Class: NScript::Rewriter

Inherits:
Object
  • Object
show all
Defined in:
lib/nscript/lexer/rewriter.rb

Constant Summary collapse

BALANCED_PAIRS = 
[['(', ')'], ['[', ']'], ['{', '}'], [:INDENT, :OUTDENT],
[:PARAM_START, :PARAM_END], [:CALL_START, :CALL_END], [:INDEX_START, :INDEX_END]]
EXPRESSION_START = 
BALANCED_PAIRS.map {|pair| pair.first }
EXPRESSION_TAIL = 
BALANCED_PAIRS.map {|pair| pair.last }
EXPRESSION_CLOSE = 
[:CATCH, :WHEN, :ELSE, :FINALLY] + EXPRESSION_TAIL
IMPLICIT_FUNC = 
[:IDENTIFIER, :SUPER, ')', :CALL_END, ']', :INDEX_END]
IMPLICIT_END = 
[:IF, :UNLESS, :FOR, :WHILE, "\n", :OUTDENT]
IMPLICIT_CALL = 
[:IDENTIFIER, :NUMBER, :STRING, :JS, :REGEX, :NEW, :PARAM_START,
:TRY, :DELETE, :TYPEOF, :SWITCH,
:TRUE, :FALSE, :YES, :NO, :ON, :OFF, '!', '!!', :NOT,
'@', '->', '=>', '[', '(', '{']
INVERSES = 
BALANCED_PAIRS.inject({}) do |memo, pair|
  memo[pair.first] = pair.last
  memo[pair.last]  = pair.first
  memo
end
SINGLE_LINERS = 
[:ELSE, "->", "=>", :TRY, :FINALLY, :THEN]
SINGLE_CLOSERS = 
["\n", :CATCH, :FINALLY, :ELSE, :OUTDENT, :LEADING_WHEN, :PARAM_START]

Instance Method Summary collapse

Instance Method Details

#add_implicit_indentation ⇒ Object 



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
 
# File 'lib/nscript/lexer/rewriter.rb', line 145

def add_implicit_indentation
  scan_tokens do |prev, token, post, i|
    next 1 unless SINGLE_LINERS.include?(token[0]) && post[0] != :INDENT &&
      !(token[0] == :ELSE && post[0] == :IF) # Elsifs shouldn't get blocks.
    starter = token[0]
    line = token[1].line
    @tokens.insert(i + 1, [:INDENT, Value.new(2, line)])
    idx = i + 1
    parens = 0
    loop do
      idx += 1
      tok = @tokens[idx]
      if (!tok || SINGLE_CLOSERS.include?(tok[0]) ||
          (tok[0] == ')' && parens == 0)) &&
          !(starter == :ELSE && tok[0] == :ELSE)
        insertion = @tokens[idx - 1][0] == "," ? idx - 1 : idx
        @tokens.insert(insertion, [:OUTDENT, Value.new(2, line)])
        break
      end
      parens += 1 if tok[0] == '('
      parens -= 1 if tok[0] == ')'
    end
    next 1 unless token[0] == :THEN
    @tokens.delete_at(i)
    next 0
  end
end

#add_implicit_parentheses ⇒ Object 



124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
 
# File 'lib/nscript/lexer/rewriter.rb', line 124

def add_implicit_parentheses
  stack = [0]
  scan_tokens do |prev, token, post, i|
    stack.push(0) if token[0] == :INDENT
    if token[0] == :OUTDENT
      last = stack.pop
      stack[-1] += last
    end
    if stack.last > 0 && (IMPLICIT_END.include?(token[0]) || post.nil?)
      idx = token[0] == :OUTDENT ? i + 1 : i
      stack.last.times { @tokens.insert(idx, [:CALL_END, Value.new(')', token[1].line)]) }
      size, stack[-1] = stack[-1] + 1, 0
      next size
    end
    next 1 unless IMPLICIT_FUNC.include?(prev[0]) && IMPLICIT_CALL.include?(token[0])
    @tokens.insert(i, [:CALL_START, Value.new('(', token[1].line)])
    stack[-1] += 1
    next 2
  end
end

#adjust_comments ⇒ Object 



51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
 
# File 'lib/nscript/lexer/rewriter.rb', line 51

def adjust_comments
  scan_tokens do |prev, token, post, i|
    next 1 unless token[0] == :COMMENT
    before, after = @tokens[i - 2], @tokens[i + 2]
    if before && after &&
        ((before[0] == :INDENT && after[0] == :OUTDENT) ||
        (before[0] == :OUTDENT && after[0] == :INDENT)) &&
        before[1] == after[1]
      @tokens.delete_at(i + 2)
      @tokens.delete_at(i - 2)
      next 0
    elsif prev[0] == "\n" && [:INDENT].include?(after[0])
      @tokens.delete_at(i + 2)
      @tokens[i - 1] = after
      next 1
    elsif !["\n", :INDENT, :OUTDENT].include?(prev[0])
      @tokens.insert(i, ["\n", Value.new("\n", token[1].line)])
      next 2
    else
      next 1
    end
  end
end

#close_open_calls_and_indexes ⇒ Object 



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
 
# File 'lib/nscript/lexer/rewriter.rb', line 97

def close_open_calls_and_indexes
  parens, brackets = [0], [0]
  scan_tokens do |prev, token, post, i|
    case token[0]
    when :CALL_START  then parens.push(0)
    when :INDEX_START then brackets.push(0)
    when '('          then parens[-1] += 1
    when '['          then brackets[-1] += 1
    when ')'
      if parens.last == 0
        parens.pop
        token[0] = :CALL_END
      else
        parens[-1] -= 1
      end
    when ']'
      if brackets.last == 0
        brackets.pop
        token[0] = :INDEX_END
      else
        brackets[-1] -= 1
      end
    end
    next 1
  end
end

#ensure_balance(*pairs) ⇒ Object

Raises:



173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
 
# File 'lib/nscript/lexer/rewriter.rb', line 173

def ensure_balance(*pairs)
  puts "\nbefore ensure_balance: #{@tokens.inspect}" if ENV['VERBOSE']
  levels, lines = Hash.new(0), Hash.new
  scan_tokens do |prev, token, post, i|
    pairs.each do |pair|
      open, close = *pair
      levels[open] += 1 if token[0] == open
      levels[open] -= 1 if token[0] == close
      lines[token[0]] = token[1].line
      raise ParseError.new(token[0], token[1], nil) if levels[open] < 0
    end
    next 1
  end
  unclosed = levels.detect {|k, v| v > 0 }
  sym = unclosed && unclosed[0]
  raise ParseError.new(sym, Value.new(sym, lines[sym]), nil, "unclosed '#{sym}'") if unclosed
end

#move_commas_outside_outdents ⇒ Object 



87
88
89
90
91
92
93
94
95
 
# File 'lib/nscript/lexer/rewriter.rb', line 87

def move_commas_outside_outdents
  scan_tokens do |prev, token, post, i|
    if token[0] == :OUTDENT && prev[0] == ','
      @tokens.delete_at(i)
      @tokens.insert(i - 1, token)
    end
    next 1
  end
end

#remove_leading_newlines ⇒ Object 



75
76
77
 
# File 'lib/nscript/lexer/rewriter.rb', line 75

def remove_leading_newlines
  @tokens.shift if @tokens[0][0] == "\n"
end

#remove_mid_expression_newlines ⇒ Object 



79
80
81
82
83
84
85
 
# File 'lib/nscript/lexer/rewriter.rb', line 79

def remove_mid_expression_newlines
  scan_tokens do |prev, token, post, i|
    next 1 unless post && EXPRESSION_CLOSE.include?(post[0]) && token[0] == "\n"
    @tokens.delete_at(i)
    next 0
  end
end

#rewrite(tokens) ⇒ Object 



28
29
30
31
32
33
34
35
36
37
38
39
40
 
# File 'lib/nscript/lexer/rewriter.rb', line 28

def rewrite(tokens)
  @tokens = tokens
  adjust_comments
  remove_leading_newlines
  remove_mid_expression_newlines
  move_commas_outside_outdents
  close_open_calls_and_indexes
  add_implicit_parentheses
  add_implicit_indentation
  ensure_balance(*BALANCED_PAIRS)
  rewrite_closing_parens
  @tokens
end

#rewrite_closing_parens ⇒ Object 



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/nscript/lexer/rewriter.rb', line 191

def rewrite_closing_parens
  verbose = ENV['VERBOSE']
  stack, debt = [], Hash.new(0)
  stack_stats = lambda { "stack: #{stack.inspect} debt: #{debt.inspect}\n\n" }
  puts "rewrite_closing_original: #{@tokens.inspect}" if verbose
  scan_tokens do |prev, token, post, i|
    tag, inv = token[0], INVERSES[token[0]]
    # Push openers onto the stack.
    if EXPRESSION_START.include?(tag)
      stack.push(token)
      puts "pushing #{tag} #{stack_stats[]}" if verbose
      next 1
    # The end of an expression, check stack and debt for a pair.
    elsif EXPRESSION_TAIL.include?(tag)
      puts @tokens[i..-1].inspect if verbose
      # If the tag is already in our debt, swallow it.
      if debt[inv] > 0
        debt[inv] -= 1
        @tokens.delete_at(i)
        puts "tag in debt #{tag} #{stack_stats[]}" if verbose
        next 0
      else
        # Pop the stack of open delimiters.
        match = stack.pop
        mtag  = match[0]
        # Continue onwards if it's the expected tag.
        if tag == INVERSES[mtag]
          puts "expected tag #{tag} #{stack_stats[]}" if verbose
          next 1
        else
          # Unexpected close, insert correct close, adding to the debt.
          debt[mtag] += 1
          puts "unexpected #{tag}, replacing with #{INVERSES[mtag]} #{stack_stats[]}" if verbose
          val = mtag == :INDENT ? match[1] : INVERSES[mtag]
          @tokens.insert(i, [INVERSES[mtag], Value.new(val, token[1].line)])
          next 1
        end
      end
    else
      # Uninteresting token:
      next 1
    end
  end
end

#scan_tokens ⇒ Object 



42
43
44
45
46
47
48
49
 
# File 'lib/nscript/lexer/rewriter.rb', line 42

def scan_tokens
  i = 0
  loop do
    break unless @tokens[i]
    move = yield(@tokens[i - 1], @tokens[i], @tokens[i + 1], i)
    i += move
  end
end