Class: Riml::Lexer

Inherits:
Object
  • Object
show all
Includes:
Constants
Defined in:
lib/riml/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::COMPILED_STRING_LOCATION, Constants::DEFINE_KEYWORDS, Constants::END_KEYWORDS, Constants::IGNORECASE_CAPABLE_OPERATORS, Constants::KEYWORDS, Constants::REGISTERS, Constants::RIML_CLASS_COMMANDS, Constants::RIML_COMMANDS, Constants::RIML_END_KEYWORDS, Constants::RIML_FILE_COMMANDS, Constants::RIML_KEYWORDS, Constants::SPECIAL_VARIABLE_PREFIXES, Constants::SPLAT_LITERAL, Constants::UNKNOWN_LOCATION_INFO, Constants::VIML_COMMANDS, Constants::VIML_END_KEYWORDS, Constants::VIML_KEYWORDS

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(code, filename = nil, parser_info = false) ⇒ Lexer

Returns a new instance of Lexer.



23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
 
# File 'lib/riml/lexer.rb', line 23

def initialize(code, filename = nil, parser_info = false)
  code.chomp!
  @s = StringScanner.new(code)
  @filename = filename || COMPILED_STRING_LOCATION
  @parser_info = parser_info
  # array of doubles and triples: [tokenname, tokenval, lineno_to_add(optional)]
  # ex: [[:NEWLINE, "\n"]] OR [[:NEWLINE, "\n", 1]]
  @token_buf = []
  # array of doubles OR triples, depending if `@parser_info` is set to true
  # doubles: [tokenname, tokenval]
  # ex: [[:NEWLINE, "\n"], ...]
  # triples: [tokenname, tokenval, parser_info]
  # ex: [[:NEWLINE, "\n", { :lineno => 1, :filename => 'main.riml' }], ...]
  @tokens = []
  @prev_token = nil
  @lineno = 1
  @current_indent = 0
  @indent_pending = false
  @dedent_pending = false
  @in_function_declaration = false
end

Instance Attribute Details

#current_indent ⇒ Object (readonly)

Returns the value of attribute current_indent.



17
18
19
 
# File 'lib/riml/lexer.rb', line 17

def current_indent
  @current_indent
end

#filename ⇒ Object (readonly)

Returns the value of attribute filename.



17
18
19
 
# File 'lib/riml/lexer.rb', line 17

def filename
  @filename
end

#ignore_indentation_check ⇒ Object

for REPL



21
22
23
 
# File 'lib/riml/lexer.rb', line 21

def ignore_indentation_check
  @ignore_indentation_check
end

#lineno ⇒ Object

Returns the value of attribute lineno.



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

def lineno
  @lineno
end

#parser_info ⇒ Object (readonly)

Returns the value of attribute parser_info.



17
18
19
 
# File 'lib/riml/lexer.rb', line 17

def parser_info
  @parser_info
end

#prev_token ⇒ Object (readonly)

Returns the value of attribute prev_token.



17
18
19
 
# File 'lib/riml/lexer.rb', line 17

def prev_token
  @prev_token
end

#tokens ⇒ Object (readonly)

Returns the value of attribute tokens.



17
18
19
 
# File 'lib/riml/lexer.rb', line 17

def tokens
  @tokens
end

Instance Method Details

#next_token ⇒ Object

TODO: fix this slow method



51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
 
# File 'lib/riml/lexer.rb', line 51

def next_token
  while (buf_empty = @token_buf.empty?) && !@s.eos?
    tokenize_chunk
  end
  if !buf_empty
    token = @token_buf.shift
    if token.size == 3
      @lineno += token.pop
    end
    if @parser_info
      @tokens << decorate_token(token)
      @prev_token = token.first(2)
      return token
    else
      @tokens << token
      return @prev_token = token
    end
  end
  check_indentation unless ignore_indentation_check
  nil
end

#prev_token_is_keyword?(n = 2) ⇒ Boolean

Checks if any of previous n tokens are keywords. If any found, return the keyword, otherwise returns ‘false`.

Returns:

  • (Boolean) 


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

def prev_token_is_keyword?(n = 2)
  return false if n <= 0
  (1..n).each do |i|
    t = tokens[-i]
    if t && t[1] && KEYWORDS.include?(t[1])
      return t[1]
    end
  end
  false
end

#tokenize ⇒ Object 



45
46
47
48
 
# File 'lib/riml/lexer.rb', line 45

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

#tokenize_chunk ⇒ Object 



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
 
# File 'lib/riml/lexer.rb', line 73

def tokenize_chunk
  # deal with line continuations
  if cont = @s.scan(/\A\r?\n*[ \t\f]*\\/m)
    @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) && @s.scan(/\A[ \t\f]*:(.*)?$/)
    @token_buf << [:EX_LITERAL, @s[1]]
    return
  end

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

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

      @token_buf << [token_name.to_sym, identifier]

    elsif BUILTIN_COMMANDS.include?(identifier) && @s.peek(identifier.size + 1)[-1, 1] != '('
      @token_buf << [:BUILTIN_COMMAND, identifier]
    elsif RIML_FILE_COMMANDS.include? identifier
      @token_buf << [:RIML_FILE_COMMAND, identifier]
    elsif RIML_CLASS_COMMANDS.include? identifier
      @token_buf << [:RIML_CLASS_COMMAND, identifier]
    elsif VIML_COMMANDS.include?(identifier) && (prev_token.nil? || prev_token[0] == :NEWLINE)
      @s.pos += identifier.size
      until_eol = @s.scan(/.*$/).to_s
      @token_buf << [:EX_LITERAL, identifier << until_eol]
      return
    # method names and variable names
    else
      @token_buf << [:IDENTIFIER, identifier]
    end

    @s.pos += identifier.size

    parse_dict_vals!

  elsif @in_function_declaration && (splat_param = @s.scan(/\A(\.{3}|\*[a-zA-Z_]\w*)/))
    @token_buf << [:SPLAT_PARAM, splat_param]
  # splat in calling context. ex: super(*args) or super(*(args + other_args)) or func('hey', *args)
  elsif !@in_function_declaration && prev_token && @s.check(/\A\*(\w+|\()/)
    @token_buf << [:SPLAT_ARG, @s.getch]
  # integer (octal)
  elsif octal = @s.scan(/\A0[0-7]+/)
    @token_buf << [:NUMBER, octal]
  # integer (hex)
  elsif hex = @s.scan(/\A0[xX][0-9a-fA-F]+/)
    @token_buf << [:NUMBER, hex]
  # integer or float (decimal)
  elsif decimal = @s.scan(/\A[0-9]+(\.[0-9]+([eE][+-]?[0-9]+)?)?/)
    @token_buf << [:NUMBER, decimal]
  elsif interpolation = @s.scan(ANCHORED_INTERPOLATION_REGEX)
    # "hey there, #{name}" = "hey there, " . name
    parts = interpolation[1...-1].split(INTERPOLATION_SPLIT_REGEX)
    handle_interpolation(*parts)
  elsif (single_line_comment = @s.check(SINGLE_LINE_COMMENT_REGEX)) && (prev_token.nil? || prev_token[0] == :NEWLINE)
    @s.pos += single_line_comment.size
    @s.pos += 1 unless @s.eos? # consume newline
    @lineno += single_line_comment.each_line.to_a.size
  elsif inline_comment = @s.scan(/\A[ \t\f]*"[^"]*?$/)
    @lineno += inline_comment.each_line.to_a.size - 1
  elsif (str = lex_string_double)
    @token_buf << [:STRING_D, str]
  elsif @s.scan(/\A'(([^']|'')*)'/)
    @token_buf << [:STRING_S, @s[1]]
  elsif newlines = @s.scan(/\A([\r\n]+)/)
    # push only 1 newline
    @token_buf << [:NEWLINE, "\n"] unless prev_token && prev_token[0] == :NEWLINE

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

    @lineno += newlines.size
  # heredoc
  elsif @s.scan(%r{\A<<(.+?)\r?\n})
    pattern = @s[1]
    @s.check(%r|(.+?\r?\n)(#{Regexp.escape(pattern)})|m)
    heredoc_string = @s[1]
    @s.pos += (pattern.size + heredoc_string.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 = @s.scan(OPERATOR_REGEX)
    @token_buf << [operator, operator]
  elsif regexp = @s.scan(%r{\A/.*?[^\\]/})
    @token_buf << [:REGEXP, regexp]
  # whitespaces
  elsif @s.scan(/\A[ \t\f]+/)
  # operators and tokens of single chars, one of: ( ) , . [ ] ! + - = < > /
  else
    value = @s.getch
    if value == '|'
      @token_buf << [:NEWLINE, "\n"]
    else
      @token_buf << [value, value]
    end
    # if we encounter `funcCall().`, the next character must be treated as
    # a dictionary retrieval operation, not a string concatenation
    # operation.
    # However, if we see `funcCall().l:localVar`, we know it must be a
    # string concatenation operation.
    if value == ']' || value == ')' && (@s.peek(1) == '.' && @s.peek(3) != ':')
      parse_dict_vals!
    end
  end
end