suma 0.1.17 → 0.1.19

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 5ecace379e8cbe19d7f3d6144ce5862be6529c02b418352c79b382f0be184e67
-  data.tar.gz: c1ce481cd14a29ce112782a088c8b11722c7d091e17df3ca771339c0446b1396
+  metadata.gz: 6c1d7370c2e26551a0f6c13b354b1cf7c78180bc941a194421948918f35e69bd
+  data.tar.gz: 80aa2f22309620d3516c2aa65943c5287a9f09c40c797d24f27d34eea3821997
 SHA512:
-  metadata.gz: 4c84af1274e4b93530d213648de1a32cd580203ca8da2123f2071a6457c007b87ed343fbe8326ac652cd706c650603c959aea0ebb54d2f5109a6e92098e5ccd7
-  data.tar.gz: 4b1ee08b731c160b772a89c9dd5af33c7720fcc388bf8b8e9cf06554a4c07c8f052cf78f953f4841a48543ef255626e666b9f4c6bd0a4e225a956684a82954bc
+  metadata.gz: 38a4ea2e865b003787c466a6c49e86b5acee535a2dd218f0f384988392b0c8b8a67f2ef2e28c068a41c1ba97f72417f8b0f11620c41122e86f0c56c78f143caf
+  data.tar.gz: b61f64d0a8cec12a54ef82ed1ff6c1dab8b45200e155588d78bbb80d80f3468236d48e46c21af83beb911c4983afdadc91e8c35f5bbd92dcad2ad2e70d005299

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 02:29:17 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: 70
 # This cop supports safe autocorrection (--autocorrect).
 # Configuration parameters: Max, AllowHeredoc, AllowURI, AllowQualifiedName, URISchemes, IgnoreCopDirectives, AllowedPatterns, SplitStrings.
 # URISchemes: http, https
@@ -44,6 +28,7 @@ Layout/LineLength:
   Exclude:
     - 'lib/suma/cli.rb'
     - 'lib/suma/cli/build.rb'
+    - 'lib/suma/cli/extract_terms.rb'
     - 'lib/suma/cli/validate.rb'
     - 'lib/suma/cli/validate_ascii.rb'
     - 'lib/suma/cli/validate_links.rb'
@@ -57,14 +42,7 @@ 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
+# Offense count: 2
 # Configuration parameters: IgnoreLiteralBranches, IgnoreConstantBranches, IgnoreDuplicateElseBranch.
 Lint/DuplicateBranch:
   Exclude:
@@ -76,20 +54,7 @@ Lint/DuplicateMethods:
     - '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: 25
 # Configuration parameters: AllowedMethods, AllowedPatterns, CountRepeatedAttributes, Max.
 Metrics/AbcSize:
   Exclude:
@@ -107,7 +72,7 @@ Metrics/AbcSize:
 Metrics/BlockLength:
   Max: 64
 
-# Offense count: 9
+# Offense count: 12
 # Configuration parameters: AllowedMethods, AllowedPatterns, Max.
 Metrics/CyclomaticComplexity:
   Exclude:
@@ -117,7 +82,7 @@ Metrics/CyclomaticComplexity:
     - 'lib/suma/glossarist_extensions.rb'
     - 'lib/suma/thor_ext.rb'
 
-# Offense count: 36
+# Offense count: 37
 # Configuration parameters: CountComments, CountAsOne, AllowedMethods, AllowedPatterns.
 Metrics/MethodLength:
   Max: 107
@@ -127,7 +92,7 @@ Metrics/MethodLength:
 Metrics/ParameterLists:
   Max: 6
 
-# Offense count: 6
+# Offense count: 9
 # Configuration parameters: AllowedMethods, AllowedPatterns, Max.
 Metrics/PerceivedComplexity:
   Exclude:
@@ -136,6 +101,17 @@ Metrics/PerceivedComplexity:
     - 'lib/suma/cli/validate_links.rb'
     - 'lib/suma/glossarist_extensions.rb'
 
+# Offense count: 1
+# Configuration parameters: NamePrefix, ForbiddenPrefixes, AllowedMethods, MethodDefinitionMacros, UseSorbetSigs.
+# NamePrefix: is_, has_, have_, does_
+# ForbiddenPrefixes: is_, has_, have_, does_
+# AllowedMethods: is_a?
+# MethodDefinitionMacros: define_method, define_singleton_method
+Naming/PredicatePrefix:
+  Exclude:
+    - 'spec/**/*'
+    - 'lib/suma/cli/extract_terms.rb'
+
 # Offense count: 5
 # Configuration parameters: EnforcedStyle, CheckMethodNames, CheckSymbols, AllowedIdentifiers, AllowedPatterns.
 # SupportedStyles: snake_case, normalcase, non_integer
@@ -176,37 +152,17 @@ Style/EmptyElse:
 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
+# Offense count: 1
 # This cop supports unsafe autocorrection (--autocorrect-all).
-# Configuration parameters: ConvertCodeThatCanStartToReturnNil, AllowedMethods, MaxChainLength.
-# AllowedMethods: present?, blank?, presence, try, try!
-Style/SafeNavigation:
+# Configuration parameters: EnforcedStyle, AllowedMethods, AllowedPatterns.
+# SupportedStyles: predicate, comparison
+Style/NumericPredicate:
   Exclude:
-    - 'lib/suma/glossarist_extensions.rb'
+    - 'spec/**/*'
+    - 'lib/suma/cli/extract_terms.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

@@ -103,6 +103,10 @@ module Suma
               language_code: language_code,
               source_ref: source_ref,
             )
+            localized_concept_id = get_localized_concept_identifier(
+              schema, entity, language_code
+            )
+            localized_concept.uuid = localized_concept_id
 
             managed_data = Glossarist::ManagedConceptData.new.tap do |data|
               data.id = get_entity_identifier(schema, entity)
@@ -111,9 +115,7 @@ module Suma
               data.localizations[language_code] = localized_concept
               # uuid is automatically set from the serialization of the object
               data.localized_concepts = {
-                language_code => get_localized_concept_identifier(
-                  schema, entity, language_code
-                ),
+                language_code => localized_concept_id,
               }
             end
 
@@ -306,7 +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
@@ -321,6 +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.
@@ -397,11 +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", "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
@@ -411,24 +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", "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)
@@ -443,23 +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 = "0.1.17"
+  VERSION = "0.1.19"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: suma
 version: !ruby/object:Gem::Version
-  version: 0.1.17
+  version: 0.1.19
 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