expressir 2.3.7 → 2.4.1

data/lib/expressir/commands/file_violations.rb

@@ -0,0 +1,70 @@
+# frozen_string_literal: true
+
+module Expressir
+  module Commands
+    class FileViolations
+      attr_reader :path, :filename, :directory, :violations
+
+      def initialize(file_path)
+        @path = file_path
+        @filename = File.basename(file_path)
+        @directory = File.dirname(file_path)
+        @characters = {}  # Map of characters to NonAsciiCharacter objects
+        @violations = []  # List of violations (line, column, etc.)
+      end
+
+      def add_violation(line_number, column, match, char_details, line)
+        violation = {
+          line_number: line_number,
+          column: column,
+          match: match,
+          char_details: char_details,
+          line: line,
+        }
+
+        @violations << violation
+
+        # Register each character
+        char_details.each do |detail|
+          char = detail[:char]
+          unless @characters[char]
+            @characters[char] = NonAsciiCharacter.new(
+              char,
+              detail[:hex],
+              detail[:utf8],
+              detail[:is_math],
+              detail[:replacement],
+              detail[:replacement_type],
+            )
+          end
+
+          @characters[char].add_occurrence(line_number, column, line)
+        end
+      end
+
+      def violation_count
+        @violations.size
+      end
+
+      def unique_characters
+        @characters.values
+      end
+
+      def display_path
+        "#{File.basename(@directory)}/#{@filename}"
+      end
+
+      def full_path
+        File.expand_path(@path)
+      end
+
+      def to_h
+        {
+          file: display_path,
+          count: violation_count,
+          non_ascii_characters: unique_characters.map(&:to_h),
+        }
+      end
+    end
+  end
+end

data/lib/expressir/commands/format.rb

@@ -35,7 +35,7 @@ module Expressir
       def format_with_iso_profile(repository)
         repository.schemas.each do |schema|
           say "\n(* Expressir formatted schema: #{schema.id} *)\n"
-          say schema.to_s(no_remarks: true)
+          say schema.format(no_remarks: true)
         end
       end
     end

data/lib/expressir/commands/non_ascii_character.rb

@@ -0,0 +1,49 @@
+# frozen_string_literal: true
+
+module Expressir
+  module Commands
+    class NonAsciiCharacter
+      attr_reader :char, :hex, :utf8, :is_math, :replacement,
+                  :replacement_type, :occurrences
+
+      def initialize(char, hex, utf8, is_math, replacement, replacement_type)
+        @char = char
+        @hex = hex
+        @utf8 = utf8
+        @is_math = is_math
+        @replacement = replacement
+        @replacement_type = replacement_type
+        @occurrences = []
+      end
+
+      def add_occurrence(line_number, column, line)
+        @occurrences << {
+          line_number: line_number,
+          column: column,
+          line: line,
+        }
+      end
+
+      def replacement_text
+        @is_math ? "AsciiMath: #{@replacement}" : "ISO 10303-11: #{@replacement}"
+      end
+
+      def occurrence_count
+        @occurrences.size
+      end
+
+      def to_h
+        {
+          character: @char,
+          hex: @hex,
+          utf8: @utf8,
+          is_math: @is_math,
+          replacement_type: @replacement_type,
+          replacement: @replacement,
+          occurrence_count: occurrence_count,
+          occurrences: @occurrences,
+        }
+      end
+    end
+  end
+end

data/lib/expressir/commands/non_ascii_violation_collection.rb

@@ -0,0 +1,301 @@
+# frozen_string_literal: true
+
+require "paint"
+
+module Expressir
+  module Commands
+    class NonAsciiViolationCollection
+      # Pre-defined mapping of common Unicode math symbols to AsciiMath.
+      UNICODE_TO_ASCIIDOC_MAP = {
+        # Greek letters
+        "α" => "alpha", "β" => "beta", "γ" => "gamma", "Γ" => "Gamma",
+        "δ" => "delta", "Δ" => "Delta", "ε" => "epsilon", "ζ" => "zeta",
+        "η" => "eta", "θ" => "theta", "Θ" => "Theta", "ι" => "iota",
+        "κ" => "kappa", "λ" => "lambda", "Λ" => "Lambda", "μ" => "mu",
+        "ν" => "nu", "ξ" => "xi", "Ξ" => "Xi", "π" => "pi", "Π" => "Pi",
+        "ρ" => "rho", "σ" => "sigma", "Σ" => "Sigma", "τ" => "tau",
+        "υ" => "upsilon", "φ" => "phi", "Φ" => "Phi", "χ" => "chi",
+        "ψ" => "psi", "Ψ" => "Psi", "ω" => "omega", "Ω" => "Omega",
+        # Math operators
+        "×" => "xx", "÷" => "div", "±" => "pm", "∓" => "mp",
+        "∞" => "oo", "≤" => "le", "≥" => "ge", "≠" => "ne",
+        "≈" => "~~", "≅" => "cong", "≡" => "equiv", "∈" => "in",
+        "∉" => "notin", "⊂" => "subset", "⊃" => "supset", "∩" => "cap",
+        "∪" => "cup", "∧" => "and", "∨" => "or", "¬" => "neg",
+        "∀" => "forall", "∃" => "exists", "∄" => "nexists", "∇" => "grad",
+        "∂" => "del", "∑" => "sum", "∏" => "prod", "∫" => "int",
+        "∮" => "oint", "√" => "sqrt", "⊥" => "perp", "‖" => "norm",
+        "→" => "rarr", "←" => "larr", "↔" => "harr",
+        "⇒" => "rArr", "⇐" => "lArr", "⇔" => "hArr"
+      }.freeze
+
+      attr_reader :file_violations, :total_files
+
+      def initialize(check_remarks: false)
+        @file_violations = {} # Map of file paths to FileViolations objects
+        @total_files = 0
+        @unicode_to_asciimath = nil
+        @check_remarks = check_remarks
+      end
+
+      def process_file(file)
+        @total_files += 1
+
+        # Initialize the mapping once
+        @unicode_to_asciimath ||= UNICODE_TO_ASCIIDOC_MAP
+
+        file_violations = process_file_violations(file)
+        return if file_violations.violations.empty?
+
+        @file_violations[file] = file_violations
+      end
+
+      def files_with_violations
+        @file_violations.size
+      end
+
+      def total_violations
+        @file_violations.values.sum(&:violation_count)
+      end
+
+      def unique_character_count
+        # Get total unique characters across all files
+        all_chars = Set.new
+        @file_violations.each_value do |file_violation|
+          file_violation.unique_characters.each do |char|
+            all_chars.add(char.char)
+          end
+        end
+        all_chars.size
+      end
+
+      def total_occurrence_count
+        # Sum all occurrences of all characters across all files
+        @file_violations.values.sum do |file_violation|
+          file_violation.unique_characters.sum(&:occurrence_count)
+        end
+      end
+
+      def to_yaml_data
+        {
+          summary: {
+            total_files: @total_files,
+            files_with_violations: files_with_violations,
+            total_violations: total_violations,
+            total_unique_characters: unique_character_count,
+            total_occurrences: total_occurrence_count,
+          },
+          violations: @file_violations.transform_keys do |k|
+            File.expand_path(k)
+          end.transform_values(&:to_h),
+        }
+      end
+
+      def print_text_output
+        # Print each file's violations if any
+        unless @file_violations.empty?
+          @file_violations.each_value do |file_violation|
+            puts "\n#{Paint[file_violation.display_path, :cyan, :bold]}:"
+
+            file_violation.violations.each do |v|
+              puts "  #{Paint['Line',
+                              :blue]} #{Paint[v[:line_number],
+                                              :yellow]}, #{Paint['Column',
+                                                                 :blue]} #{Paint[v[:column],
+                                                                                 :yellow]}:"
+              puts "    #{v[:line]}"
+              puts "    #{' ' * v[:column]}#{Paint['^' * v[:match].length,
+                                                   :red]} #{Paint['Non-ASCII sequence',
+                                                                  :red]}"
+
+              v[:char_details].each do |cd|
+                character = file_violation.unique_characters.find do |c|
+                  c.char == cd[:char]
+                end
+                next unless character
+
+                puts "      #{Paint["\"#{cd[:char]}\"",
+                                    :yellow]} - Hex: #{Paint[cd[:hex],
+                                                             :magenta]}, UTF-8 bytes: #{Paint[cd[:utf8],
+                                                                                              :magenta]}"
+                puts "      #{Paint['Replacement:',
+                                    :green]} #{character.replacement_text}"
+              end
+              puts ""
+            end
+
+            puts "  #{Paint['Found',
+                            :green]} #{Paint[file_violation.violation_count,
+                                             :red]} #{Paint['non-ASCII sequence(s) in',
+                                                            :green]} #{Paint[file_violation.filename,
+                                                                             :cyan]}\n"
+          end
+        end
+
+        # Always print summary
+        validation_scope = @check_remarks ? "code and all remarks" : "code and tail remarks (embedded remarks excluded)"
+        puts "\n#{Paint['Summary:', :blue, :bold]}"
+        puts "  #{Paint['Validation scope:',
+                        :green]} #{Paint[validation_scope,
+                                         :cyan]}"
+        puts "  #{Paint['Scanned',
+                        :green]} #{Paint[@total_files,
+                                         :yellow]} #{Paint['EXPRESS file(s)',
+                                                           :green]}"
+        puts "  #{Paint['Found',
+                        :green]} #{Paint[total_violations,
+                                         :red]} #{Paint['non-ASCII sequence(s) in',
+                                                        :green]} #{Paint[files_with_violations,
+                                                                         :red]} #{Paint['file(s)',
+                                                                                        :green]}"
+      end
+
+      def print_table_output
+        return if @file_violations.empty?
+
+        # Build rows array
+        rows = []
+        total_occurrences = 0
+
+        @file_violations.each_value do |file_violation|
+          file_violation.unique_characters.each do |character|
+            occurrence_count = character.occurrence_count
+            total_occurrences += occurrence_count
+
+            rows << {
+              file: file_violation.display_path,
+              symbol: "\"#{character.char}\" (#{character.hex})",
+              replacement: character.replacement_text,
+              occurrences: occurrence_count,
+            }
+          end
+        end
+
+        # Add total row
+        rows << {
+          file: "TOTAL",
+          symbol: "#{unique_character_count} unique",
+          replacement: "",
+          occurrences: total_occurrences,
+        }
+
+        # Use TableTennis to render
+        options = {
+          title: "Non-ASCII Characters Summary",
+          columns: %i[file symbol replacement occurrences],
+          headers: {
+            file: "File",
+            symbol: "Symbol",
+            replacement: "Replacement",
+            occurrences: "Occurrences",
+          },
+          mark: ->(row) { row[:file] == "TOTAL" },
+        }
+
+        puts "\n#{TableTennis.new(rows, options)}\n"
+      end
+
+      private
+
+      # Build a masked copy of `source` where every byte inside an embedded
+      # remark region is replaced with an ASCII space. Byte positions and the
+      # string's encoding are preserved, so masked lines still align 1:1 with
+      # the original file for reporting.
+      def mask_embedded_remarks(source)
+        remarks = Expressir::Express::RemarkScanner.new(source).scan
+        embedded = remarks.select(&:embedded?)
+        return source if embedded.empty?
+
+        masked = source.b.dup
+        embedded.each do |r|
+          (r.position...r.end_position).each { |i| masked.setbyte(i, 0x20) }
+        end
+        masked.force_encoding(source.encoding)
+      end
+
+      def scan_lines_for_non_ascii(file_violations, lines)
+        lines.each_with_index do |line, line_idx|
+          line_number = line_idx + 1
+          next unless /[^\x00-\x7F]/.match?(line)
+
+          line.chomp.scan(/([^\x00-\x7F]+)/) do |match|
+            match = match[0]
+            column = line.index(match)
+
+            char_details = match.chars.filter_map do |c|
+              process_non_ascii_char(c)
+            end
+            next if char_details.empty?
+
+            file_violations.add_violation(line_number, column, match,
+                                          char_details, line.chomp)
+          end
+        end
+      end
+
+      def process_file_violations(file)
+        file_violations = FileViolations.new(file)
+        source = File.read(file, encoding: "UTF-8")
+
+        # With --check-remarks: scan every line including embedded remarks.
+        # Without: scan code + tail remarks only — mask out embedded remark
+        # regions so non-ASCII inside `(* ... *)` documentation blocks is
+        # not flagged. Tail remarks (`--`) are part of code lines and stay
+        # in scope (issue #285).
+        scan_source = @check_remarks ? source : mask_embedded_remarks(source)
+        scan_lines_for_non_ascii(file_violations, scan_source.lines)
+        file_violations
+      end
+
+      def process_non_ascii_char(char)
+        # Skip ASCII characters
+        return nil if char.ord <= 0x7F
+
+        code_point = char.ord
+        hex = "0x#{code_point.to_s(16)}"
+        utf8 = code_point.chr(Encoding::UTF_8).bytes.map do |b|
+          "0x#{b.to_s(16)}"
+        end.join(" ")
+
+        # Check if it's a math symbol
+        if asciimath = @unicode_to_asciimath[char]
+          return {
+            char: char,
+            hex: hex,
+            utf8: utf8,
+            is_math: true,
+            replacement: asciimath,
+            replacement_type: "asciimath",
+          }
+        end
+
+        # Not a math symbol, use ISO encoding
+        {
+          char: char,
+          hex: hex,
+          utf8: utf8,
+          is_math: false,
+          replacement: encode_express_format(char),
+          replacement_type: "iso-10303-11",
+        }
+      end
+
+      def encode_express_format(char)
+        code_point = char.ord
+
+        # Format the encoded value with double quotes
+        if code_point < 0x10000
+          "\"#{sprintf('%08X', code_point)}\"" # e.g., "00000041" for 'A'
+        else
+          # For higher code points, use all four octets
+          group = (code_point >> 24) & 0xFF
+          plane = (code_point >> 16) & 0xFF
+          row = (code_point >> 8) & 0xFF
+          cell = code_point & 0xFF
+
+          "\"#{sprintf('%02X%02X%02X%02X', group, plane, row, cell)}\""
+        end
+      end
+    end
+  end
+end