Class: Bade::Parser

Inherits:
Object show all
Defined in:
lib/bade/parser.rb,
lib/bade/parser/parser_tag.rb,
lib/bade/parser/parser_text.rb,
lib/bade/parser/parser_lines.rb,
lib/bade/parser/parser_mixin.rb,
lib/bade/parser/parser_constants.rb,
lib/bade/parser/parser_ruby_code.rb

Overview

Class to parse input string into AST::Document

Defined Under Namespace

Modules: LineIndicatorRegexps, MixinRegexps, ParseRubyCodeRegexps, TagRegexps, TextRegexps Classes: ParserInternalError, SyntaxError

Constant Summary collapse

WORD_RE = 
''.respond_to?(:encoding) ? '\p{Word}' : '\w'
NAME_RE_STRING = 
"(#{WORD_RE}(?:#{WORD_RE}|:|-|_)*)".freeze
ATTR_NAME_RE_STRING = 
"\\A\\s*#{NAME_RE_STRING}".freeze
CODE_ATTR_RE = 
/#{ATTR_NAME_RE_STRING}\s*&?:\s*/.freeze
TAG_RE = 
/\A#{NAME_RE_STRING}/.freeze
CLASS_TAG_RE = 
/\A\.#{NAME_RE_STRING}/.freeze
ID_TAG_RE = 
/\A##{NAME_RE_STRING}/.freeze
RUBY_DELIMITERS_REVERSE = 
{
  '(' => ')',
  '[' => ']',
  '{' => '}',
}.freeze
RUBY_QUOTES = 
%w[' "].freeze
RUBY_NOT_NESTABLE_DELIMITERS = 
RUBY_QUOTES
RUBY_START_DELIMITERS = 
(%w(\( [ {) + RUBY_NOT_NESTABLE_DELIMITERS).freeze
RUBY_END_DELIMITERS = 
(%w(\) ] }) + RUBY_NOT_NESTABLE_DELIMITERS).freeze
RUBY_ALL_DELIMITERS = 
(RUBY_START_DELIMITERS + RUBY_END_DELIMITERS).uniq.freeze
RUBY_START_DELIMITERS_RE = 
/\A[#{Regexp.escape RUBY_START_DELIMITERS.join}]/.freeze
RUBY_END_DELIMITERS_RE = 
/\A[#{Regexp.escape RUBY_END_DELIMITERS.join}]/.freeze

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(tabsize: 4, file_path: nil) ⇒ Parser

Returns a new instance of Parser.

Parameters:

  • tabsize (Fixnum) (defaults to: 4)
  • file_path (String) (defaults to: nil) 


58
59
60
61
62
63
64
65
66
67
68
 
# File 'lib/bade/parser.rb', line 58

def initialize(tabsize: 4, file_path: nil)
  @line = ''

  @tabsize = tabsize
  @file_path = file_path

  @tab_re = /\G((?: {#{tabsize}})*) {0,#{tabsize - 1}}\t/
  @tab = "\1#{' ' * tabsize}"

  reset
end

Instance Attribute Details

#dependency_paths ⇒ Array<String> (readonly)

Returns:



49
50
51
 
# File 'lib/bade/parser.rb', line 49

def dependency_paths
  @dependency_paths
end

#file_path ⇒ String (readonly)

Returns:



53
54
55
 
# File 'lib/bade/parser.rb', line 53

def file_path
  @file_path
end

Instance Method Details

#append_node(type, indent: @indents.length, add: false, value: nil) ⇒ Object

Append element to stacks and result tree

Parameters:

  • type (Symbol) 


106
107
108
109
110
111
112
113
114
115
116
117
118
119
 
# File 'lib/bade/parser.rb', line 106

def append_node(type, indent: @indents.length, add: false, value: nil)
  # add necessary stack items to match required indent
  @stacks << @stacks.last.dup while indent >= @stacks.length

  parent = @stacks[indent].last
  node = AST::NodeRegistrator.create(type, parent, lineno: @lineno, filename: @file_path)
  parent.children << node

  node.value = value unless value.nil?

  @stacks[indent] << node if add

  node
end

#fixed_trailing_colon(value) ⇒ Object

Parameters:



141
142
143
144
145
146
147
148
 
# File 'lib/bade/parser.rb', line 141

def fixed_trailing_colon(value)
  if value.is_a?(String) && value.end_with?(':')
    value = value.remove_last
    @line.prepend(':')
  end

  value
end

#get_indent(line) ⇒ Int

Calculate indent for line

Parameters:

Returns:

  • (Int) —

    indent size



98
99
100
 
# File 'lib/bade/parser.rb', line 98

def get_indent(line)
  line.get_indent(@tabsize)
end

#next_line ⇒ Object 



54
55
56
57
58
59
60
61
62
63
64
65
66
67
 
# File 'lib/bade/parser/parser_lines.rb', line 54

def next_line
  if @lines.empty?
    @orig_line = @line = nil

    last_newlines = remove_last_newlines
    @root.children += last_newlines

    nil
  else
    @orig_line = @lines.shift
    @lineno += 1
    @line = @orig_line.dup
  end
end

#parse(str) ⇒ Bade::AST::Document

Returns root node.

Parameters:

Returns:



73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
 
# File 'lib/bade/parser.rb', line 73

def parse(str)
  @document = AST::Document.new(file_path: file_path)
  @root = @document.root

  @dependency_paths = []

  if str.is_a?(Array)
    reset(str, [[@root]])
  else
    reset(str.split(/\r?\n/, -1), [[@root]]) # -1 is for not suppressing empty lines
  end

  parse_line while next_line

  reset

  @document
end

#parse_import ⇒ Object 



129
130
131
132
133
134
135
136
137
 
# File 'lib/bade/parser.rb', line 129

def parse_import
  # TODO: change this to something better
  # rubocop:disable Security/Eval
  path = eval(@line)
  # rubocop:enable Security/Eval
  append_node(:import, value: path)

  @dependency_paths << path unless @dependency_paths.include?(path)
end

#parse_line ⇒ Object 



69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
 
# File 'lib/bade/parser/parser_lines.rb', line 69

def parse_line
  if @line.strip.empty?
    append_node(:newline) unless @lines.empty?
    return
  end

  indent = get_indent(@line)

  # left strip
  @line.remove_indent!(indent, @tabsize)

  # If there's more stacks than indents, it means that the previous
  # line is expecting this line to be indented.
  expecting_indentation = @stacks.length > @indents.length

  if indent > @indents.last
    @indents << indent
  else
    # This line was *not* indented more than the line before,
    # so we'll just forget about the stack that the previous line pushed.
    if expecting_indentation
      last_newlines = remove_last_newlines

      @stacks.pop

      new_node = @stacks.last.last
      new_node.children += last_newlines
    end

    # This line was deindented.
    # Now we're have to go through the all the indents and figure out
    # how many levels we've deindented.
    while indent < @indents.last
      last_newlines = remove_last_newlines

      @indents.pop
      @stacks.pop

      new_node = @stacks.last.last
      new_node.children += last_newlines
    end

    # Remove old stacks we don't need
    while !@stacks[indent].nil? && indent < @stacks[indent].length - 1
      last_newlines = remove_last_newlines

      @stacks[indent].pop

      new_node = @stacks.last.last
      new_node.children += last_newlines
    end

    # This line's indentation happens lie "between" two other line's
    # indentation:
    #
    #   hello
    #       world
    #     this      # <- This should not be possible!
    syntax_error('Malformed indentation') if indent != @indents.last
  end

  parse_line_indicators
end

#parse_line_indicators(add_newline: true) ⇒ Object 



133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
 
# File 'lib/bade/parser/parser_lines.rb', line 133

def parse_line_indicators(add_newline: true)
  case @line
  when LineIndicatorRegexps::IMPORT
    @line = $'
    parse_import

  when LineIndicatorRegexps::YIELD
    @line = $'
    node = append_node(:yield, add: true)
    node.conditional = $1.nil?

  when LineIndicatorRegexps::MIXIN_DECL
    # Mixin declaration
    @line = $'
    parse_mixin_declaration($1)

  when LineIndicatorRegexps::MIXIN_CALL
    # Mixin call
    @line = $'
    parse_mixin_call($1)

  when LineIndicatorRegexps::BLOCK_DECLARATION
    @line = $'
    if @stacks.last.last.type == :mixin_call
      node = append_node(:mixin_block, add: true)
      node.name = $1
    else
      # keyword block used outside of mixin call
      parse_tag($&)
    end

  when LineIndicatorRegexps::HTML_COMMENT
    # HTML comment
    append_node(:html_comment, add: true)
    parse_text_block $', @indents.last + @tabsize

  when LineIndicatorRegexps::NORMAL_COMMENT
    # Comment
    append_node(:comment, add: true)
    parse_text_block $', @indents.last + @tabsize

  when LineIndicatorRegexps::TEXT_BLOCK_START
    # Found a text block.
    parse_text_block $', @indents.last + @tabsize

  when LineIndicatorRegexps::INLINE_HTML
    # Inline html
    parse_text

  when LineIndicatorRegexps::CODE_BLOCK
    # Found a code block.
    append_node(:code, value: $'.strip)

  when LineIndicatorRegexps::OUTPUT_BLOCK
    # Found an output block.
    # We expect the line to be broken or the next line to be indented.
    @line = $'
    output_node = append_node(:output)
    output_node.conditional = $1.length == 1
    output_node.escaped = $2.length == 1
    output_node.value = parse_ruby_code(ParseRubyCodeRegexps::END_NEW_LINE)

  when LineIndicatorRegexps::DOCTYPE
    # Found doctype declaration
    append_node(:doctype, value: $'.strip)

  when TAG_RE
    # Found a HTML tag.
    @line = $' if $1
    parse_tag($&)

  when LineIndicatorRegexps::TAG_CLASS_START_BLOCK, LineIndicatorRegexps::TAG_ID_START_BLOCK
    # Found a class or id selector.
    parse_tag 'div'

  else
    syntax_error 'Unknown line indicator'
  end

  append_node(:newline) if add_newline && !@lines.empty?
end

#parse_mixin_call(mixin_name) ⇒ Object 



24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
 
# File 'lib/bade/parser/parser_mixin.rb', line 24

def parse_mixin_call(mixin_name)
  mixin_name = fixed_trailing_colon(mixin_name)

  mixin_node = append_node(:mixin_call, add: true)
  mixin_node.name = mixin_name

  parse_mixin_call_params

  case @line
  when MixinRegexps::TEXT_START
    @line = $'
    parse_text

  when MixinRegexps::BLOCK_EXPANSION
    # Block expansion
    @line = $'
    parse_line_indicators(add_newline: false)

  when MixinRegexps::OUTPUT_CODE
    # Handle output code
    parse_line_indicators(add_newline: false)

  when ''
    # nothing

  else
    syntax_error "Unknown symbol after mixin calling, line = `#{@line}'"
  end
end

#parse_mixin_call_params ⇒ Object 



54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
 
# File 'lib/bade/parser/parser_mixin.rb', line 54

def parse_mixin_call_params
  # between tag name and attribute must not be space
  # and skip when is nothing other
  return unless @line.start_with?('(')

  # remove starting bracket
  @line.remove_first!

  loop do
    case @line
    when MixinRegexps::PARAMS_KEY_PARAM_NAME
      @line = $'
      attr_node = append_node(:mixin_key_param)
      attr_node.name = fixed_trailing_colon($1)
      attr_node.value = parse_ruby_code(ParseRubyCodeRegexps::END_PARAMS_ARG, allow_multiline: true)

    when MixinRegexps::PARAMS_ARGS_DELIMITER
      # args delimiter
      @line = $'
      next

    when MixinRegexps::PARAMS_END_SPACES
      # spaces and/or end of line
      next_line
      next

    when MixinRegexps::PARAMS_END
      # Find ending delimiter
      @line = $'
      break

    else
      attr_node = append_node(:mixin_param)
      attr_node.value = parse_ruby_code(ParseRubyCodeRegexps::END_PARAMS_ARG, allow_multiline: true)
    end
  end
end

#parse_mixin_declaration(mixin_name) ⇒ Object 



92
93
94
95
96
97
 
# File 'lib/bade/parser/parser_mixin.rb', line 92

def parse_mixin_declaration(mixin_name)
  mixin_node = append_node(:mixin_decl, add: true)
  mixin_node.name = mixin_name

  parse_mixin_declaration_params
end

#parse_mixin_declaration_params ⇒ Object 



99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
 
# File 'lib/bade/parser/parser_mixin.rb', line 99

def parse_mixin_declaration_params
  # between tag name and attribute must not be space
  # and skip when is nothing other
  return unless @line.start_with?('(')

  # remove starting bracket
  @line.remove_first!

  loop do
    case @line
    when MixinRegexps::PARAMS_KEY_PARAM_NAME
      # Value ruby code
      @line = $'
      attr_node = append_node(:mixin_key_param)
      attr_node.name = fixed_trailing_colon($1)
      attr_node.value = parse_ruby_code(ParseRubyCodeRegexps::END_PARAMS_ARG)

    when MixinRegexps::PARAMS_PARAM_NAME
      @line = $'
      attr_node = append_node(:mixin_param, value: $1)

      if @line =~ MixinRegexps::PARAMS_PARAM_DEFAULT_START
        @line = $'
        attr_node.default_value = parse_ruby_code(ParseRubyCodeRegexps::END_PARAMS_ARG)
      end

    when MixinRegexps::PARAMS_BLOCK_NAME
      @line = $'
      append_node(:mixin_block_param, value: $1)

    when MixinRegexps::PARAMS_ARGS_DELIMITER
      # args delimiter
      @line = $'
      next

    when MixinRegexps::PARAMS_END
      # Find ending delimiter
      @line = $'
      break

    else
      syntax_error('wrong mixin attribute syntax')
    end
  end
end

#parse_ruby_code(outer_delimiters, allow_multiline: false) ⇒ Void

Parse ruby code, ended with outer delimiters

Parameters:

  • outer_delimiters (String, Regexp)

Returns:

  • (Void) —

    parsed ruby code



19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
 
# File 'lib/bade/parser/parser_ruby_code.rb', line 19

def parse_ruby_code(outer_delimiters, allow_multiline: false)
  code = String.new
  end_re = if outer_delimiters.is_a?(Regexp)
             outer_delimiters
           else
             /\A\s*[#{Regexp.escape outer_delimiters.to_s}]/
           end

  delimiters = []
  string_start_quote_char = nil

  loop do
    break if !allow_multiline && @line.empty?
    break if allow_multiline && @line.empty? && (@lines && @lines.empty?)
    break if delimiters.empty? && @line =~ end_re

    if @line.empty? && allow_multiline && !(@lines && @lines.empty?)
      next_line
      code << "\n"
    end

    char = @line[0]

    # backslash escaped delimiter
    if char == '\\' && RUBY_ALL_DELIMITERS.include?(@line[1])
      code << @line.slice!(0, 2)
      next
    end

    case char
    when RUBY_START_DELIMITERS_RE
      if RUBY_NOT_NESTABLE_DELIMITERS.include?(char) && delimiters.last == char
        # end char of not nestable delimiter
        delimiters.pop
        string_start_quote_char = nil
      else
        # diving into nestable delimiters
        delimiters << char if string_start_quote_char.nil?

        # mark start char of the not nestable delimiters, for example strings
        if RUBY_NOT_NESTABLE_DELIMITERS.include?(char) && string_start_quote_char.nil?
          string_start_quote_char = char
        end
      end

    when RUBY_END_DELIMITERS_RE
      # rising
      delimiters.pop if char == RUBY_DELIMITERS_REVERSE[delimiters.last]
    end

    code << @line.slice!(0)
  end

  syntax_error('Unexpected end of ruby code') unless delimiters.empty?

  code.strip
end

#parse_tag(tag) ⇒ Object

Parameters:



19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
 
# File 'lib/bade/parser/parser_tag.rb', line 19

def parse_tag(tag)
  tag = fixed_trailing_colon(tag)

  if tag.is_a?(AST::Node)
    tag_node = tag
  else
    tag_node = append_node(:tag, add: true)
    tag_node.name = tag
  end

  parse_tag_attributes

  case @line
  when TagRegexps::BLOCK_EXPANSION
    # Block expansion
    @line = $'
    parse_line_indicators(add_newline: false)

  when TagRegexps::OUTPUT_CODE
    # Handle output code
    parse_line_indicators(add_newline: false)

  when CLASS_TAG_RE
    # Class name
    @line = $'

    attr_node = append_node(:tag_attr)
    attr_node.name = 'class'
    attr_node.value = fixed_trailing_colon($1).single_quote

    parse_tag tag_node

  when ID_TAG_RE
    # Id name
    @line = $'

    attr_node = append_node(:tag_attr)
    attr_node.name = 'id'
    attr_node.value = fixed_trailing_colon($1).single_quote

    parse_tag tag_node

  when TagRegexps::TEXT_START
    # Text content
    @line = $'
    parse_text

  when ''
    # nothing

  else
    syntax_error "Unknown symbol after tag definition #{@line}"
  end
end

#parse_tag_attributes ⇒ Object 



74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
 
# File 'lib/bade/parser/parser_tag.rb', line 74

def parse_tag_attributes
  # Check to see if there is a delimiter right after the tag name

  # between tag name and attribute must not be space
  # and skip when is nothing other
  return unless @line.start_with?('(')

  # remove starting bracket
  @line.remove_first!

  loop do
    case @line
    when CODE_ATTR_RE
      # Value ruby code
      @line = $'
      attr_node = append_node(:tag_attr)
      attr_node.name = $1
      attr_node.value = parse_ruby_code(ParseRubyCodeRegexps::END_PARAMS_ARG)

    when TagRegexps::PARAMS_ARGS_DELIMITER
      # args delimiter
      @line = $'
      next

    when TagRegexps::PARAMS_END
      # Find ending delimiter
      @line = $'
      break

    else
      # Found something where an attribute should be
      @line.lstrip!
      syntax_error('Expected attribute') unless @line.empty?

      # Attributes span multiple lines
      append_node(:newline)
      syntax_error('Expected closing tag attributes delimiter `)`') if @lines.empty?
      next_line
    end
  end
end

#parse_text ⇒ Object 



13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
 
# File 'lib/bade/parser/parser_text.rb', line 13

def parse_text
  new_index = @line.index(TextRegexps::INTERPOLATION_START)

  # the interpolation sequence is not in text, mark whole text as static
  if new_index.nil?
    append_node(:static_text, value: @line)
    return
  end

  unparsed_part = String.new

  while (new_index = @line.index(TextRegexps::INTERPOLATION_START))
    if $1.nil?
      static_part = unparsed_part + @line.remove_first!(new_index)
      append_node(:static_text, value: static_part)

      @line.remove_first!(2) # #{ or &{

      dynamic_part = parse_ruby_code(TextRegexps::INTERPOLATION_END)
      node = append_node(:output, value: dynamic_part)
      node.escaped = $2 == '&'

      @line.remove_first! # ending }

      unparsed_part = String.new
    else
      unparsed_part << @line.remove_first!(new_index)
      @line.remove_first! # symbol \
      unparsed_part << @line.remove_first!(2) # #{ or &{
    end
  end

  # add the rest of line
  append_node(:static_text, value: unparsed_part + @line) unless @line.empty?
end

#parse_text_block(first_line, text_indent = nil) ⇒ Object 



49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
 
# File 'lib/bade/parser/parser_text.rb', line 49

def parse_text_block(first_line, text_indent = nil)
  if !first_line || first_line.empty?
    text_indent = nil
  else
    @line = first_line
    parse_text
  end

  until @lines.empty?
    if @lines.first.blank?
      next_line
      append_node(:newline)
    else
      indent = get_indent(@lines.first)
      break if indent <= @indents.last

      next_line

      @line.remove_indent!(text_indent || indent, @tabsize)

      parse_text

      # The indentation of first line of the text block
      # determines the text base indentation.
      text_indent ||= indent
    end
  end
end

#remove_last_newlines ⇒ Array<AST::Node>

Returns:



123
124
125
126
127
 
# File 'lib/bade/parser.rb', line 123

def remove_last_newlines
  last_node = @stacks.last.last
  last_newlines_count = last_node.children.rcount_matching { |n| n.type == :newline }
  last_node.children.pop(last_newlines_count)
end

#reset(lines = nil, stacks = nil) ⇒ Object 



25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
 
# File 'lib/bade/parser/parser_lines.rb', line 25

def reset(lines = nil, stacks = nil)
  # Since you can indent however you like in Slim, we need to keep a list
  # of how deeply indented you are. For instance, in a template like this:
  #
  #   doctype       # 0 spaces
  #   html          # 0 spaces
  #    head         # 1 space
  #       title     # 4 spaces
  #
  # indents will then contain [0, 1, 4] (when it's processing the last line.)
  #
  # We uses this information to figure out how many steps we must "jump"
  # out when we see an de-indented line.
  @indents = [0]

  # Whenever we want to output something, we'll *always* output it to the
  # last stack in this array. So when there's a line that expects
  # indentation, we simply push a new stack onto this array. When it
  # processes the next line, the content will then be outputted into that
  # stack.
  @stacks = stacks

  @lineno = 0
  @lines = lines

  # @return [String]
  @line = @orig_line = nil
end

#syntax_error(message) ⇒ Object

Raise specific error

Parameters:

Raises:



156
157
158
159
 
# File 'lib/bade/parser.rb', line 156

def syntax_error(message)
  column = @orig_line && @line ? @orig_line.size - @line.size : 0
  raise SyntaxError.new(message, file_path, @orig_line, @lineno, column)
end