Basic parser setup without tests.

Who needs tests anyway!

Rakefile

@@ -8,10 +8,9 @@ GEMSPEC = Gem::Specification.load('oga.gemspec')
 LEXER_INPUT  = 'lib/oga/lexer.rl'
 LEXER_OUTPUT = 'lib/oga/lexer.rb'
 
-#PARSER_INPUT  = 'lib/oga/parser.y'
-#PARSER_OUTPUT = 'lib/oga/parser.rb'
+HTML_PARSER = 'lib/oga/parser/html.rb'
 
-GENERATED_FILES = ['coverage', 'yardoc', LEXER_OUTPUT]
+GENERATED_FILES = ['coverage', 'yardoc', LEXER_OUTPUT, HTML_PARSER]
 
 GENERATED_FILES.each do |file|
   CLEAN << file if File.exist?(file)

lib/oga.rb

@@ -2,3 +2,4 @@ require 'ast'
 
 require_relative 'oga/ast/node'
 require_relative 'oga/lexer'
+require_relative 'oga/parser/html'

lib/oga/ast/node.rb

@@ -3,7 +3,7 @@ module Oga
     ##
     #
     class Node < ::AST::Node
-
+      attr_reader :line, :column
     end # Node
   end # AST
 end # Oga

lib/oga/parser/html.rb

@@ -0,0 +1,223 @@
+#
+# DO NOT MODIFY!!!!
+# This file is automatically generated by Racc 1.4.11
+# from Racc grammer file "".
+#
+
+require 'racc/parser.rb'
+module Oga
+  module Parser
+    class HTML < Racc::Parser
+
+
+  def initialize
+    @lexer = Lexer.new
+  end
+
+  def reset
+    @lines  = []
+    @line   = 1
+    @column = 1
+  end
+
+  def s(type, *children)
+    return AST::Node.new(
+      type,
+      children.flatten,
+      :line   => @line,
+      :column => @column
+    )
+  end
+
+  def next_token
+    type, value, line, column = @tokens.shift
+
+    @line   = line if line
+    @column = column if column
+
+    return type ? [type, value] : [false, false]
+  end
+
+  def on_error(type, value, stack)
+    name      = token_to_str(type)
+    line_str  = @lines[@line - 1]
+    indicator = '~' * (@column - 1) + '^'
+
+    raise Racc::ParseError, <<-EOF.strip
+Failed to parse the supplied input.
+
+Reason:   unexpected #{name} with value #{value.inspect}
+Location: line #{@line}, column #{@column}
+
+Offending code:
+
+#{line_str}
+#{indicator}
+
+Current stack:
+
+#{stack.inspect}
+    EOF
+  end
+
+  def parse(string)
+    @lines  = string.lines
+    @tokens = @lexer.lex(string)
+    ast     = do_parse
+
+    reset
+
+    return ast
+  end
+
+# vim: set ft=racc:
+##### State transition tables begin ###
+
+racc_action_table = [
+     5,     7,     5,     9,    11,    12,    13,    14,    15,    16,
+    17 ]
+
+racc_action_check = [
+     0,     1,     2,     5,     6,     7,     9,    10,    14,    15,
+    16 ]
+
+racc_action_pointer = [
+    -4,     1,    -2,   nil,   nil,   -12,   -11,     5,   nil,     1,
+     3,   nil,   nil,   nil,     2,    -6,     5,   nil ]
+
+racc_action_default = [
+    -2,    -9,    -1,    -4,    -5,    -9,    -9,    -9,    -3,    -9,
+    -9,    -8,    18,    -6,    -9,    -9,    -9,    -7 ]
+
+racc_goto_table = [
+     3,     1,     8,     2,    10 ]
+
+racc_goto_check = [
+     3,     1,     3,     2,     6 ]
+
+racc_goto_pointer = [
+   nil,     1,     3,     0,   nil,   nil,    -2 ]
+
+racc_goto_default = [
+   nil,   nil,   nil,   nil,     4,     6,   nil ]
+
+racc_reduce_table = [
+  0, 0, :racc_error,
+  1, 17, :_reduce_1,
+  0, 17, :_reduce_2,
+  2, 18, :_reduce_3,
+  1, 18, :_reduce_4,
+  1, 19, :_reduce_none,
+  3, 21, :_reduce_6,
+  6, 20, :_reduce_7,
+  1, 22, :_reduce_none ]
+
+racc_reduce_n = 9
+
+racc_shift_n = 18
+
+racc_token_table = {
+  false => 0,
+  :error => 1,
+  :T_SPACE => 2,
+  :T_NEWLINE => 3,
+  :T_SMALLER => 4,
+  :T_GREATER => 5,
+  :T_SLASH => 6,
+  :T_DQUOTE => 7,
+  :T_SQUOTE => 8,
+  :T_DASH => 9,
+  :T_RBRACKET => 10,
+  :T_LBRACKET => 11,
+  :T_COLON => 12,
+  :T_BANG => 13,
+  :T_EQUALS => 14,
+  :T_TEXT => 15 }
+
+racc_nt_base = 16
+
+racc_use_result_var = false
+
+Racc_arg = [
+  racc_action_table,
+  racc_action_check,
+  racc_action_default,
+  racc_action_pointer,
+  racc_goto_table,
+  racc_goto_check,
+  racc_goto_default,
+  racc_goto_pointer,
+  racc_nt_base,
+  racc_reduce_table,
+  racc_token_table,
+  racc_shift_n,
+  racc_reduce_n,
+  racc_use_result_var ]
+
+Racc_token_to_s_table = [
+  "$end",
+  "error",
+  "T_SPACE",
+  "T_NEWLINE",
+  "T_SMALLER",
+  "T_GREATER",
+  "T_SLASH",
+  "T_DQUOTE",
+  "T_SQUOTE",
+  "T_DASH",
+  "T_RBRACKET",
+  "T_LBRACKET",
+  "T_COLON",
+  "T_BANG",
+  "T_EQUALS",
+  "T_TEXT",
+  "$start",
+  "document",
+  "expressions",
+  "expression",
+  "tag",
+  "tag_start",
+  "tag_body" ]
+
+Racc_debug_parser = false
+
+##### State transition tables end #####
+
+# reduce 0 omitted
+
+def _reduce_1(val, _values)
+ s(:document, val[0]) 
+end
+
+def _reduce_2(val, _values)
+ s(:document) 
+end
+
+def _reduce_3(val, _values)
+ val.compact 
+end
+
+def _reduce_4(val, _values)
+ val[0] 
+end
+
+# reduce 5 omitted
+
+def _reduce_6(val, _values)
+ val[1] 
+end
+
+def _reduce_7(val, _values)
+        s(:element, val[0], val[1])
+      
+end
+
+# reduce 8 omitted
+
+def _reduce_none(val, _values)
+  val[0]
+end
+
+    end   # class HTML
+    end   # module Parser
+  end   # module Oga

lib/oga/parser/html.y

@@ -0,0 +1,104 @@
+class Oga::Parser::HTML
+
+token T_SPACE T_NEWLINE T_SMALLER T_GREATER T_SLASH
+token T_DQUOTE T_SQUOTE T_DASH T_RBRACKET T_LBRACKET
+token T_COLON T_BANG T_EQUALS T_TEXT
+
+options no_result_var
+
+rule
+  document
+    : expressions { s(:document, val[0]) }
+    | /* none */  { s(:document) }
+    ;
+
+  expressions
+    : expressions expression { val.compact }
+    | expression             { val[0] }
+    ;
+
+  expression
+    : tag
+    ;
+
+  tag_start
+    # <p>
+    : T_SMALLER T_TEXT T_GREATER { val[1] }
+    ;
+
+  tag
+    # <p>foo</p>
+    : tag_start tag_body T_SMALLER T_SLASH T_TEXT T_GREATER
+      {
+        s(:element, val[0], val[1])
+      }
+    ;
+
+  tag_body
+    : T_TEXT
+    ;
+end
+
+---- inner
+
+  def initialize
+    @lexer = Lexer.new
+  end
+
+  def reset
+    @lines  = []
+    @line   = 1
+    @column = 1
+  end
+
+  def s(type, *children)
+    return AST::Node.new(
+      type,
+      children.flatten,
+      :line   => @line,
+      :column => @column
+    )
+  end
+
+  def next_token
+    type, value, line, column = @tokens.shift
+
+    @line   = line if line
+    @column = column if column
+
+    return type ? [type, value] : [false, false]
+  end
+
+  def on_error(type, value, stack)
+    name      = token_to_str(type)
+    line_str  = @lines[@line - 1]
+    indicator = '~' * (@column - 1) + '^'
+
+    raise Racc::ParseError, <<-EOF.strip
+Failed to parse the supplied input.
+
+Reason:   unexpected #{name} with value #{value.inspect}
+Location: line #{@line}, column #{@column}
+
+Offending code:
+
+#{line_str}
+#{indicator}
+
+Current stack:
+
+#{stack.inspect}
+    EOF
+  end
+
+  def parse(string)
+    @lines  = string.lines
+    @tokens = @lexer.lex(string)
+    ast     = do_parse
+
+    reset
+
+    return ast
+  end
+
+# vim: set ft=racc:

task/generate.rake

@@ -1,5 +1,5 @@
 desc 'Generates auto-generated files'
-task :generate => [:lexer]
+task :generate => [:lexer, :parser]
 
 desc 'Regenerates auto-generated files'
 task :regenerate => [:clean, :generate]

task/parser.rake

@@ -1,4 +1,3 @@
-=begin
 rule '.rb' => '.y' do |task|
   Cliver.assert('racc', '~> 1.4')
 
@@ -6,5 +5,4 @@ rule '.rb' => '.y' do |task|
 end
 
 desc 'Generates the parser'
-task :parser => [PARSER_OUTPUT]
-=end
+task :parser => [HTML_PARSER]