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.
5 # Copyright 2006 Google, Inc. All Rights Reserved.
6 # Licensed to PSF under a Contributor Agreement.
10 This provides a high-level interface to parse a file into a syntax tree.
14 __author__ = "Guido van Rossum <guido@python.org>"
16 __all__ = ["Driver", "load_grammar"]
27 from . import grammar, parse, token, tokenize, pgen
31 def __init__(self, grammar, convert=None, logger=None):
32 self.grammar = grammar
34 logger = logging.getLogger(__name__)
36 self.convert = convert
38 def parse_tokens(self, tokens, debug=False):
39 """Parse a series of tokens and return the syntax tree."""
40 # XXX Move the prefix computation into a wrapper around tokenize.
41 p = parse.Parser(self.grammar, self.convert)
46 type = value = start = end = line_text = None
48 for quintuple in tokens:
49 type, value, start, end, line_text = quintuple
50 if start != (lineno, column):
51 assert (lineno, column) <= start, ((lineno, column), start)
52 s_lineno, s_column = start
54 prefix += "\n" * (s_lineno - lineno)
58 prefix += line_text[column:s_column]
60 if type in (tokenize.COMMENT, tokenize.NL):
63 if value.endswith("\n"):
68 type = grammar.opmap[value]
71 "%s %r (prefix=%r)", token.tok_name[type], value, prefix
73 if type == token.INDENT:
74 indent_columns.append(len(value))
75 _prefix = prefix + value
78 elif type == token.DEDENT:
79 _indent_col = indent_columns.pop()
80 prefix, _prefix = self._partially_consume_prefix(prefix, _indent_col)
81 if p.addtoken(type, value, (prefix, start)):
83 self.logger.debug("Stop.")
86 if type in {token.INDENT, token.DEDENT}:
89 if value.endswith("\n"):
93 # We never broke out -- EOF is too soon (how can this happen???)
94 raise parse.ParseError("incomplete input", type, value, (prefix, start))
97 def parse_stream_raw(self, stream, debug=False):
98 """Parse a stream and return the syntax tree."""
99 tokens = tokenize.generate_tokens(stream.readline, grammar=self.grammar)
100 return self.parse_tokens(tokens, debug)
102 def parse_stream(self, stream, debug=False):
103 """Parse a stream and return the syntax tree."""
104 return self.parse_stream_raw(stream, debug)
106 def parse_file(self, filename, encoding=None, debug=False):
107 """Parse a file and return the syntax tree."""
108 with io.open(filename, "r", encoding=encoding) as stream:
109 return self.parse_stream(stream, debug)
111 def parse_string(self, text, debug=False):
112 """Parse a string and return the syntax tree."""
113 tokens = tokenize.generate_tokens(
114 io.StringIO(text).readline, grammar=self.grammar
116 return self.parse_tokens(tokens, debug)
118 def _partially_consume_prefix(self, prefix, column):
127 if current_line.strip() and current_column < column:
129 return res, prefix[len(res) :]
131 lines.append(current_line)
138 # unexpected empty line
143 return "".join(lines), current_line
146 def _generate_pickle_name(gt, cache_dir=None):
147 head, tail = os.path.splitext(gt)
150 name = head + tail + ".".join(map(str, sys.version_info)) + ".pickle"
152 return os.path.join(cache_dir, os.path.basename(name))
157 def load_grammar(gt="Grammar.txt", gp=None, save=True, force=False, logger=None):
158 """Load the grammar (maybe from a pickle)."""
160 logger = logging.getLogger(__name__)
161 gp = _generate_pickle_name(gt) if gp is None else gp
162 if force or not _newer(gp, gt):
163 logger.info("Generating grammar tables from %s", gt)
164 g = pgen.generate_grammar(gt)
166 logger.info("Writing grammar tables to %s", gp)
170 logger.info("Writing failed: %s", e)
172 g = grammar.Grammar()
178 """Inquire whether file a was written since file b."""
179 if not os.path.exists(a):
181 if not os.path.exists(b):
183 return os.path.getmtime(a) >= os.path.getmtime(b)
186 def load_packaged_grammar(package, grammar_source, cache_dir=None):
187 """Normally, loads a pickled grammar by doing
188 pkgutil.get_data(package, pickled_grammar)
189 where *pickled_grammar* is computed from *grammar_source* by adding the
190 Python version and using a ``.pickle`` extension.
192 However, if *grammar_source* is an extant file, load_grammar(grammar_source)
193 is called instead. This facilitates using a packaged grammar file when needed
194 but preserves load_grammar's automatic regeneration behavior when possible.
197 if os.path.isfile(grammar_source):
198 gp = _generate_pickle_name(grammar_source, cache_dir) if cache_dir else None
199 return load_grammar(grammar_source, gp=gp)
200 pickled_name = _generate_pickle_name(os.path.basename(grammar_source), cache_dir)
201 data = pkgutil.get_data(package, pickled_name)
202 g = grammar.Grammar()
208 """Main program, when run as a script: produce grammar pickle files.
210 Calls load_grammar for each argument, a path to a grammar text file.
214 logging.basicConfig(level=logging.INFO, stream=sys.stdout, format="%(message)s")
216 load_grammar(gt, save=True, force=True)
220 if __name__ == "__main__":
221 sys.exit(int(not main()))