Mercurial > lcfOS
annotate python/pyyacc.py @ 399:a7c444404df9
Fix hexwrite
| author | Windel Bouwman |
|---|---|
| date | Fri, 20 Jun 2014 16:36:49 +0200 |
| parents | 3b0c495e3008 |
| children |
| rev | line source |
|---|---|
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
1 """ |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
2 Parser generator script |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
3 """ |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
4 |
| 191 | 5 from ppci import Token |
| 6 | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
7 EPS = 'EPS' |
| 184 | 8 EOF = 'EOF' |
| 318 | 9 |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
10 |
| 185 | 11 class ParserGenerationException(Exception): |
| 186 | 12 """ Raised when something goes wrong during parser generation """ |
| 185 | 13 pass |
| 14 | |
| 15 | |
| 16 class ParserException(Exception): | |
| 186 | 17 """ Raised during a failure in the parsing process """ |
| 185 | 18 pass |
| 19 | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
20 |
| 318 | 21 class Action: |
| 22 def __repr__(self): | |
| 23 return 'Action' | |
| 24 | |
| 25 def __eq__(self, other): | |
| 26 return str(self) == str(other) | |
| 27 | |
| 28 | |
| 29 class Shift(Action): | |
| 30 def __init__(self, to_state): | |
| 31 self.to_state = to_state | |
| 32 | |
| 33 def __repr__(self): | |
| 34 return 'Shift({})'.format(self.to_state) | |
| 35 | |
| 36 | |
| 37 class Reduce(Action): | |
| 38 def __init__(self, rule): | |
| 39 self.rule = rule | |
| 40 | |
| 41 def __repr__(self): | |
| 42 return 'Reduce({})'.format(self.rule) | |
| 43 | |
| 44 | |
| 45 class Accept(Reduce): | |
| 46 def __repr__(self): | |
| 47 return 'Accept({})'.format(self.rule) | |
| 48 | |
| 49 | |
| 50 def print_grammar(g): | |
| 51 """ Pretty print a grammar """ | |
| 52 print(g) | |
| 53 for production in g.productions: | |
| 54 print(production) | |
| 55 | |
| 319 | 56 |
| 57 def calculate_first_sets(grammar): | |
| 58 """ | |
| 59 Calculate first sets for each grammar symbol | |
| 60 This is a dictionary which maps each grammar symbol | |
| 61 to a set of terminals that can be encountered first | |
| 62 when looking for the symbol. | |
| 63 """ | |
| 64 first = {} | |
| 65 nullable = {} | |
| 66 for terminal in grammar.terminals | {EOF, EPS}: | |
| 67 first[terminal] = set([terminal]) | |
| 68 nullable[terminal] = False | |
| 69 for nt in grammar.nonterminals: | |
| 70 first[nt] = set() | |
| 71 nullable[nt] = False | |
| 72 while True: | |
| 73 some_change = False | |
| 74 for rule in grammar.productions: | |
| 75 # Check for null-ability: | |
| 76 if all(nullable[beta] for beta in rule.symbols): | |
| 77 if not nullable[rule.name]: | |
| 78 nullable[rule.name] = True | |
| 79 some_change = True | |
| 80 # Update first sets: | |
| 81 for beta in rule.symbols: | |
| 82 if not nullable[beta]: | |
| 83 if first[beta] - first[rule.name]: | |
| 84 first[rule.name] |= first[beta] | |
| 85 some_change = True | |
| 86 break | |
| 87 if not some_change: | |
| 88 break | |
| 89 return first | |
| 90 | |
| 91 | |
| 178 | 92 class Grammar: |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
93 """ Defines a grammar of a language """ |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
94 def __init__(self, terminals): |
| 319 | 95 self.terminals = set(terminals) |
| 96 self.nonterminals = set() | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
97 self.productions = [] |
| 185 | 98 self._first = None # Cached first set |
| 318 | 99 self.start_symbol = None |
| 100 | |
| 101 def __repr__(self): | |
| 102 return 'Grammar with {} rules'.format(len(self.productions)) | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
103 |
| 194 | 104 def add_production(self, name, symbols, f=None): |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
105 """ Add a production rule to the grammar """ |
| 194 | 106 production = Production(name, symbols, f) |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
107 self.productions.append(production) |
| 192 | 108 if name in self.terminals: |
| 109 raise ParserGenerationException("Cannot redefine terminal {0}".format(name)) | |
| 319 | 110 self.nonterminals.add(name) |
| 194 | 111 self._first = None # Invalidate cached version |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
112 |
| 383 | 113 def add_one_or_more(self, element_nonterm, list_nonterm): |
| 114 """ Helper to add the rule | |
| 115 lst: elem | |
| 116 lst: lst elem | |
| 117 """ | |
| 118 def a(el): | |
| 119 return [el] | |
| 120 def b(ls, el): | |
| 121 ls.append(el) | |
| 122 return ls | |
| 123 self.add_production(list_nonterm, [element_nonterm], a) | |
| 124 self.add_production(list_nonterm, [list_nonterm, element_nonterm], b) | |
| 125 | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
126 def productionsForName(self, name): |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
127 """ Retrieve all productions for a non terminal """ |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
128 return [p for p in self.productions if p.name == name] |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
129 |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
130 @property |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
131 def Symbols(self): |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
132 """ Get all the symbols defined by this grammar """ |
| 319 | 133 return self.nonterminals | self.terminals |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
134 |
| 185 | 135 @property |
| 136 def first(self): | |
| 318 | 137 """ |
| 186 | 138 The first set is a mapping from a grammar symbol to a set of |
| 139 set of all terminal symbols that can be the first terminal when | |
| 318 | 140 looking for the grammar symbol |
| 186 | 141 """ |
| 185 | 142 if not self._first: |
| 319 | 143 self._first = calculate_first_sets(self) |
| 185 | 144 return self._first |
| 145 | |
| 184 | 146 def closure(self, itemset): |
| 147 """ Expand itemset by using epsilon moves """ | |
| 148 worklist = list(itemset) | |
| 149 def addIt(itm): | |
| 150 if not itm in itemset: | |
| 151 itemset.add(itm) | |
| 152 worklist.append(itm) | |
| 186 | 153 def first2(itm): |
| 184 | 154 # When using the first sets, create a copy: |
| 186 | 155 f = set(self.first[itm.NextNext]) |
| 184 | 156 if EPS in f: |
| 157 f.discard(EPS) | |
| 186 | 158 f.add(itm.look_ahead) |
| 184 | 159 return f |
| 319 | 160 # Start of algorithm: |
| 184 | 161 while worklist: |
| 162 item = worklist.pop(0) | |
| 163 if not item.IsShift: | |
| 164 continue | |
| 165 if not (item.Next in self.nonterminals): | |
| 166 continue | |
| 167 C = item.Next | |
| 168 for add_p in self.productionsForName(C): | |
| 186 | 169 for b in first2(item): |
| 184 | 170 addIt(Item(add_p, 0, b)) |
| 171 return frozenset(itemset) | |
| 172 | |
| 173 def initialItemSet(self): | |
| 174 """ Calculates the initial item set """ | |
| 175 iis = set() | |
| 176 for p in self.productionsForName(self.start_symbol): | |
| 177 iis.add(Item(p, 0, EOF)) | |
| 178 return self.closure(iis) | |
| 179 | |
| 180 def nextItemSet(self, itemset, symbol): | |
| 318 | 181 """ |
| 319 | 182 Determines the next itemset for the current set and a symbol |
| 184 | 183 This is the goto procedure |
| 184 """ | |
| 185 next_set = set() | |
| 186 for item in itemset: | |
| 187 if item.can_shift_over(symbol): | |
| 188 next_set.add(item.shifted()) | |
| 189 return self.closure(next_set) | |
| 318 | 190 |
| 184 | 191 def genCanonicalSet(self, iis): |
| 185 | 192 states = [] |
| 184 | 193 worklist = [] |
| 185 | 194 transitions = {} |
| 184 | 195 def addSt(s): |
| 196 if not (s in states): | |
| 197 worklist.append(s) | |
| 185 | 198 states.append(s) |
| 184 | 199 addSt(iis) |
| 200 while len(worklist) > 0: | |
| 201 itemset = worklist.pop(0) | |
| 202 for symbol in self.Symbols: | |
| 203 nis = self.nextItemSet(itemset, symbol) | |
| 204 if not nis: | |
| 205 continue | |
| 206 addSt(nis) | |
| 185 | 207 transitions[(states.index(itemset), symbol)] = states.index(nis) |
| 208 return states, transitions | |
| 318 | 209 |
| 192 | 210 def checkSymbols(self): |
| 211 """ Checks no symbols are undefined """ | |
| 212 for production in self.productions: | |
| 213 for symbol in production.symbols: | |
| 319 | 214 if symbol not in self.Symbols: |
| 192 | 215 raise ParserGenerationException('Symbol {0} undefined'.format(symbol)) |
| 216 | |
| 340 | 217 def generate_parser(self): |
| 218 """ Generates a parser from the grammar """ | |
| 219 action_table, goto_table = self.generate_tables() | |
| 218 | 220 p = LRParser(action_table, goto_table, self.start_symbol) |
| 221 p.grammar = self | |
| 222 return p | |
| 223 | |
| 340 | 224 def generate_tables(self): |
| 225 """ Generate parsing tables """ | |
| 318 | 226 if not self.start_symbol: |
| 227 self.start_symbol = self.productions[0].name | |
| 192 | 228 self.checkSymbols() |
| 184 | 229 action_table = {} |
| 185 | 230 goto_table = {} |
| 184 | 231 iis = self.initialItemSet() |
| 232 | |
| 233 # First generate all item sets by using the nextItemset function: | |
| 185 | 234 states, transitions = self.genCanonicalSet(iis) |
| 194 | 235 |
| 185 | 236 def setAction(state, t, action): |
| 318 | 237 assert isinstance(action, Action) |
| 185 | 238 key = (state, t) |
| 218 | 239 assert type(state) is int |
| 240 assert type(t) is str | |
| 185 | 241 if key in action_table: |
| 242 action2 = action_table[key] | |
| 243 if action != action2: | |
| 318 | 244 if (type(action2) is Reduce) and (type(action) is Shift): |
| 194 | 245 # Automatically resolve and do the shift action! |
| 246 # Simple, but almost always what you want!! | |
| 247 action_table[key] = action | |
| 318 | 248 elif isinstance(action2, Shift) and isinstance(action, Reduce): |
| 249 pass | |
| 194 | 250 else: |
| 318 | 251 a1 = str(action) |
| 252 a2 = str(action2) | |
| 253 raise ParserGenerationException('LR construction conflict {0} vs {1}'.format(a1, a2)) | |
| 185 | 254 else: |
| 255 action_table[key] = action | |
| 184 | 256 |
| 257 # Fill action table: | |
| 258 for state in states: | |
| 185 | 259 # Detect conflicts: |
| 184 | 260 for item in state: |
| 261 if item.IsShift and item.Next in self.terminals: | |
| 185 | 262 # Rule 1, a shift item: |
| 263 nextstate = transitions[(states.index(state), item.Next)] | |
| 318 | 264 setAction(states.index(state), item.Next, Shift(nextstate)) |
| 185 | 265 if item.IsReduce: |
| 266 if item.production.name == self.start_symbol and item.look_ahead == EOF: | |
| 267 # Rule 3: accept: | |
| 318 | 268 act = Accept(self.productions.index(item.production)) |
| 184 | 269 else: |
| 185 | 270 # Rule 2, reduce item: |
| 318 | 271 act = Reduce(self.productions.index(item.production)) |
| 272 setAction(states.index(state), item.look_ahead, act) | |
| 185 | 273 for nt in self.nonterminals: |
| 274 key = (states.index(state), nt) | |
| 275 if key in transitions: | |
| 276 goto_table[key] = transitions[key] | |
| 218 | 277 return action_table, goto_table |
| 185 | 278 |
| 178 | 279 |
| 280 class Production: | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
281 """ Production rule for a grammar """ |
| 341 | 282 def __init__(self, name, symbols, f): |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
283 self.name = name |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
284 self.symbols = symbols |
| 194 | 285 self.f = f |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
286 |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
287 def __repr__(self): |
| 318 | 288 action = ' ' + str(self.f) if self.f else '' |
| 289 return '{0} -> {1}'.format(self.name, self.symbols) + action | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
290 |
| 179 | 291 |
| 178 | 292 class Item: |
| 318 | 293 """ |
| 294 Represents a partially parsed item | |
| 186 | 295 It has a production it is looking for, a position |
| 296 in this production called the 'dot' and a look ahead | |
| 297 symbol that must follow this item. | |
| 298 """ | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
299 def __init__(self, production, dotpos, look_ahead): |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
300 self.production = production |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
301 self.dotpos = dotpos |
| 184 | 302 assert self.dotpos <= len(self.production.symbols) |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
303 self.look_ahead = look_ahead |
| 346 | 304 self._is_shift = self.dotpos < len(self.production.symbols) |
| 395 | 305 self.IsShift = self._is_shift |
| 306 if self.IsShift: | |
| 307 self.Next = self.production.symbols[self.dotpos] | |
| 308 self._data = (self.production, self.dotpos, self.look_ahead) | |
| 309 self._hash = self._data.__hash__() | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
310 |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
311 def __eq__(self, other): |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
312 if type(other) is type(self): |
| 395 | 313 return self._data == other._data |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
314 return False |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
315 |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
316 def __hash__(self): |
| 395 | 317 return self._hash |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
318 |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
319 @property |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
320 def IsReduce(self): |
| 186 | 321 """ Check if this item has the dot at the end """ |
| 346 | 322 return not self._is_shift |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
323 |
| 184 | 324 def can_shift_over(self, symbol): |
| 325 """ Determines if this item can shift over the given symbol """ | |
| 395 | 326 return self._is_shift and self.Next == symbol |
| 184 | 327 |
| 328 def shifted(self): | |
| 329 """ Creates a new item that is shifted one position """ | |
| 330 return Item(self.production, self.dotpos + 1, self.look_ahead) | |
| 331 | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
332 @property |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
333 def NextNext(self): |
| 186 | 334 """ Gets the symbol after the next symbol, or EPS if at the end """ |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
335 if self.dotpos + 1 >= len(self.production.symbols): |
| 184 | 336 return EPS |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
337 else: |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
338 return self.production.symbols[self.dotpos + 1] |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
339 |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
340 def __repr__(self): |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
341 prod = self.production |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
342 predot = ' '.join(prod.symbols[0:self.dotpos]) |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
343 postdot = ' '.join(prod.symbols[self.dotpos:]) |
| 186 | 344 name = prod.name |
| 345 args = (name, predot, postdot, self.look_ahead) | |
| 185 | 346 return '[{0} -> {1} . {2} -> {3}]'.format(*args) |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
347 |
| 179 | 348 |
| 349 class LRParser: | |
| 395 | 350 """ LR parser automata. This class takes goto and action table |
| 351 and can then process a sequence of tokens. | |
| 352 """ | |
| 195 | 353 def __init__(self, action_table, goto_table, start_symbol): |
| 184 | 354 self.action_table = action_table |
| 185 | 355 self.goto_table = goto_table |
| 195 | 356 self.start_symbol = start_symbol |
| 184 | 357 |
| 318 | 358 def parse(self, lexer): |
| 191 | 359 """ Parse an iterable with tokens """ |
| 318 | 360 assert hasattr(lexer, 'next_token'), '{0} is no lexer'.format(type(lexer)) |
| 185 | 361 stack = [0] |
| 195 | 362 r_data_stack = [] |
| 318 | 363 look_ahead = lexer.next_token() |
| 191 | 364 assert type(look_ahead) is Token |
| 195 | 365 # TODO: exit on this condition: |
| 377 | 366 while stack != [0, self.start_symbol, 0]: |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
367 state = stack[-1] # top of stack |
| 191 | 368 key = (state, look_ahead.typ) |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
369 if not key in self.action_table: |
| 194 | 370 raise ParserException('Error parsing at character {0}'.format(look_ahead)) |
| 318 | 371 action = self.action_table[key] |
| 372 if type(action) is Reduce: | |
| 194 | 373 f_args = [] |
| 318 | 374 prod = self.grammar.productions[action.rule] |
| 375 for s in prod.symbols: | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
376 stack.pop() |
|
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
377 stack.pop() |
| 195 | 378 f_args.append(r_data_stack.pop()) |
| 379 f_args.reverse() | |
| 380 r_data = None | |
| 318 | 381 if prod.f: |
| 382 r_data = prod.f(*f_args) | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
383 state = stack[-1] |
| 318 | 384 stack.append(prod.name) |
| 385 stack.append(self.goto_table[(state, prod.name)]) | |
| 195 | 386 r_data_stack.append(r_data) |
| 318 | 387 elif type(action) is Shift: |
| 191 | 388 stack.append(look_ahead.typ) |
| 318 | 389 stack.append(action.to_state) |
| 390 r_data_stack.append(look_ahead) | |
| 391 look_ahead = lexer.next_token() | |
| 191 | 392 assert type(look_ahead) is Token |
| 318 | 393 elif type(action) is Accept: |
| 195 | 394 # Pop last rule data off the stack: |
| 395 f_args = [] | |
| 318 | 396 param = self.grammar.productions[action.rule] |
| 195 | 397 for s in param.symbols: |
| 398 stack.pop() | |
| 399 stack.pop() | |
| 400 f_args.append(r_data_stack.pop()) | |
| 401 f_args.reverse() | |
| 402 if param.f: | |
| 318 | 403 ret_val = param.f(*f_args) |
| 404 else: | |
| 405 ret_val = None | |
| 195 | 406 # Break out! |
| 377 | 407 stack.append(param.name) |
| 408 stack.append(0) | |
|
181
216da5e46efc
Changed indent to 4 spaces to comply to convention
Windel Bouwman
parents:
180
diff
changeset
|
409 break |
| 195 | 410 # At exit, the stack must be 1 long |
| 411 # TODO: fix that this holds: | |
| 377 | 412 #assert stack == [0, self.start_symbol, 0] |
| 318 | 413 return ret_val |
| 377 | 414 |