Class: Riml::Lexer

Inherits:
Object
  • Object
show all
Includes:
Constants
Defined in:
lib/lexer.rb

Constant Summary collapse

SINGLE_LINE_COMMENT_REGEX = 
/\A[ \t\f]*"(.*)$/
OPERATOR_REGEX = 
/\A#{Regexp.union(['||', '&&', '===', '+=', '-=', '.='] + COMPARISON_OPERATORS)}/
INTERPOLATION_REGEX = 
/"([^"]*?)(\#\{([^"]*?)\})([^"]*?)"/m
ANCHORED_INTERPOLATION_REGEX = 
/\A#{INTERPOLATION_REGEX}/m
INTERPOLATION_SPLIT_REGEX = 
/(\#\{.*?\})/m

Constants included from Constants

Constants::BUILTIN_COMMANDS, Constants::BUILTIN_FUNCTIONS, Constants::COMPARISON_OPERATORS, Constants::DEFINE_KEYWORDS, Constants::END_KEYWORDS, Constants::IGNORECASE_CAPABLE_OPERATORS, Constants::KEYWORDS, Constants::REGISTERS, Constants::RIML_COMMANDS, Constants::RIML_END_KEYWORDS, Constants::RIML_KEYWORDS, Constants::SPECIAL_VARIABLE_PREFIXES, Constants::SPLAT_LITERAL, Constants::VIML_COMMANDS, Constants::VIML_END_KEYWORDS, Constants::VIML_KEYWORDS

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(code) ⇒ Lexer

Returns a new instance of Lexer.



18
19
20
21
22
 
# File 'lib/lexer.rb', line 18

def initialize(code)
  @code = code
  @code.chomp!
  set_start_state!
end

Instance Attribute Details

#chunk ⇒ Object (readonly)

Returns the value of attribute chunk.



14
15
16
 
# File 'lib/lexer.rb', line 14

def chunk
  @chunk
end

#current_indent ⇒ Object (readonly)

Returns the value of attribute current_indent.



14
15
16
 
# File 'lib/lexer.rb', line 14

def current_indent
  @current_indent
end

#ignore_indentation_check ⇒ Object

for REPL



16
17
18
 
# File 'lib/lexer.rb', line 16

def ignore_indentation_check
  @ignore_indentation_check
end

#invalid_keyword ⇒ Object (readonly)

Returns the value of attribute invalid_keyword.



14
15
16
 
# File 'lib/lexer.rb', line 14

def invalid_keyword
  @invalid_keyword
end

#lineno ⇒ Object (readonly)

Returns the value of attribute lineno.



14
15
16
 
# File 'lib/lexer.rb', line 14

def lineno
  @lineno
end

#prev_token ⇒ Object (readonly)

Returns the value of attribute prev_token.



14
15
16
 
# File 'lib/lexer.rb', line 14

def prev_token
  @prev_token
end

#tokens ⇒ Object (readonly)

Returns the value of attribute tokens.



14
15
16
 
# File 'lib/lexer.rb', line 14

def tokens
  @tokens
end

Instance Method Details

#next_token ⇒ Object 



47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
 
# File 'lib/lexer.rb', line 47

def next_token
  while @token_buf.empty? && more_code_to_tokenize?
    tokenize_chunk(get_new_chunk)
  end
  if !@token_buf.empty?
    token = @token_buf.shift
    if token.size == 3
      @lineno += token.pop
    end
    tokens << token
    return @prev_token = token
  end
  check_indentation unless ignore_indentation_check
  nil
end

#prev_token_is_keyword? ⇒ Boolean

Returns:

  • (Boolean) 


236
237
238
239
240
241
242
243
244
245
246
 
# File 'lib/lexer.rb', line 236

def prev_token_is_keyword?
  if prev_token && prev_token[1]
    if KEYWORDS.include?(prev_token[1])
      return @invalid_keyword = prev_token[1]
    end
    prev_prev_token = tokens[-2]
    if prev_prev_token && KEYWORDS.include?(prev_prev_token[1])
      return @invalid_keyword = prev_prev_token[1]
    end
  end
end

#set_start_state! ⇒ Object 



24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
 
# File 'lib/lexer.rb', line 24

def set_start_state!
  # number of characters consumed
  @i = 0
  # array of doubles and triples: [tokenname, tokenval, lineno_to_add(optional)]
  # ex: [[:NEWLINE, "\n"]] OR [[:NEWLINE, "\n", 1]]
  @token_buf = []
  @tokens = []
  @prev_token = nil
  @lineno = 1
  @current_indent = 0
  @indent_pending = false
  @dedent_pending = false
  @one_line_conditional_end_pending = false
  @in_function_declaration = false
  @invalid_keyword = nil
end

#tokenize ⇒ Object 



41
42
43
44
45
 
# File 'lib/lexer.rb', line 41

def tokenize
  set_start_state!
  while next_token != nil; end
  @tokens
end

#tokenize_chunk(chunk) ⇒ Object 



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
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
132
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
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
 
# File 'lib/lexer.rb', line 63

def tokenize_chunk(chunk)
  @chunk = chunk
  # deal with line continuations
  if cont = chunk[/\A\r?\n*[ \t\f]*\\/m]
    @i += cont.size
    @lineno += cont.each_line.to_a.size - 1
    return
  end

  # all lines that start with ':' pass right through unmodified
  if (prev_token.nil? || prev_token[0] == :NEWLINE) && (ex_literal = chunk[/\A[ \t\f]*:(.*)?$/])
    @i += ex_literal.size
    @token_buf << [:EX_LITERAL, $1]
    return
  end

  if splat_var = chunk[/\Aa:\d+/]
    @i += splat_var.size
    @token_buf << [:SCOPE_MODIFIER, 'a:'] << [:IDENTIFIER, splat_var[2..-1]]
  # the 'n' scope modifier is added by riml
  elsif scope_modifier = chunk[/\A([bwtglsavn]:)(\w|{)/, 1]
    @i += 2
    @token_buf << [:SCOPE_MODIFIER, scope_modifier]
  elsif scope_modifier_literal = chunk[/\A([bwtglsavn]:)/]
    @i += scope_modifier_literal.size
    @token_buf << [:SCOPE_MODIFIER_LITERAL, scope_modifier_literal]
  elsif special_var_prefix = chunk[/\A(&(\w:)?(?!&)|\$|@)/]
    @token_buf << [:SPECIAL_VAR_PREFIX, special_var_prefix.strip]
    @i += special_var_prefix.size
    if special_var_prefix == '@'
      new_chunk = get_new_chunk
      next_char = new_chunk[0]
      if REGISTERS.include?(next_char)
        @token_buf << [:IDENTIFIER, next_char]
        @i += 1
      end
    else
      @expecting_identifier = true
    end
  elsif function_method = chunk[/\A(function)\(/, 1]
    @token_buf << [:IDENTIFIER, function_method]
    @i += function_method.size
  elsif identifier = chunk[/\A[a-zA-Z_][\w#]*(\?|!)?/]
    # keyword identifiers
    if KEYWORDS.include?(identifier)
      if identifier.match(/\Afunction/)
        old_identifier = identifier.dup
        identifier.sub!(/function/, "def")
        @i += (old_identifier.size - identifier.size)
      end

      if DEFINE_KEYWORDS.include?(identifier)
        @in_function_declaration = true
      end

      # strip '?' out of token names and replace '!' with '_bang'
      token_name = identifier.sub(/\?\Z/, "").sub(/!\Z/, "_bang").upcase
      track_indent_level(chunk, identifier)

      if VIML_END_KEYWORDS.include?(identifier)
        token_name = :END
      end

      @token_buf << [token_name.intern, identifier]

    elsif BUILTIN_COMMANDS.include?(identifier) && !chunk[/\A#{Regexp.escape(identifier)}\(/]
      @token_buf << [:BUILTIN_COMMAND, identifier]
    elsif RIML_COMMANDS.include? identifier
      @token_buf << [:RIML_COMMAND, identifier]
    elsif VIML_COMMANDS.include?(identifier) && (prev_token.nil? || prev_token[0] == :NEWLINE)
      @i += identifier.size
      new_chunk = get_new_chunk
      until_eol = new_chunk[/.*$/].to_s
      @token_buf << [:EX_LITERAL, identifier << until_eol]
      @i += until_eol.size
      return
    # method names and variable names
    else
      @token_buf << [:IDENTIFIER, identifier]
    end

    @i += identifier.size

    parse_dict_vals!

    if @in_function_declaration
      @in_function_declaration = false unless DEFINE_KEYWORDS.include?(identifier) && @token_buf.size == 1
    end
  elsif splat = chunk[/\A(\.{3}|\*[a-zA-Z_]\w*)/]
    @token_buf << [:SPLAT, splat]
    @i += splat.size
  # integer (octal)
  elsif octal = chunk[/\A0[0-7]+/]
    @token_buf << [:NUMBER, octal]
    @i += octal.size
  # integer (hex)
  elsif hex = chunk[/\A0[xX]\h+/]
    @token_buf << [:NUMBER, hex]
    @i += hex.size
  # integer or float (decimal)
  elsif decimal = chunk[/\A[0-9]+(\.[0-9]+([eE][+-]?[0-9]+)?)?/]
    @token_buf << [:NUMBER, decimal]
    @i += decimal.size
  elsif interpolation = chunk[ANCHORED_INTERPOLATION_REGEX]
    # "hey there, #{name}" = "hey there, " . name
    parts = interpolation[1...-1].split(INTERPOLATION_SPLIT_REGEX)
    handle_interpolation(*parts)
    @i += interpolation.size
  elsif (single_line_comment = chunk[SINGLE_LINE_COMMENT_REGEX]) && (prev_token.nil? || prev_token[0] == :NEWLINE)
    @i += single_line_comment.size + 1 # consume next newline character
    @lineno += single_line_comment.each_line.to_a.size
  elsif inline_comment = chunk[/\A[ \t\f]*"[^"]*?$/]
    @i += inline_comment.size # inline comment, don't consume newline character
    @lineno += inline_comment.each_line.to_a.size - 1
  elsif string_double = chunk[/\A"(.*?)(?<!\\)"/, 1]
    @token_buf << [:STRING_D, string_double]
    @i += string_double.size + 2
  elsif string_single = chunk[/\A'(([^']|'')*)'/, 1]
    @token_buf << [:STRING_S, string_single]
    @i += string_single.size + 2
  elsif newlines = chunk[/\A([\r\n]+)/, 1]
    # push only 1 newline
    @token_buf << [:NEWLINE, "\n"] unless prev_token && prev_token[0] == :NEWLINE

    # pending indents/dedents
    if @one_line_conditional_end_pending
      @one_line_conditional_end_pending = false
    elsif @indent_pending
      @indent_pending = false
    elsif @dedent_pending
      @dedent_pending = false
    end

    @i += newlines.size
    @lineno += newlines.size
  elsif heredoc_pattern = chunk[%r{\A<<(.+?)\r?\n}]
    pattern = $1
    @i += heredoc_pattern.size
    new_chunk = get_new_chunk
    heredoc_string = new_chunk[%r|(.+?\r?\n)(#{Regexp.escape(pattern)})|m, 1]
    @i += heredoc_string.size + pattern.size
    heredoc_string.chomp!
    if heredoc_string =~ INTERPOLATION_REGEX || %Q("#{heredoc_string}") =~ INTERPOLATION_REGEX
      parts = heredoc_string.split(INTERPOLATION_SPLIT_REGEX)
      handle_interpolation(*parts)
    else
      @token_buf << [:STRING_D, escape_chars!(heredoc_string)]
    end
    @lineno += heredoc_string.each_line.to_a.size
  # operators of more than 1 char
  elsif operator = chunk[OPERATOR_REGEX]
    @token_buf << [operator, operator]
    @i += operator.size
  elsif regexp = chunk[%r{\A/.*?[^\\]/}]
    @token_buf << [:REGEXP, regexp]
    @i += regexp.size
  elsif whitespaces = chunk[/\A[ \t\f]+/]
    @i += whitespaces.size
  # operators and tokens of single chars, one of: ( ) , . [ ] ! + - = < > /
  else
    value = chunk[0, 1]
    if value == '|'
      @token_buf << [:NEWLINE, "\n"]
    else
      @token_buf << [value, value]
    end
    @i += 1
    if value == ']' || value == ')' && (chunk[1, 1] == '.' && chunk[3, 1] != ':')
      parse_dict_vals!
    end
  end
end