Pickle raw tuples in FileData cache (#3877)

[etc/vim.git] / src / blib2to3 / pgen2 / driver.py

diff --git a/src/blib2to3/pgen2/driver.py b/src/blib2to3/pgen2/driver.py
index 5edd75b1333991d692a8f57ac84101b1a609e916..e629843f8b986ea792d59a150ff64018d2108bcb 100644 (file)
--- a/src/blib2to3/pgen2/driver.py
+++ b/src/blib2to3/pgen2/driver.py
@@ -17,31 +17,21 @@ __all__ = ["Driver", "load_grammar"]
 
 # Python imports
 import io
 
 # Python imports
 import io

-import os

 import logging
 import logging

+import os

 import pkgutil
 import sys
 import pkgutil
 import sys

-from typing import (
-    Any,
-    IO,
-    Iterable,
-    List,
-    Optional,
-    Text,
-    Iterator,
-    Tuple,
-    TypeVar,
-    Generic,
-    Union,
-)
+from contextlib import contextmanager

 from dataclasses import dataclass, field
 from dataclasses import dataclass, field

-
-# Pgen imports
-from . import grammar, parse, token, tokenize, pgen

 from logging import Logger
 from logging import Logger

-from blib2to3.pytree import _Convert, NL
+from typing import IO, Any, Iterable, Iterator, List, Optional, Tuple, Union, cast
+

 from blib2to3.pgen2.grammar import Grammar
 from blib2to3.pgen2.grammar import Grammar

-from contextlib import contextmanager
+from blib2to3.pgen2.tokenize import GoodTokenInfo
+from blib2to3.pytree import NL
+
+# Pgen imports
+from . import grammar, parse, pgen, token, tokenize

 
 Path = Union[str, "os.PathLike[str]"]
 
 
 Path = Union[str, "os.PathLike[str]"]
 


@@ -105,7 +95,7 @@ class TokenProxy:
 
     def can_advance(self, to: int) -> bool:
         # Try to eat, fail if it can't. The eat operation is cached
 
     def can_advance(self, to: int) -> bool:
         # Try to eat, fail if it can't. The eat operation is cached

-        # so there wont be any additional cost of eating here
+        # so there won't be any additional cost of eating here

         try:
             self.eat(to)
         except StopIteration:
         try:
             self.eat(to)
         except StopIteration:


@@ -114,30 +104,24 @@ class TokenProxy:
             return True
 
 
             return True
 
 

-class Driver(object):
-    def __init__(
-        self,
-        grammar: Grammar,
-        convert: Optional[_Convert] = None,
-        logger: Optional[Logger] = None,
-    ) -> None:
+class Driver:
+    def __init__(self, grammar: Grammar, logger: Optional[Logger] = None) -> None:

         self.grammar = grammar
         if logger is None:
             logger = logging.getLogger(__name__)
         self.logger = logger
         self.grammar = grammar
         if logger is None:
             logger = logging.getLogger(__name__)
         self.logger = logger

-        self.convert = convert

 
 

-    def parse_tokens(self, tokens: Iterable[Any], debug: bool = False) -> NL:
+    def parse_tokens(self, tokens: Iterable[GoodTokenInfo], debug: bool = False) -> NL:

         """Parse a series of tokens and return the syntax tree."""
         # XXX Move the prefix computation into a wrapper around tokenize.
         proxy = TokenProxy(tokens)
 
         """Parse a series of tokens and return the syntax tree."""
         # XXX Move the prefix computation into a wrapper around tokenize.
         proxy = TokenProxy(tokens)
 

-        p = parse.Parser(self.grammar, self.convert)
+        p = parse.Parser(self.grammar)

         p.setup(proxy=proxy)
 
         lineno = 1
         column = 0
         p.setup(proxy=proxy)
 
         lineno = 1
         column = 0

-        indent_columns = []
+        indent_columns: List[int] = []

         type = value = start = end = line_text = None
         prefix = ""
 
         type = value = start = end = line_text = None
         prefix = ""
 


@@ -163,6 +147,7 @@ class Driver(object):
             if type == token.OP:
                 type = grammar.opmap[value]
             if debug:
             if type == token.OP:
                 type = grammar.opmap[value]
             if debug:

+                assert type is not None

                 self.logger.debug(
                     "%s %r (prefix=%r)", token.tok_name[type], value, prefix
                 )
                 self.logger.debug(
                     "%s %r (prefix=%r)", token.tok_name[type], value, prefix
                 )


@@ -174,7 +159,7 @@ class Driver(object):
             elif type == token.DEDENT:
                 _indent_col = indent_columns.pop()
                 prefix, _prefix = self._partially_consume_prefix(prefix, _indent_col)
             elif type == token.DEDENT:
                 _indent_col = indent_columns.pop()
                 prefix, _prefix = self._partially_consume_prefix(prefix, _indent_col)

-            if p.addtoken(type, value, (prefix, start)):
+            if p.addtoken(cast(int, type), value, (prefix, start)):

                 if debug:
                     self.logger.debug("Stop.")
                 break
                 if debug:
                     self.logger.debug("Stop.")
                 break


@@ -192,30 +177,30 @@ class Driver(object):
         assert p.rootnode is not None
         return p.rootnode
 
         assert p.rootnode is not None
         return p.rootnode
 

-    def parse_stream_raw(self, stream: IO[Text], debug: bool = False) -> NL:
+    def parse_stream_raw(self, stream: IO[str], debug: bool = False) -> NL:

         """Parse a stream and return the syntax tree."""
         tokens = tokenize.generate_tokens(stream.readline, grammar=self.grammar)
         return self.parse_tokens(tokens, debug)
 
         """Parse a stream and return the syntax tree."""
         tokens = tokenize.generate_tokens(stream.readline, grammar=self.grammar)
         return self.parse_tokens(tokens, debug)
 

-    def parse_stream(self, stream: IO[Text], debug: bool = False) -> NL:
+    def parse_stream(self, stream: IO[str], debug: bool = False) -> NL:

         """Parse a stream and return the syntax tree."""
         return self.parse_stream_raw(stream, debug)
 
     def parse_file(
         """Parse a stream and return the syntax tree."""
         return self.parse_stream_raw(stream, debug)
 
     def parse_file(

-        self, filename: Path, encoding: Optional[Text] = None, debug: bool = False
+        self, filename: Path, encoding: Optional[str] = None, debug: bool = False

     ) -> NL:
         """Parse a file and return the syntax tree."""
     ) -> NL:
         """Parse a file and return the syntax tree."""

-        with io.open(filename, "r", encoding=encoding) as stream:
+        with open(filename, encoding=encoding) as stream:

             return self.parse_stream(stream, debug)
 
             return self.parse_stream(stream, debug)
 

-    def parse_string(self, text: Text, debug: bool = False) -> NL:
+    def parse_string(self, text: str, debug: bool = False) -> NL:

         """Parse a string and return the syntax tree."""
         tokens = tokenize.generate_tokens(
             io.StringIO(text).readline, grammar=self.grammar
         )
         return self.parse_tokens(tokens, debug)
 
         """Parse a string and return the syntax tree."""
         tokens = tokenize.generate_tokens(
             io.StringIO(text).readline, grammar=self.grammar
         )
         return self.parse_tokens(tokens, debug)
 

-    def _partially_consume_prefix(self, prefix: Text, column: int) -> Tuple[Text, Text]:
+    def _partially_consume_prefix(self, prefix: str, column: int) -> Tuple[str, str]:

         lines: List[str] = []
         current_line = ""
         current_column = 0
         lines: List[str] = []
         current_line = ""
         current_column = 0


@@ -243,7 +228,7 @@ class Driver(object):
         return "".join(lines), current_line
 
 
         return "".join(lines), current_line
 
 

-def _generate_pickle_name(gt: Path, cache_dir: Optional[Path] = None) -> Text:
+def _generate_pickle_name(gt: Path, cache_dir: Optional[Path] = None) -> str:

     head, tail = os.path.splitext(gt)
     if tail == ".txt":
         tail = ""
     head, tail = os.path.splitext(gt)
     if tail == ".txt":
         tail = ""


@@ -255,8 +240,8 @@ def _generate_pickle_name(gt: Path, cache_dir: Optional[Path] = None) -> Text:
 
 
 def load_grammar(
 
 
 def load_grammar(

-    gt: Text = "Grammar.txt",
-    gp: Optional[Text] = None,
+    gt: str = "Grammar.txt",
+    gp: Optional[str] = None,

     save: bool = True,
     force: bool = False,
     logger: Optional[Logger] = None,
     save: bool = True,
     force: bool = False,
     logger: Optional[Logger] = None,


@@ -266,21 +251,20 @@ def load_grammar(
         logger = logging.getLogger(__name__)
     gp = _generate_pickle_name(gt) if gp is None else gp
     if force or not _newer(gp, gt):
         logger = logging.getLogger(__name__)
     gp = _generate_pickle_name(gt) if gp is None else gp
     if force or not _newer(gp, gt):

-        logger.info("Generating grammar tables from %s", gt)

         g: grammar.Grammar = pgen.generate_grammar(gt)
         if save:
         g: grammar.Grammar = pgen.generate_grammar(gt)
         if save:

-            logger.info("Writing grammar tables to %s", gp)

             try:
                 g.dump(gp)
             try:
                 g.dump(gp)

-            except OSError as e:
-                logger.info("Writing failed: %s", e)
+            except OSError:
+                # Ignore error, caching is not vital.
+                pass

     else:
         g = grammar.Grammar()
         g.load(gp)
     return g
 
 
     else:
         g = grammar.Grammar()
         g.load(gp)
     return g
 
 

-def _newer(a: Text, b: Text) -> bool:
+def _newer(a: str, b: str) -> bool:

     """Inquire whether file a was written since file b."""
     if not os.path.exists(a):
         return False
     """Inquire whether file a was written since file b."""
     if not os.path.exists(a):
         return False


@@ -290,7 +274,7 @@ def _newer(a: Text, b: Text) -> bool:
 
 
 def load_packaged_grammar(
 
 
 def load_packaged_grammar(

-    package: str, grammar_source: Text, cache_dir: Optional[Path] = None
+    package: str, grammar_source: str, cache_dir: Optional[Path] = None

 ) -> grammar.Grammar:
     """Normally, loads a pickled grammar by doing
         pkgutil.get_data(package, pickled_grammar)
 ) -> grammar.Grammar:
     """Normally, loads a pickled grammar by doing
         pkgutil.get_data(package, pickled_grammar)


@@ -313,7 +297,7 @@ def load_packaged_grammar(
     return g
 
 
     return g
 
 

-def main(*args: Text) -> bool:
+def main(*args: str) -> bool:

     """Main program, when run as a script: produce grammar pickle files.
 
     Calls load_grammar for each argument, a path to a grammar text file.
     """Main program, when run as a script: produce grammar pickle files.
 
     Calls load_grammar for each argument, a path to a grammar text file.