b528df159a9d83210a7f91f03cdb57cfa497dcaa312ad831392b4ab3ae
-  data.tar.gz: e536b9101471873b9ba3df2f422ec13221db3941e240c8731bcba1e79e361431
+  metadata.gz: 930231a5ca9c90a49c8e52b4a0881e2f93fd87508506872c02e693b9c0b8b25e
+  data.tar.gz: 13202f05c7673f950e8d4de87a7088a044db5e0b393cf768889c3c0483baf174
 SHA512:
-  metadata.gz: 0a759070a8972bba5f92705f9abc5bc7e23038d1c92873b35457709dfeed1e3c5a03acfd426feca37760fe849e5d173faf742e84231d97544b13f38e91cba192
-  data.tar.gz: cca4eeff40165f757f60311ee9db9a33802cfb1733c5de09aba15f63dce4aca1d1d38c74af1701b8ce689fac41ee994fd6b6de90f231c6a3ce1b76a026b1fce1
+  metadata.gz: 63c387978764d7bf16fb2b8ea129038dbbf771a360415ddf09254ab14dcf33cdae44166fbdb60439e52026925c7755e4b631985ac0b41d9b2cc43311e3e2374c
+  data.tar.gz: 8632ad3a285fa3a970805c7491812f07ed9501268f6cf3ce12fcb2be22b48d30a494e47bfbb40de1f4fe3f4c0e4822f5c2e3978637605f3afac7b2eb28edb8b9

data/.gitignore

@@ -9,7 +9,6 @@
 
 # rspec failure tracking
 .rspec_status
-.rubocop_todo.yml
 .rubocop-https---*-yml
 .ruby-version
 Gemfile.lock

data/.rubocop.yml

@@ -1,6 +1,6 @@
 inherit_from:
-  - https://raw.githubusercontent.com/riboseinc/oss-guides/main/ci/rubocop.yml
   - .rubocop_todo.yml
+  - https://raw.githubusercontent.com/riboseinc/oss-guides/main/ci/rubocop.yml
 
 plugins:
 - rubocop-performance
@@ -11,4 +11,9 @@ AllCops:
   TargetRubyVersion: 3.1
   NewCops: enable
   Exclude:
+    - 'lib/suma/cli/reformat.rb'
+    - 'lib/suma/cli/generate_schemas.rb'
+    - 'lib/suma/cli/validate_ascii.rb'
+    - 'lib/suma/cli/validate_links.rb'
+    - 'lib/suma/cli/extract_terms.rb'
     - 'vendor/**/*'

data/.rubocop_todo.yml

@@ -1,6 +1,6 @@
 # This configuration was generated by
 # `rubocop --auto-gen-config`
-# on 2025-07-17 08:10:53 UTC using RuboCop version 1.78.0.
+# on 2025-07-18 05:24:45 UTC using RuboCop version 1.78.0.
 # The point is for the user to remove these configuration records
 # one by one as the offenses are removed from the code base.
 # Note that changes in the inspected code, or installation of new
@@ -20,23 +20,7 @@ Gemspec/RequiredRubyVersion:
   Exclude:
     - 'suma.gemspec'
 
-# Offense count: 17
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EnforcedStyle, IndentationWidth.
-# SupportedStyles: with_first_argument, with_fixed_indentation
-Layout/ArgumentAlignment:
-  Exclude:
-    - 'lib/suma/cli/validate_ascii.rb'
-
-# Offense count: 1
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: EnforcedStyle.
-# SupportedStyles: normal, indented_internal_methods
-Layout/IndentationConsistency:
-  Exclude:
-    - 'lib/suma/glossarist_extensions.rb'
-
-# Offense count: 90
+# Offense count: 28
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: Max, AllowHeredoc, AllowURI, AllowQualifiedName, URISchemes, IgnoreCopDirectives, AllowedPatterns, SplitStrings.
 # URISchemes: http, https
@@ -45,9 +29,6 @@ Layout/LineLength:
     - 'lib/suma/cli.rb'
     - 'lib/suma/cli/build.rb'
     - 'lib/suma/cli/validate.rb'
-    - 'lib/suma/cli/validate_ascii.rb'
-    - 'lib/suma/cli/validate_links.rb'
-    - 'lib/suma/glossarist_extensions.rb'
     - 'lib/suma/processor.rb'
     - 'lib/suma/schema_attachment.rb'
     - 'lib/suma/schema_collection.rb'
@@ -57,92 +38,31 @@ Layout/LineLength:
     - 'spec/suma/cli/validate_ascii_spec.rb'
     - 'suma.gemspec'
 
-# Offense count: 22
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: AllowInHeredoc.
-Layout/TrailingWhitespace:
-  Exclude:
-    - 'lib/suma/cli/validate_ascii.rb'
-
 # Offense count: 1
-# Configuration parameters: IgnoreLiteralBranches, IgnoreConstantBranches, IgnoreDuplicateElseBranch.
-Lint/DuplicateBranch:
-  Exclude:
-    - 'lib/suma/cli/extract_terms.rb'
-
-# Offense count: 2
 Lint/DuplicateMethods:
   Exclude:
-    - 'lib/suma/cli/validate_ascii.rb'
     - 'lib/suma/express_schema.rb'
 
-# Offense count: 2
-# This cop supports unsafe autocorrection (--autocorrect-all).
-Lint/NonAtomicFileOperation:
-  Exclude:
-    - 'lib/suma/glossarist_extensions.rb'
-
-# Offense count: 1
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: AutoCorrect, IgnoreEmptyBlocks, AllowUnusedKeywordArguments.
-Lint/UnusedBlockArgument:
-  Exclude:
-    - 'lib/suma/glossarist_extensions.rb'
-
-# Offense count: 23
+# Offense count: 4
 # Configuration parameters: AllowedMethods, AllowedPatterns, CountRepeatedAttributes, Max.
 Metrics/AbcSize:
   Exclude:
-    - 'lib/suma/cli/extract_terms.rb'
-    - 'lib/suma/cli/validate_ascii.rb'
-    - 'lib/suma/cli/validate_links.rb'
-    - 'lib/suma/glossarist_extensions.rb'
     - 'lib/suma/schema_attachment.rb'
     - 'lib/suma/schema_document.rb'
     - 'lib/suma/thor_ext.rb'
 
-# Offense count: 3
-# Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns, inherit_mode.
-# AllowedMethods: refine
-Metrics/BlockLength:
-  Max: 64
-
-# Offense count: 9
+# Offense count: 1
 # Configuration parameters: AllowedMethods, AllowedPatterns, Max.
 Metrics/CyclomaticComplexity:
   Exclude:
-    - 'lib/suma/cli/extract_terms.rb'
-    - 'lib/suma/cli/validate_ascii.rb'
-    - 'lib/suma/cli/validate_links.rb'
-    - 'lib/suma/glossarist_extensions.rb'
     - 'lib/suma/thor_ext.rb'
 
-# Offense count: 36
-# Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns.
-Metrics/MethodLength:
-  Max: 107
-
 # Offense count: 4
-# Configuration parameters: CountKeywordArgs, MaxOptionalParameters.
-Metrics/ParameterLists:
-  Max: 6
-
-# Offense count: 6
-# Configuration parameters: AllowedMethods, AllowedPatterns, Max.
-Metrics/PerceivedComplexity:
-  Exclude:
-    - 'lib/suma/cli/extract_terms.rb'
-    - 'lib/suma/cli/validate_ascii.rb'
-    - 'lib/suma/cli/validate_links.rb'
-    - 'lib/suma/glossarist_extensions.rb'
-
-# Offense count: 5
 # Configuration parameters: EnforcedStyle, CheckMethodNames, CheckSymbols, AllowedIdentifiers, AllowedPatterns.
 # SupportedStyles: snake_case, normalcase, non_integer
 # AllowedIdentifiers: TLS1_1, TLS1_2, capture3, iso8601, rfc1123_date, rfc822, rfc2822, rfc3339, x86_64
 Naming/VariableNumber:
   Exclude:
-    - 'lib/suma/cli/validate_ascii.rb'
     - 'spec/suma/cli/validate_ascii_spec.rb'
 
 # Offense count: 1
@@ -159,54 +79,3 @@ RSpec/ExampleLength:
 # Offense count: 6
 RSpec/MultipleExpectations:
   Max: 12
-
-# Offense count: 2
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: AutoCorrect, EnforcedStyle, AllowComments.
-# SupportedStyles: empty, nil, both
-Style/EmptyElse:
-  Exclude:
-    - 'lib/suma/cli/validate_links.rb'
-    - 'lib/suma/glossarist_extensions.rb'
-
-# Offense count: 4
-# This cop supports safe autocorrection (--autocorrect).
-# Configuration parameters: MaxUnannotatedPlaceholdersAllowed, Mode, AllowedMethods, AllowedPatterns.
-# SupportedStyles: annotated, template, unannotated
-Style/FormatStringToken:
-  EnforcedStyle: unannotated
-
-# Offense count: 2
-# This cop supports unsafe autocorrection (--autocorrect-all).
-# Configuration parameters: AllowedReceivers.
-# AllowedReceivers: Thread.current
-Style/HashEachMethods:
-  Exclude:
-    - 'lib/suma/glossarist_extensions.rb'
-
-# Offense count: 2
-# This cop supports unsafe autocorrection (--autocorrect-all).
-Style/IdenticalConditionalBranches:
-  Exclude:
-    - 'lib/suma/glossarist_extensions.rb'
-
-# Offense count: 3
-# This cop supports unsafe autocorrection (--autocorrect-all).
-# Configuration parameters: ConvertCodeThatCanStartToReturnNil, AllowedMethods, MaxChainLength.
-# AllowedMethods: present?, blank?, presence, try, try!
-Style/SafeNavigation:
-  Exclude:
-    - 'lib/suma/glossarist_extensions.rb'
-
-# Offense count: 1
-# Configuration parameters: Max.
-Style/SafeNavigationChainLength:
-  Exclude:
-    - 'lib/suma/cli/extract_terms.rb'
-
-# Offense count: 4
-# This cop supports unsafe autocorrection (--autocorrect-all).
-# Configuration parameters: Mode.
-Style/StringConcatenation:
-  Exclude:
-    - 'lib/suma/cli/validate_ascii.rb'

data/lib/suma/cli/extract_terms.rb

@@ -308,7 +308,14 @@ module Suma
       # (in EXPRESS remark) becomes NOTE 1 in ISO 10303-2 of the entity.
       def only_keep_first_sentence(notes)
         notes.each do |note|
-          # Split by period and take the first sentence
+          # Skip truncation only for content that starts with a paragraph ending in ":"
+          # followed by a list (complete list structures that should be preserved)
+          if note&.content && should_preserve_complete_structure?(note.content)
+            # For complete list structures, keep the content as-is
+            next
+          end
+
+          # Split by period and take the first sentence for all other content
           # Avoid splitting by pattern like "abc.def"
           if note&.content
             new_content = note.content
@@ -323,6 +330,30 @@ module Suma
         end
       end
 
+      def should_preserve_complete_structure?(content)
+        return false if content.nil? || content.empty?
+
+        # Check if content starts with a single introductory sentence ending in ":"
+        # followed by a list. This indicates a complete list structure that should be preserved.
+        lines = content.split("\n")
+        first_paragraph = lines.first&.strip
+
+        # Look for pattern: Single sentence ending with ":" (introductory pattern)
+        if first_paragraph&.end_with?(":") && lines.length > 1
+          # Check if the first paragraph contains multiple sentences (periods before the colon)
+          # If it does, this is NOT an introductory paragraph - extract first sentence only
+          if first_paragraph.count(".").positive?
+            return false
+          end
+
+          # Check if there's a list after the colon
+          remaining_content = lines[1..].join("\n")
+          return starts_with_list?(remaining_content.strip)
+        end
+
+        false
+      end
+
       # https://github.com/metanorma/iso-10303/issues/621
       # 2. If this first sentence matches the 7-word magic sentence
       # (2-3 forms of that), it is discarded so there will not be a NOTE 1.
@@ -399,11 +430,112 @@ module Suma
         schema_id.end_with?("_bom")
       end
 
+      def contains_list?(content)
+        return false if content.nil? || content.empty?
+
+        # Check if content contains list markers
+        content.match?(/^\s*[\*\-\+]\s+/m) || content.match?(/^\s*\d+\.\s+/m)
+      end
+
+      def starts_with_list?(content)
+        return false if content.nil? || content.empty?
+
+        # Check if content starts with list markers
+        content.match?(/^\s*[\*\-\+]\s+/) || content.match?(/^\s*\d+\.\s+/)
+      end
+
+      def is_list_continuation?(content)
+        return false if content.nil? || content.empty?
+
+        # Check for AsciiDoc list continuation patterns
+        content.match?(/^\+\s*$/) ||
+          content.match?(/^--\s*$/) ||
+          content.match?(/^\s{2,}/) || # Indented content (continuation)
+          content.start_with?("which", "where", "that") # Logical continuation
+      end
+
+      def extract_complete_list(paragraphs, start_index)
+        return paragraphs[start_index] if start_index >= paragraphs.length
+
+        combined = paragraphs[start_index].dup
+        current_index = start_index + 1
+
+        # Check if the first paragraph already contains an opening continuation block
+        in_continuation_block = combined.include?("--") && !combined.match?(/--.*--/m)
+
+        # Continue collecting paragraphs while we're in a list context
+        while current_index < paragraphs.length
+          next_para = paragraphs[current_index]
+
+          # Check if we're entering or exiting a continuation block
+          if next_para.match?(/^--\s*$/) || next_para.end_with?("--")
+            in_continuation_block = !in_continuation_block
+            combined += "\n\n#{next_para}"
+            current_index += 1
+            next
+          end
+
+          # If we're in a continuation block, include all content until we hit the closing --
+          if in_continuation_block
+            combined += "\n\n#{next_para}"
+            current_index += 1
+            next
+          end
+
+          # Check if this is a list item or list continuation
+          if starts_with_list?(next_para) || is_list_continuation?(next_para)
+            combined += "\n\n#{next_para}"
+            current_index += 1
+
+            # Check if this paragraph contains an opening continuation block
+            if next_para.include?("--") && !next_para.match?(/--.*--/m)
+              in_continuation_block = true
+            end
+          else
+            # This paragraph is not part of the list structure
+            break
+          end
+        end
+
+        combined
+      end
+
+      def ends_list_structure?(current_para, next_para)
+        return true if next_para.nil?
+
+        # List ends if:
+        # 1. Current paragraph doesn't end with continuation markers
+        # 2. Next paragraph starts a new section (not list or continuation)
+        !current_para.match?(/\+\s*$/) &&
+          !starts_with_list?(next_para) &&
+          !is_list_continuation?(next_para)
+      end
+
+      def apply_first_sentence_logic(paragraph)
+        # Apply the original first-sentence extraction logic
+        # Split by period and take the first sentence
+        # Avoid splitting by pattern like "abc.def"
+        new_content = paragraph
+          .split(".\n").first.strip
+          .split(". ").first.strip
+
+        if new_content.end_with?(".")
+          new_content
+        else
+          "#{new_content}."
+        end
+      end
+
       # rubocop:disable Metrics/MethodLength
       def combine_paragraphs(full_paragraph, next_paragraph)
+        # Check if we're dealing with a list structure
+        if contains_list?(full_paragraph) || starts_with_list?(next_paragraph)
+          return combine_list_content(full_paragraph, next_paragraph)
+        end
+
+        # For regular paragraphs, apply the original first-sentence logic
         # If full_paragraph already contains a period, extract that.
         if m = full_paragraph.match(/\A(?<inner_first>[^\n]*?\.)\s/)
-          # puts "CONDITION 1"
           if m[:inner_first]
             return m[:inner_first]
           else
@@ -413,24 +545,26 @@ module Suma
 
         # If full_paragraph ends with a period, this is the last.
         if /\.\s*\Z/.match?(full_paragraph)
-          # puts "CONDITION 2"
           return full_paragraph
         end
 
-        # If next_paragraph is a list
-        if next_paragraph.start_with?("*")
-          # puts "CONDITION 3"
+        # If next_paragraph is a continuation of a paragraph
+        if next_paragraph&.start_with?("which", "where", "that")
           return "#{full_paragraph}\n\n#{next_paragraph}"
         end
 
-        # If next_paragraph is a continuation of a list
-        if next_paragraph.start_with?("which", "that")
-          # puts "CONDITION 4"
-          return "#{full_paragraph}\n\n#{next_paragraph}"
+        full_paragraph
+      end
+
+      def combine_list_content(full_paragraph, next_paragraph)
+        combined = full_paragraph.dup
+
+        # If we have a next paragraph, add it
+        unless next_paragraph.nil? || next_paragraph.empty?
+          combined += "\n\n#{next_paragraph}"
         end
 
-        # puts "CONDITION 5"
-        full_paragraph
+        combined
       end
 
       def trim_definition(definition)
@@ -445,23 +579,22 @@ module Suma
 
         return nil if definition_str.empty?
 
-        # Unless the first paragraph ends with "between" and is followed by a
-        # list, don't split
         paragraphs = definition_str.split("\n\n")
-
-        # puts paragraphs.inspect
-
         first_paragraph = paragraphs.first
 
-        combined = if paragraphs.length > 1
-                     paragraphs[1..].inject(first_paragraph) do |acc, p|
-                       combine_paragraphs(acc, p)
-                     end
-                   else
-                     combine_paragraphs(first_paragraph, "")
-                   end
-
-        # puts "combined--------- #{combined}"
+        # If we only have one paragraph, apply the original logic
+        if paragraphs.length == 1
+          combined = apply_first_sentence_logic(first_paragraph)
+        elsif first_paragraph.end_with?(":") && paragraphs.length > 1 && starts_with_list?(paragraphs[1])
+          # Case 1: First paragraph ends with ":" and leads into a list
+          # Extract the complete list structure (this is an introductory paragraph)
+          complete_list = extract_complete_list(paragraphs, 1)
+          combined = "#{first_paragraph}\n\n#{complete_list}"
+        else
+          # Case 2: For all other cases (including sentences followed by lists)
+          # Extract only the first sentence from the first paragraph
+          combined = apply_first_sentence_logic(first_paragraph)
+        end
 
         # Remove comments until end of line
         combined = "#{combined}\n"

data/lib/suma/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Suma
-  VERSION = "0.1.18"
+  VERSION = "0.1.20"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: suma
 version: !ruby/object:Gem::Version
-  version: 0.1.18
+  version: 0.1.20
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2025-07-17 00:00:00.000000000 Z
+date: 2025-07-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: expressir