Tokenise error messages in Console lexer

lib/rouge/lexers/console.rb

@@ -9,12 +9,14 @@ module Lexers
     # line before passing the remainder of the line to the language lexer for
     # the shell (by default, the {Shell} lexer).
     #
-    # The {ConsoleLexer} class accepts four options:
+    # The {ConsoleLexer} class accepts five options:
     # 1. **lang**: the shell language to lex (default: `shell`);
     # 2. **output**: the output language (default: `plaintext?token=Generic.Output`);
     # 3. **prompt**: comma-separated list of strings that indicate the end of a
     #    prompt (default: `$,#,>,;`);
     # 4. **comments**: whether to enable comments.
+    # 5. **error**: comma-separated list of strings that indicate the start of an
+    #    error message
     #
     # The comments option, if enabled, will lex lines that begin with a `#` as a
     # comment. Please note that this option will only work if the prompt is
@@ -39,27 +41,38 @@ class ConsoleLexer < Lexer
       tag 'console'
       aliases 'terminal', 'shell_session', 'shell-session'
       filenames '*.cap'
-      desc 'A generic lexer for shell sessions. Accepts ?lang and ?output lexer options, a ?prompt option, and ?comments to enable # comments.'
+      desc 'A generic lexer for shell sessions. Accepts ?lang and ?output lexer options, a ?prompt option, ?comments to enable # comments, and ?error to handle error messages.'
 
       option :lang, 'the shell language to lex (default: shell)'
       option :output, 'the output language (default: plaintext?token=Generic.Output)'
       option :prompt, 'comma-separated list of strings that indicate the end of a prompt. (default: $,#,>,;)'
       option :comments, 'enable hash-comments at the start of a line - otherwise interpreted as a prompt. (default: false, implied by ?prompt not containing `#`)'
+      option :error, 'comma-separated list of strings that indicate the start of an error message'
 
       def initialize(*)
         super
         @prompt = list_option(:prompt) { nil }
         @lang = lexer_option(:lang) { 'shell' }
         @output = lexer_option(:output) { PlainText.new(token: Generic::Output) }
         @comments = bool_option(:comments) { :guess }
+        @error = list_option(:error) { nil }
       end
 
-      def prompt_regex
-        @prompt_regex ||= begin
-          /^#{prompt_prefix_regex}(?:#{end_chars.map(&Regexp.method(:escape)).join('|')})/
+      # whether to allow comments. if manually specifying a prompt that isn't
+      # simply "#", we flag this to on
+      def allow_comments?
+        case @comments
+        when :guess
+          @prompt && !@prompt.empty? && !end_chars.include?('#')
+        else
+          @comments
         end
       end
 
+      def comment_regex
+        /\A\s*?#/
+      end
+
       def end_chars
         @end_chars ||= if @prompt.any?
           @prompt.reject { |c| c.empty? }
@@ -70,22 +83,9 @@ def end_chars
         end
       end
 
-      # whether to allow comments. if manually specifying a prompt that isn't
-      # simply "#", we flag this to on
-      def allow_comments?
-        case @comments
-        when :guess
-          @prompt && !@prompt.empty? && !end_chars.include?('#')
-        else
-          @comments
-        end
-      end
-
-      def prompt_prefix_regex
-        if allow_comments?
-          /[^<#]*?/m
-        else
-          /.*?/m
+      def error_regex
+        @error_regex ||= if @error.any?
+          /^(?:#{@error.map(&Regexp.method(:escape)).join('|')})/
         end
       end
 
@@ -102,6 +102,10 @@ def lang_lexer
         end
       end
 
+      def line_regex
+        /(\\.|[^\\])*?(\n|$)/m
+      end
+
       def output_lexer
         @output_lexer ||= case @output
         when nil
@@ -115,22 +119,6 @@ def output_lexer
         end
       end
 
-      def line_regex
-        /(\\.|[^\\])*?(\n|$)/m
-      end
-
-      def comment_regex
-        /\A\s*?#/
-      end
-
-      def stream_tokens(input, &output)
-        input = StringScanner.new(input)
-        lang_lexer.reset!
-        output_lexer.reset!
-
-        process_line(input, &output) while !input.eos?
-      end
-
       def process_line(input, &output)
         input.scan(line_regex)
 
@@ -162,13 +150,41 @@ def process_line(input, &output)
           lang_lexer.reset!
 
           yield Comment, input[0]
+        elsif error_regex =~ input[0]
+          puts "console: matched error #{input[0].inspect}" if @debug
+          output_lexer.reset!
+          lang_lexer.reset!
+
+          yield Generic::Error, input[0]
         else
           puts "console: matched output #{input[0].inspect}" if @debug
           lang_lexer.reset!
 
           output_lexer.continue_lex(input[0], &output)
         end
       end
+
+      def prompt_prefix_regex
+        if allow_comments?
+          /[^<#]*?/m
+        else
+          /.*?/m
+        end
+      end
+
+      def prompt_regex
+        @prompt_regex ||= begin
+          /^#{prompt_prefix_regex}(?:#{end_chars.map(&Regexp.method(:escape)).join('|')})/
+        end
+      end
+
+      def stream_tokens(input, &output)
+        input = StringScanner.new(input)
+        lang_lexer.reset!
+        output_lexer.reset!
+
+        process_line(input, &output) while !input.eos?
+      end
     end
   end
 end

spec/lexers/console_spec.rb

@@ -22,6 +22,16 @@
       ['Text', 'foo']
   end
 
+  it 'parses a custom error' do
+    subject_with_options = klass.new({ error: 'No command,Unhandled' })
+    assert_tokens_equal 'No command \'foo\' found, did you mean:', subject_with_options,
+      ['Generic.Error', 'No command \'foo\' found, did you mean:']
+    assert_tokens_equal 'Unhandled condition in test.lisp', subject_with_options,
+      ['Generic.Error', 'Unhandled condition in test.lisp']
+    assert_tokens_equal 'foo', subject_with_options,
+      ['Generic.Output', 'foo']
+  end
+
   it 'parses single-line comments' do
     subject_with_options = klass.new({ comments: true })
     assert_tokens_equal '# this is a comment', subject_with_options,