Finish 2.1.1

dryruby · Jul 26, 2020 · e2414fb · e2414fb
2 parents 8aae6c0 + f2fdfbd
commit e2414fb
Show file tree

Hide file tree

Showing 8 changed files with 66 additions and 47 deletions.
diff --git a/.travis.yml b/.travis.yml
@@ -1,8 +1,9 @@
 language: ruby
 script: "bundle exec rspec spec"
 env:
-  - NOKOGIRI_USE_SYSTEM_LIBRARIES=true
   - CI=true
+  global:
+    - NOKOGIRI_USE_SYSTEM_LIBRARIES=true
 rvm:
   - 2.4
   - 2.5

diff --git a/VERSION b/VERSION
@@ -1 +1 @@
-2.1.0
+2.1.1
diff --git a/ebnf.gemspec b/ebnf.gemspec
@@ -9,7 +9,7 @@ Gem::Specification.new do |gem|
   gem.homepage              = "https://github.com/dryruby/ebnf"
   gem.license               = 'Unlicense'
   gem.summary               = "EBNF parser and parser generator."
-  gem.description           = %q{EBNF is a Ruby parser for W3C EBNF and a parser generator for compliant LL(1) grammars.}
+  gem.description           = %q{EBNF is a Ruby parser for W3C EBNF and a parser generator for PEG and LL(1). Also includes parsing modes for ISO EBNF and ABNF.}
 
   gem.authors               = ['Gregg Kellogg']
   gem.email                 = 'public-rdf-ruby@w3.org'
@@ -27,10 +27,11 @@ Gem::Specification.new do |gem|
   gem.add_runtime_dependency     'sxp',             '~> 1.1'
   gem.add_runtime_dependency     'scanf',           '~> 1.0'
   gem.add_runtime_dependency     'rdf',             '~> 3.1' # Required by sxp
+  gem.add_runtime_dependency     'htmlentities',    '~> 4.3'
   gem.add_development_dependency 'rdf-spec',        '~> 3.1'
   gem.add_development_dependency 'rdf-turtle',      '~> 3.1'
-  gem.add_development_dependency 'erubis',          '~> 2.7'
   gem.add_development_dependency 'nokogiri',        '~> 1.10'
+  gem.add_development_dependency 'erubis',          '~> 2.7'
   gem.add_development_dependency 'rspec',           '~> 3.9'
   gem.add_development_dependency 'rspec-its',       '~> 1.3'
   gem.add_development_dependency 'yard',            '~> 0.9'

diff --git a/etc/doap.ttl b/etc/doap.ttl
@@ -13,10 +13,17 @@
   doap:homepage      <https://github.com/dryruby/ebnf> ;
   doap:license       <https://unlicense.org/1.0/> ;
   doap:shortdesc     "EBNF parser and parser generator"@en ;
-  doap:description   "EBNF is a Ruby parser for W3C EBNF and a parser generator for compliant LL(1) grammars."@en ;
+  doap:description   "EBNF is a Ruby parser for W3C EBNF and a parser generator for PEG and LL(1). Also includes parsing modes for ISO EBNF and ABNF."@en ;
   doap:created       "2011-08-29"^^xsd:date ;
   doap:programming-language "Ruby" ;
-  doap:implements    <http://dbpedia.org/resource/Compiler-compiler> ;
+  doap:implements    <http://dbpedia.org/resource/Compiler-compiler>,
+                     <https://en.wikipedia.org/wiki/LL_parser>,
+                     <https://en.wikipedia.org/wiki/Parsing_expression_grammar>,
+                     <https://pdos.csail.mit.edu/~baford/packrat/thesis/>,
+                     <https://www.w3.org/TR/REC-xml/#sec-notation>,
+                     <https://en.wikipedia.org/wiki/Backus–Naur_form>,
+                     <https://www.iso.org/standard/26153.html>,
+                     <https://www.rfc-editor.org/rfc/rfc5234>;
   doap:category      <http://dbpedia.org/resource/Resource_Description_Framework>,
                      <http://dbpedia.org/resource/Ruby_(programming_language)> ;
   doap:download-page <> ;
@@ -28,6 +35,6 @@
   doap:documenter    <https://greggkellogg.net/foaf#me> ;
   foaf:maker         <https://greggkellogg.net/foaf#me> ;
   dc:title           "ebnf" ;
-  dc:description     "EBNF is a Ruby parser for W3C EBNF and a parser generator for compliant LL(1) grammars."@en ;
+  dc:description     "EBNF is a Ruby parser for W3C EBNF and a parser generator for PEG and LL(1). Also includes parsing modes for ISO EBNF and ABNF."@en ;
   dc:date            "2011-08-29"^^xsd:date ;
   dc:creator         <https://greggkellogg.net/foaf#me> .
diff --git a/lib/ebnf/ll1/scanner.rb b/lib/ebnf/ll1/scanner.rb
@@ -69,7 +69,6 @@ def eos?
     # @return [String]
     def rest
       feed_me
-      @lineno += 1 if eos?
       encode_utf8 super
     end
 

diff --git a/lib/ebnf/peg/parser.rb b/lib/ebnf/peg/parser.rb
@@ -75,8 +75,6 @@ def terminal_regexps; (@terminal_regexps ||= {}); end
       # @option options [Hash{String => String}] :map ({})
       #   A mapping from terminals, in lower-case form, to
       #   their canonical value
-      # @option options [Boolean] :unescape
-      #   Cause strings and codepoints to be unescaped.
       # @yield [value, prod]
       # @yieldparam [String] value
       #   The scanned terminal value.
@@ -269,7 +267,8 @@ def clear_packrat; @packrat.clear; end
     # @param [String] message Error string
     # @param [Hash{Symbol => Object}] options
     # @option options [URI, #to_s] :production
-    # @option options [Token] :token
+    # @option options [Boolean] :raise abort furhter processing
+    # @option options [Array] :backtrace state where error occured
     # @see #debug
     def error(node, message, **options)
       lineno = options[:lineno] || (scanner.lineno if scanner)
@@ -282,7 +281,11 @@ def error(node, message, **options)
       @recovering = true
       debug(node, m, level: 3, **options)
       if options[:raise] || @options[:validate]
-        raise Error.new(m, lineno: lineno, rest: options[:rest], production: options[:production])
+        raise Error.new(m,
+                lineno: lineno,
+                rest: options[:rest],
+                production: options[:production],
+                backtrace: options[:backtrace])
       end
     end
 
@@ -365,25 +368,27 @@ def onStart(prod)
       @productions << prod
       debug("#{prod}(:start)", "",
         lineno: (scanner.lineno if scanner),
-        pos: (scanner.pos if scanner),
-        depth: (depth + 1)) {"#{prod}, pos: #{scanner ? scanner.pos : '?'}, rest: #{scanner ? scanner.rest[0..20].inspect : '?'}"}
+        pos: (scanner.pos if scanner)
+      ) do
+          "#{prod}, pos: #{scanner ? scanner.pos : '?'}, rest: #{scanner ? scanner.rest[0..20].inspect : '?'}"
+      end
       if handler
         # Create a new production data element, potentially allowing handler
         # to customize before pushing on the @prod_data stack
-        data = {}
+        data = {_production: prod}
         begin
           self.class.eval_with_binding(self) {
             handler.call(data, @parse_callback)
           }
         rescue ArgumentError, Error => e
-          error("start", "#{e.class}: #{e.message}", production: prod)
+          error("start", "#{e.class}: #{e.message}", production: prod, backtrace: e.backtrace)
           @recovering = false
         end
         @prod_data << data
       elsif self.class.production_handlers[prod]
         # Make sure we push as many was we pop, even if there is no
         # explicit start handler
-        @prod_data << {}
+        @prod_data << {_production: prod}
       end
       return self.class.start_options.fetch(prod, {}) # any options on this production
     end
@@ -397,21 +402,23 @@ def onFinish(result)
       prod = @productions.last
       handler, clear_packrat = self.class.production_handlers[prod]
       data = @prod_data.pop if handler || self.class.start_handlers[prod]
+      error("finish",
+        "prod_data production mismatch: expected #{prod.inspect}, got #{data[:_production].inspect}",
+        production: prod, prod_data: @prod_data) if data && prod != data[:_production]
       if handler && !@recovering && result != :unmatched
         # Pop production data element from stack, potentially allowing handler to use it
         result = begin
           self.class.eval_with_binding(self) {
             handler.call(result, data, @parse_callback)
           }
         rescue ArgumentError, Error => e
-          error("finish", "#{e.class}: #{e.message}", production: prod)
+          error("finish", "#{e.class}: #{e.message}", production: prod, backtrace: e.backtrace)
           @recovering = false
         end
       end
-      progress("#{prod}(:finish)", "",
-               depth: (depth + 1),
-               lineno: (scanner.lineno if scanner),
-               level: result == :unmatched ? 0 : 1) do
+      debug("#{prod}(:finish)", "",
+             lineno: (scanner.lineno if scanner),
+             level: result == :unmatched ? 0 : 1) do
         "#{result.inspect}@(#{scanner ? scanner.pos : '?'}), rest: #{scanner ? scanner.rest[0..20].inspect : '?'}"
       end
       self.clear_packrat if clear_packrat
@@ -433,12 +440,12 @@ def onTerminal(prod, value)
             handler.call(value, parentProd, @parse_callback)
           }
         rescue ArgumentError, Error => e
-          error("terminal", "#{e.class}: #{e.message}", value: value, production: prod)
+          error("terminal", "#{e.class}: #{e.message}", value: value, production: prod, backtrace: e.backtrace)
           @recovering = false
         end
       end
       progress("#{prod}(:terminal)", "",
-               depth: (depth + 2),
+               depth: (depth + 1),
                lineno: (scanner.lineno if scanner),
                level: value == :unmatched ? 0 : 1) do
         "#{value.inspect}@(#{scanner ? scanner.pos : '?'})"

diff --git a/lib/ebnf/writer.rb b/lib/ebnf/writer.rb
@@ -8,6 +8,7 @@
 module EBNF
   class Writer
     LINE_LENGTH = 80
+    LINE_LENGTH_HTML = 200
 
     # ASCII escape names
     ASCII_ESCAPE_NAMES = [
@@ -118,19 +119,21 @@ def initialize(rules, out: $stdout, html: false, format: :ebnf, **options)
         lhs_fmt = "%<id>-#{max_id+2}s " + lhs_fmt
         lhs_length += max_id + 3
       end
-      rhs_length = LINE_LENGTH - lhs_length
+      rhs_length = (html ? LINE_LENGTH_HTML : LINE_LENGTH) - lhs_length
 
       if html
         # Output as formatted HTML
         begin
           require 'erubis'
+          require 'htmlentities'
+          @coder = HTMLEntities.new
           eruby = Erubis::Eruby.new(ERB_DESC)
           formatted_rules = rules.map do |rule|
             if rule.kind == :terminals || rule.kind == :pass
               OpenStruct.new(id: ("@#{rule.kind}"),
                              sym: nil,
                              assign: nil,
-                             formatted: ("<strong>Productions for terminals</strong>" if rule.kind == :terminals))
+                             formatted: ("<strong># Productions for terminals</strong>" if rule.kind == :terminals))
             else
               formatted_expr = self.send(format_meth, rule.expr)
               # Measure text without markup
@@ -151,7 +154,7 @@ def initialize(rules, out: $stdout, html: false, format: :ebnf, **options)
                     formatted.sub!(%r{\s*<code>\|</code>\s*}, '')
                     (ndx > 0 ? (rule.alt? ? '|' : '') : '=')
                   end
-                  lines << OpenStruct.new(id: ("[#{rule.id}]" if rule.id),
+                  lines << OpenStruct.new(id: ((ndx == 0 ? "[#{rule.id}]" : "") if rule.id),
                                           sym: (rule.sym if ndx == 0 || format == :abnf),
                                           assign: assign,
                                           formatted: formatted)
@@ -171,7 +174,7 @@ def initialize(rules, out: $stdout, html: false, format: :ebnf, **options)
           out.write eruby.evaluate(format: format, rules: formatted_rules)
           return
         rescue LoadError
-          $stderr.puts "Generating HTML requires erubis gem to be loaded"
+          $stderr.puts "Generating HTML requires erubis and htmlentities gems to be loaded"
         end
       end
 
@@ -216,7 +219,7 @@ def initialize(rules, out: $stdout, html: false, format: :ebnf, **options)
 
     # Format the expression part of a rule
     def format_ebnf(expr, sep: nil, embedded: false)
-      return (@options[:html] ? %(<a href="#grammar-production-#{expr}">#{expr}</a>) : expr.to_s) if expr.is_a?(Symbol)
+      return (@options[:html] ? %(<a href="#grammar-production-#{@coder.encode expr}">#{@coder.encode expr}</a>) : expr.to_s) if expr.is_a?(Symbol)
       if expr.is_a?(String)
         return expr.length == 1 ?
           format_ebnf_char(expr) :
@@ -290,10 +293,10 @@ def format_ebnf(expr, sep: nil, embedded: false)
     # Format a single-character string, prefering hex for non-main ASCII
     def format_ebnf_char(c)
       case c.ord
-      when (0x21)         then (@options[:html] ? %("<code class="grammar-literal">#{c}</code>") : %{"#{c}"})
-      when 0x22           then (@options[:html] ? %('<code class="grammar-literal">"</code>') : %{'"'})
-      when (0x23..0x7e)   then (@options[:html] ? %("<code class="grammar-literal">#{c}</code>") : %{"#{c}"})
-      when (0x80..0xFFFD) then (@options[:html] ? %("<code class="grammar-literal">#{c}</code>") : %{"#{c}"})
+      when (0x21)         then (@options[:html] ? %("<code class="grammar-literal">#{@coder.encode c}</code>") : %{"#{c}"})
+      when 0x22           then (@options[:html] ? %('<code class="grammar-literal">&quot;</code>') : %{'"'})
+      when (0x23..0x7e)   then (@options[:html] ? %("<code class="grammar-literal">#{@coder.encode c}</code>") : %{"#{c}"})
+      when (0x80..0xFFFD) then (@options[:html] ? %("<code class="grammar-literal">#{@coder.encode c}</code>") : %{"#{c}"})
       else                     escape_ebnf_hex(c)
       end
     end
@@ -308,7 +311,7 @@ def format_ebnf_range(string)
       while !s.eos?
         case
         when s.scan(/\A[!"\u0024-\u007e]+/)
-          buffer << (@options[:html] ? %(<code class="grammar-literal">#{s.matched}</code>) : s.matched)
+          buffer << (@options[:html] ? %(<code class="grammar-literal">#{@coder.encode s.matched}</code>) : s.matched)
         when s.scan(/\A#x\h+/)
           buffer << escape_ebnf_hex(s.matched[2..-1].hex.chr(Encoding::UTF_8))
         else
@@ -328,7 +331,8 @@ def format_ebnf_string(string, quote = '"')
         end
       end
 
-      "#{quote}#{string}#{quote}"
+      res = "#{quote}#{string}#{quote}"
+      @options[:html] ? @coder.encode(res) : res
     end
 
     def escape_ebnf_hex(u)
@@ -341,11 +345,11 @@ def escape_ebnf_hex(u)
       char = fmt % u.ord
       if @options[:html]
         if u.ord <= 0x20
-          char = %(<abbr title="#{ASCII_ESCAPE_NAMES[u.ord]}">#{char}</abbr>)
+          char = %(<abbr title="#{ASCII_ESCAPE_NAMES[u.ord]}">#{@coder.encode char}</abbr>)
         elsif u.ord < 0x7F
-          char = %(<abbr title="ascii '#{u}'">#{char}</abbr>)
+          char = %(<abbr title="ascii '#{@coder.encode u}'">#{@coder.encode char}</abbr>)
         elsif u.ord == 0x7F
-          char = %(<abbr title="delete">#{char}</abbr>)
+          char = %(<abbr title="delete">#{@coder.encode char}</abbr>)
         elsif u.ord <= 0xFF
           char = %(<abbr title="extended ascii '#{u}'">#{char}</abbr>)
         else
@@ -363,7 +367,7 @@ def escape_ebnf_hex(u)
 
     # Format the expression part of a rule
     def format_abnf(expr, sep: nil, embedded: false, sensitive: true)
-      return (@options[:html] ? %(<a href="#grammar-production-#{expr}">#{expr}</a>) : expr.to_s) if expr.is_a?(Symbol)
+      return (@options[:html] ? %(<a href="#grammar-production-#{@coder.encode expr}">#{@coder.encode expr}</a>) : expr.to_s) if expr.is_a?(Symbol)
       if expr.is_a?(String)
         if expr.length == 1
           return format_abnf_char(expr)
@@ -380,7 +384,7 @@ def format_abnf(expr, sep: nil, embedded: false, sensitive: true)
           seq.unshift(:seq)
           return format_abnf(seq, sep: nil, embedded: false)
         else
-          return (@options[:html] ? %("<code class="grammar-literal">#{'%s' if sensitive}#{expr}</code>") : %(#{'%s' if sensitive}"#{expr}"))
+          return (@options[:html] ? %("<code class="grammar-literal">#{'%s' if sensitive}#{@coder.encode expr}</code>") : %(#{'%s' if sensitive}"#{expr}"))
         end
       end
       parts = {
@@ -528,11 +532,11 @@ def escape_abnf_hex(u)
       char =  "%x" + (fmt % u.ord)
       if @options[:html]
         if u.ord <= 0x20
-          char = %(<abbr title="#{ASCII_ESCAPE_NAMES[u.ord]}">#{char}</abbr>)
+          char = %(<abbr title="#{ASCII_ESCAPE_NAMES[u.ord]}">#{@coder.encode char}</abbr>)
         elsif u.ord <= 0x7F
-          char = %(<abbr title="ascii '#{u}'">#{char}</abbr>)
+          char = %(<abbr title="ascii '#{u}'">#{@coder.encode char}</abbr>)
         elsif u.ord == 0x7F
-          char = %(<abbr title="delete">#{char}</abbr>)
+          char = %(<abbr title="delete">#{@coder.encode char}</abbr>)
         elsif u.ord <= 0xFF
           char = %(<abbr title="extended ascii '#{u}'">#{char}</abbr>)
         else
@@ -550,17 +554,17 @@ def escape_abnf_hex(u)
 
     # Format the expression part of a rule
     def format_isoebnf(expr, sep: nil, embedded: false)
-      return (@options[:html] ? %(<a href="#grammar-production-#{expr}">#{expr}</a>) : expr.to_s) if expr.is_a?(Symbol)
+      return (@options[:html] ? %(<a href="#grammar-production-#{@coder.encode expr}">#{@coder.encode expr}</a>) : expr.to_s) if expr.is_a?(Symbol)
       if expr.is_a?(String)
         expr = expr[2..-1].hex.chr if expr =~ /\A#x\h+/
         expr.chars.each do |c|
           raise RangeError, "cannot format #{expr.inspect} as an ISO EBNF String: #{c.inspect} is out of range" unless
             ISOEBNF::TERMINAL_CHARACTER.match?(c)
         end
         if expr =~ /"/
-          return (@options[:html] ? %('<code class="grammar-literal">#{expr}</code>') : %('#{expr}'))
+          return (@options[:html] ? %('<code class="grammar-literal">#{@coder.encode expr}</code>') : %('#{expr}'))
         else
-          return (@options[:html] ? %("<code class="grammar-literal">#{expr}</code>") : %("#{expr}"))
+          return (@options[:html] ? %("<code class="grammar-literal">#{@coder.encode expr}</code>") : %("#{expr}"))
         end
       end
       parts = {

diff --git a/spec/ll1/scanner_spec.rb b/spec/ll1/scanner_spec.rb
@@ -21,7 +21,7 @@
       f = double("input")
       expect(f).to receive(:read).and_return("ascii".force_encoding(Encoding::ASCII_8BIT))
       expect(f).to receive(:gets).and_return("utf8".force_encoding(Encoding::UTF_8))
-      expect(f).to receive(:eof?).and_return(false, false, true, true)
+      expect(f).to receive(:eof?).and_return(false, false, true)
       scanner = EBNF::LL1::Scanner.new(f)
       s = scanner.rest
       expect(s).to eq "asciiutf8"