All patches and comments are welcome. Please squash your changes to logical
commits before using git-format-patch and git-send-email to
patches@git.madduck.net.
If you'd read over the Git project's submission guidelines and adhered to them,
I'd be especially grateful.
1 # Copyright 2004-2005 Elemental Security, Inc. All Rights Reserved.
2 # Licensed to PSF under a Contributor Agreement.
4 """This module defines the data structures used to represent a grammar.
6 These are a bit arcane because they are derived from the data
7 structures used by Python's 'pgen' parser generator.
9 There's also a table here mapping operators to their names in the
10 token module; the Python tokenize module reports all operators as the
11 fallback token code OP, but the parser needs the actual token code.
24 class Grammar(object):
25 """Pgen parsing tables conversion class.
27 Once initialized, this class supplies the grammar tables for the
28 parsing engine implemented by parse.py. The parsing engine
29 accesses the instance variables directly. The class here does not
30 provide initialization of the tables; several subclasses exist to
31 do this (see the conv and pgen modules).
33 The load() method reads the tables from a pickle file, which is
34 much faster than the other ways offered by subclasses. The pickle
35 file is written by calling dump() (after loading the grammar
36 tables using a subclass). The report() method prints a readable
37 representation of the tables to stdout, for debugging.
39 The instance variables are as follows:
41 symbol2number -- a dict mapping symbol names to numbers. Symbol
42 numbers are always 256 or higher, to distinguish
43 them from token numbers, which are between 0 and
46 number2symbol -- a dict mapping numbers to symbol names;
47 these two are each other's inverse.
49 states -- a list of DFAs, where each DFA is a list of
50 states, each state is a list of arcs, and each
51 arc is a (i, j) pair where i is a label and j is
52 a state number. The DFA number is the index into
53 this list. (This name is slightly confusing.)
54 Final states are represented by a special arc of
55 the form (0, j) where j is its own state number.
57 dfas -- a dict mapping symbol numbers to (DFA, first)
58 pairs, where DFA is an item from the states list
59 above, and first is a set of tokens that can
60 begin this grammar rule (represented by a dict
61 whose values are always 1).
63 labels -- a list of (x, y) pairs where x is either a token
64 number or a symbol number, and y is either None
65 or a string; the strings are keywords. The label
66 number is the index in this list; label numbers
67 are used to mark state transitions (arcs) in the
70 start -- the number of the grammar's start symbol.
72 keywords -- a dict mapping keyword strings to arc labels.
74 tokens -- a dict mapping token numbers to arc labels.
79 self.symbol2number = {}
80 self.number2symbol = {}
83 self.labels = [(0, "EMPTY")]
86 self.symbol2label = {}
88 # Python 3.7+ parses async as a keyword, not an identifier
89 self.async_keywords = False
91 def dump(self, filename):
92 """Dump the grammar tables to a pickle file."""
93 with tempfile.NamedTemporaryFile(dir=os.path.dirname(filename), delete=False) as f:
94 pickle.dump(self.__dict__, f, pickle.HIGHEST_PROTOCOL)
95 os.replace(f.name, filename)
97 def load(self, filename):
98 """Load the grammar tables from a pickle file."""
99 with open(filename, "rb") as f:
101 self.__dict__.update(d)
103 def loads(self, pkl):
104 """Load the grammar tables from a pickle bytes object."""
105 self.__dict__.update(pickle.loads(pkl))
111 new = self.__class__()
112 for dict_attr in ("symbol2number", "number2symbol", "dfas", "keywords",
113 "tokens", "symbol2label"):
114 setattr(new, dict_attr, getattr(self, dict_attr).copy())
115 new.labels = self.labels[:]
116 new.states = self.states[:]
117 new.start = self.start
118 new.async_keywords = self.async_keywords
122 """Dump the grammar tables to standard output, for debugging."""
123 from pprint import pprint
125 pprint(self.symbol2number)
127 pprint(self.number2symbol)
134 print("start", self.start)
137 # Map from operator to number (since tokenize doesn't do this)
190 for line in opmap_raw.splitlines():
192 op, name = line.split()
193 opmap[op] = getattr(token, name)