comment_strip-ruby 0.2.0 → 0.2.1

data/lib/comment_strip/language_families/c.rb

@@ -1,14 +1,15 @@
+# frozen_string_literal: true
 # ######################################################################## #
 # File:     comment_strip/language_families/c.rb
 #
 # Purpose:  Definition of strip() function for C-family languages.
 #
 # Created:  14th September 2020
-# Updated:  31st March 2024
+# Updated:  30th August 2026
 #
-# Home:     http://github.com/synesissoftware/comment_strip.r
+# Home:     https://github.com/synesissoftware/comment_strip.r
 #
-# Copyright (c) 2020-2024, Matthew Wilson and Synesis Information Systems
+# Copyright (c) 2020-2026, Matthew Wilson and Synesis Information Systems
 # All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
@@ -47,6 +48,23 @@ module LanguageFamilies
 
 module C
 
+  def self.decimal_digit? character
+
+    character && character >= ?0 && character <= ?9
+  end
+
+  def self.identifier_character? character
+
+    return false unless character
+
+    case character
+    when ?a..?z, ?A..?Z, ?0..?9, '_'
+      true
+    else
+      false
+    end
+  end
+
   def self.strip input, lf, **options
 
     return input if input.nil?
@@ -57,29 +75,63 @@ module C
 
     # States:
     #
-    # - :c_comment          - in a C comment, i.e. from immediately after "/*"
-    # - :c_comment_star     - in a C comment when just received '*', e.g. from immediately after "/* comment *"
-    # - :cpp_comment        - in a C++ comment, i.e. from immediately after "//"
-    # - :dq_string          - within a double-quoted string
-    # - :dq_string_escape   - within a double-quoted string when just received a '"', e.g. from immediately after '"the next word is quoted \'
-    # - :slash_start        - having found a slash (not in a string)
-    # - :sq_string_closing  - waiting for final '\'' in a single-quoted string
-    # - :sq_string_escape   - within a escaped single-quoted string, i.e. from immediately after "'\"
-    # - :sq_string_open     - within a single-quoted string, i.e. from immediately after "'"
-    # - :text               - regular part of the code
+    # - :c_comment              - in a C comment, i.e. from immediately after "/*"
+    # - :c_comment_star         - in a C comment when just received '*', e.g. from immediately after "/* comment *"
+    # - :cpp_comment            - in a C++ comment, i.e. from immediately after "//"
+    # - :dq_string              - within a double-quoted string
+    # - :dq_string_escape       - within a double-quoted string when just received a '"', e.g. from immediately after '"the next word is quoted \'
+    # - :raw_string             - within a C++ raw string
+    # - :raw_string_closing     - waiting for the final '"' in a C++ raw string
+    # - :backtick_string        - within a Go raw string
+    # - :backtick_string_open   - waiting for the opening '`' in a Go raw string
+    # - :lifetime               - within a Rust lifetime name
+    # - :verbatim_string        - within a C# verbatim string
+    # - :verbatim_string_open   - waiting for the opening '"' in a C# verbatim string
+    # - :verbatim_string_quote  - waiting to determine whether a C# quote is doubled
+    # - :slash_start            - having found a slash (not in a string)
+    # - :sq_string_closing      - waiting for final '\'' in a single-quoted string
+    # - :sq_string_escape       - within a escaped single-quoted string, i.e. from immediately after "'\"
+    # - :sq_string_open         - within a single-quoted string, i.e. from immediately after "'"
+    # - :text                   - regular part of the code
 
     state   =   :text
 
-    r       =   ''
+    r       =   String.new
+    r.force_encoding input.encoding
 
     cc_lines =   0
+    previous_was_cr = false
+
+    i = 0
+    while i < input.length
+      c = input[i]
+
+      if ?\\ == c
+        next_c = input[i + 1]
+
+        if ?\n == next_c
+          i += 2
+          previous_was_cr = false
+          next
+        elsif ?\r == next_c
+          i += 2
+          previous_was_cr = false
+
+          if ?\n == input[i]
+            i += 1
+            next
+          end
 
-    input.each_char do |c|
+          next if i >= input.length
+
+          c = input[i]
+        end
+      end
 
       case c
       when ?\r, ?\n
 
-        line += 1
+        line += 1 unless ?\n == c && previous_was_cr
         column = 0
       else
 
@@ -88,13 +140,23 @@ module C
 
       skip = false
 
+      if :text == state && 'R' == c && '"' == input[i + 1] && '(' == input[i + 2]
+        state = :raw_string
+      elsif :text == state && '@' == c && '"' == input[i + 1]
+        state = :verbatim_string_open
+      elsif :text == state && i > 0 && '<' == input[i - 1] && input[i + 1] && input[i + 1] >= ?a && input[i + 1] <= ?z
+        state = :lifetime
+      elsif :text == state && '`' == c
+        state = :backtick_string_open
+      end
+
       case c
       when ?\r, ?\n
 
         case state
         when :c_comment, :c_comment_star
 
-          cc_lines += 1
+          cc_lines += 1 unless ?\n == c && previous_was_cr
 
           state = :c_comment
         when :cpp_comment
@@ -121,6 +183,37 @@ module C
         # - for comment-star
 
         case state
+        when :backtick_string_open
+
+          state = :backtick_string if '`' == c
+        when :backtick_string
+
+          state = :text if '`' == c
+        when :lifetime
+
+          if !identifier_character?(c)
+            state = :text
+          elsif !identifier_character?(input[i + 1])
+            state = :text
+          end
+        when :verbatim_string_open
+
+          state = :verbatim_string if '"' == c
+        when :verbatim_string
+
+          if '"' == c
+            state = :verbatim_string_quote if '"' == input[i + 1]
+            state = :text unless '"' == input[i + 1]
+          end
+        when :verbatim_string_quote
+
+          state = :verbatim_string if '"' == c
+        when :raw_string
+
+          state = :raw_string_closing if ')' == c && '"' == input[i + 1]
+        when :raw_string_closing
+
+          state = :text if '"' == c
         when :sq_string_open
 
           state = (?\\ == c) ? :sq_string_escape : :sq_string_closing
@@ -195,7 +288,11 @@ module C
               case state
               when :text
 
-                state = :sq_string_open
+                if i > 0 && decimal_digit?(input[i - 1]) && decimal_digit?(input[i + 1])
+                  ;
+                else
+                  state = :sq_string_open
+                end
               when :sq_string_closing
 
                 state = :text
@@ -259,8 +356,14 @@ module C
 
         r << c unless skip
       end
+
+      previous_was_cr = ?\r == c
+      i += 1
     end
 
+    r << '/' if :slash_start == state
+    r << ?\n * cc_lines if [ :c_comment, :c_comment_star, ].include?(state)
+
     r
   end
 end # module C

data/lib/comment_strip/language_families/hash_line.rb

@@ -1,3 +1,4 @@
+# frozen_string_literal: true
 # ######################################################################## #
 # File:     comment_strip/language_families/hash_line.rb
 #
@@ -5,11 +6,11 @@
 #           line comments beginning at the # character.
 #
 # Created:  1st December 2023
-# Updated:  31st March 2024
+# Updated:  30th August 2026
 #
-# Home:     http://github.com/synesissoftware/comment_strip.r
+# Home:     https://github.com/synesissoftware/comment_strip.r
 #
-# Copyright (c) 2023-2024, Matthew Wilson and Synesis Information Systems
+# Copyright (c) 2023-2026, Matthew Wilson and Synesis Information Systems
 # All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
@@ -48,6 +49,18 @@ module LanguageFamilies
 
 module HashLine
 
+  def self.identifier_character? character
+
+    return false unless character
+
+    case character
+    when ?a..?z, ?A..?Z, ?0..?9, '_'
+      true
+    else
+      false
+    end
+  end
+
   def self.strip input, lf, **options
 
     return input if input.nil?
@@ -58,21 +71,66 @@ module HashLine
 
     # States:
     #
-    # - :dq_string          - within a double-quoted string
-    # - :dq_string_escape   - within a double-quoted string when just received a '"', e.g. from immediately after '"the next word is quoted \'
-    # - :hash_comment       - in a #-comment, i.e. from immediately after "#"
-    # - :sq_string_closing  - waiting for final '\'' in a single-quoted string
-    # - :sq_string_escape   - within a escaped single-quoted string, i.e. from immediately after "'\"
-    # - :sq_string_open     - within a single-quoted string, i.e. from immediately after "'"
-    # - :text               - regular part of the code
+    # - :dq_string            - within a double-quoted string
+    # - :dq_string_escape     - within a double-quoted string when just received a '"', e.g. from immediately after '"the next word is quoted \'
+    # - :hash_comment         - in a #-comment, i.e. from immediately after "#"
+    # - :heredoc_body         - within a Ruby heredoc body
+    # - :heredoc_header       - within a Ruby heredoc header
+    # - :percent_string       - within a Ruby percent string
+    # - :percent_string_open  - waiting for the opening delimiter of a Ruby percent string
+    # - :sq_string_closing    - waiting for final '\'' in a single-quoted string
+    # - :sq_string_escape     - within a escaped single-quoted string, i.e. from immediately after "'\"
+    # - :sq_string_open       - within a single-quoted string, i.e. from immediately after "'"
+    # - :text                 - regular part of the code
 
     state   =   :text
 
-    r       =   ''
+    r       =   String.new
+    r.force_encoding input.encoding
+
+    percent_open = nil
+    percent_closing = nil
+    heredoc_terminator = nil
+    at_line_start = true
+
+    i = 0
+    while i < input.length
+      c = input[i]
+      percent_start = false
+
+      if :text == state && '%' == c
+        percent_type = input[i + 1]
+        percent_delimiter = input[i + 2]
+
+        if [ 'q', 'w', ].include?(percent_type) && percent_delimiter
+          percent_open = percent_delimiter
+          percent_closing = case percent_delimiter
+                            when '{' then '}'
+                            when '[' then ']'
+                            when '(' then ')'
+                            else percent_open
+                            end
+          percent_start = true
+        end
+      end
 
-    cc_lines =   0
+      if :text == state && '<' == c && '<' == input[i + 1]
+        heredoc_offset = i + 2
+        heredoc_offset += 1 if '~' == input[heredoc_offset]
+        heredoc_quote = input[heredoc_offset]
+        heredoc_quote = nil unless [ '\'', '"', ].include?(heredoc_quote)
+        heredoc_offset += 1 if [ '\'', '"', ].include?(heredoc_quote)
+        heredoc_start = heredoc_offset
 
-    input.each_char do |c|
+        while heredoc_offset < input.length && identifier_character?(input[heredoc_offset])
+          heredoc_offset += 1
+        end
+
+        if heredoc_start < heredoc_offset && (!heredoc_quote || heredoc_quote == input[heredoc_offset])
+          heredoc_terminator = input[heredoc_start...heredoc_offset]
+          state = :heredoc_header
+        end
+      end
 
       case c
       when ?\r, ?\n
@@ -99,9 +157,9 @@ module HashLine
         when :dq_string_escape
 
           state = :dq_string
-        when :hash_comment
+        when :heredoc_header
 
-          state = :text
+          state = :heredoc_body
         end
       else
 
@@ -111,12 +169,39 @@ module HashLine
         # - for slash-start
 
         case state
+        when :heredoc_body
+
+          if at_line_start
+            heredoc_match = true
+            heredoc_index = 0
+
+            while heredoc_index < heredoc_terminator.length
+              heredoc_match = false unless input[i + heredoc_index] ==
+                                            heredoc_terminator[heredoc_index]
+              heredoc_index += 1
+            end
+
+            if heredoc_match && [ nil, ?\r, ?\n, ].include?(input[i + heredoc_terminator.length])
+              state = :text
+            end
+          end
+        when :percent_string_open
+
+          state = :percent_string if c == percent_open
+        when :percent_string
+
+          state = :text if c == percent_closing
         when :sq_string_open
 
-          state = (?\\ == c) ? :sq_string_escape : :sq_string_closing
+          case c
+          when ?\\
+            state = :sq_string_escape
+          when ?\'
+            state = :text
+          end
         when :sq_string_escape
 
-          state = :sq_string_closing
+          state = :sq_string_open
         when :dq_string_escape
 
           state = :dq_string
@@ -139,9 +224,6 @@ module HashLine
             when :text
 
               state = :sq_string_open
-            when :sq_string_closing
-
-              state = :text
             else
 
               ;
@@ -198,6 +280,10 @@ module HashLine
 
         r << c unless skip
       end
+
+      state = :percent_string_open if percent_start
+      at_line_start = ?\r == c || ?\n == c
+      i += 1
     end
 
     r

data/lib/comment_strip/strip.rb

@@ -1,14 +1,15 @@
+# frozen_string_literal: true
 # ######################################################################## #
 # File:     comment_strip/strip.rb
 #
 # Purpose:  Definition of strip() function
 #
 # Created:  14th September 2020
-# Updated:  30th March 2024
+# Updated:  19th August 2026
 #
-# Home:     http://github.com/synesissoftware/comment_strip.r
+# Home:     https://github.com/synesissoftware/comment_strip.r
 #
-# Copyright (c) 2020-2024, Matthew Wilson and Synesis Information Systems
+# Copyright (c) 2020-2026, Matthew Wilson and Synesis Information Systems
 # All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
@@ -57,22 +58,25 @@ module CommentStrip
     include ::Xqsr3::Quality::ParameterChecking
 
     # Strips comments from an input string, according to the rules and
-    # conventions of a given language-family
+    # conventions of a given language-family.
     #
     # === Signature
     #
     # * *Parameters:*
-    #   - +input+ (+String+, +nil+) the input source code
-    #   - +lf+ (+String+) the name of the language family, which must be one of the following:
-    #      - +'C'+
-    #      - +'Hash_Line'+
-    #   - +options+ (+Hash+) options that moderate the behaviour
+    #   - +input+ (+String+, +nil+) the input source code;
+    #   - +lf+ (+String+) the name of the language family, which must be one of the following listed in the section below;
+    #   - +options+ (+Hash+) options that moderate the behaviour;
     #
     # * *Options:*
     # None currently defined.
     #
-    # === Signature
+    # === Return
     # (+String+) The stripped for of the input.
+    #
+    # === Supported language families
+    # Currently supported language families:
+    #   - +'C'+ - including C, C++, C#, Go, Java, Rust;
+    #   - +'Hash_Line'+ - include Ruby, Python, Ruby, shell;
     def strip input, lf, **options
 
         check_parameter input, 'input', responds_to: [ :each_char, :empty?, :nil?, ], nil: true
@@ -87,7 +91,7 @@ module CommentStrip
             LanguageFamilies::HashLine.strip input, lf, **options
         else
 
-            raise "language family '#{lf}' unrecognised or not supported1"
+            raise "language family '#{lf}' unrecognised or not supported"
         end
     end
 

data/lib/comment_strip/version.rb

@@ -1,14 +1,15 @@
+# frozen_string_literal: true
 # ######################################################################## #
 # File:     comment_strip/version.rb
 #
 # Purpose:  Version for comment_strip.r library
 #
 # Created:  14th September 2020
-# Updated:  30th March 2024
+# Updated:  30th August 2026
 #
-# Home:     http://github.com/synesissoftware/comment_strip.r
+# Home:     https://github.com/synesissoftware/comment_strip.r
 #
-# Copyright (c) 2020-2024, Matthew Wilson and Synesis Information Systems
+# Copyright (c) 2020-2026, Matthew Wilson and Synesis Information Systems
 # All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without
@@ -43,7 +44,7 @@
 module CommentStrip
 
     # Current version of the comment_strip.r library
-    VERSION                 =   '0.2.0'
+    VERSION                 =   '0.2.1'
 
     private
     VERSION_PARTS_          =   VERSION.split(/[.]/).collect { |n| n.to_i } # :nodoc:

data/lib/comment_strip.rb

@@ -1,14 +1,15 @@
+# frozen_string_literal: true
 # ######################################################################## #
-# File:         comment_strip.rb
+# File:     comment_strip.rb
 #
-# Purpose:      Top-level include for comment_strip.r library
+# Purpose:  Top-level include for comment_strip.r library
 #
-# Created:      14th September 2020
-# Updated:      11th July 2022
+# Created:  14th September 2020
+# Updated:  19th August 2026
 #
-# Home:         http://github.com/synesissoftware/comment_strip.r
+# Home:     https://github.com/synesissoftware/comment_strip.r
 #
-# Copyright (c) 2020-2022, Matthew Wilson and Synesis Information Systems
+# Copyright (c) 2020-2026, Matthew Wilson and Synesis Information Systems
 # All rights reserved.
 #
 # Redistribution and use in source and binary forms, with or without

data/test/unit/tc_strip_c.rb

@@ -1,5 +1,12 @@
 #! /usr/bin/env ruby
 
+# NOTE: TAB characters in this test file are intentional and occur only
+# inside HEREDOC fixtures containing imported C/C++ source. They preserve
+# the fixtures' original whitespace so comment stripping can be tested
+# accurately. All Ruby test code and all non-fixture lines must use spaces,
+# with two-space indentation; do not add TABs elsewhere.
+
+
 $:.unshift File.join(__dir__, '../..', 'lib')
 
 
@@ -865,8 +872,6 @@ EOF_main
 #include <new>
 EOF_main
 
-    actual = strip(input, 'C')
-
     assert_equal expected, strip(input, 'C')
   end
 
@@ -2384,7 +2389,7 @@ EOF_main
     assert_equal expected, actual
   end
 
-  def _test_real_sample_6
+  def test_real_sample_6
 
     input = <<-EOF_main
 
@@ -2474,8 +2479,6 @@ namespace {
 
             default:
                 // Check for control characters and invalid utf-8
-EOF_main
-    x1 = <<-EOF_main
                 // Escape control characters in standard ascii
                 // see http://stackoverflow.com/questions/404107/why-are-control-characters-illegal-in-xml-1-0
                 if (c < 0x09 || (c > 0x0D && c < 0x20) || c == 0x7F) {
@@ -2764,9 +2767,6 @@ namespace {
 
             default:
                 
-EOF_main
-    x1 = <<-EOF_main
-
                 
                 
                 if (c < 0x09 || (c > 0x0D && c < 0x20) || c == 0x7F) {