Don't let TokenError bubble up from lib2to3_parse (GH-2343)

[etc/vim.git] / src / black / parsing.py

diff --git a/src/black/parsing.py b/src/black/parsing.py
index ee6aae1e7ff9db26be2f5f4c439a25d31baf2b7d..b673027022f44b4a2ebd5a9b4b3bd44d0b163b3b 100644 (file)
--- a/src/black/parsing.py
+++ b/src/black/parsing.py
@@ -4,18 +4,27 @@ Parse Python code and perform AST validation.
 import ast
 import platform
 import sys
 import ast
 import platform
 import sys

-from typing import Iterable, Iterator, List, Set, Union, Tuple
+from typing import Any, Iterable, Iterator, List, Set, Tuple, Type, Union
+
+if sys.version_info < (3, 8):
+    from typing_extensions import Final
+else:
+    from typing import Final

 
 # lib2to3 fork
 from blib2to3.pytree import Node, Leaf
 
 # lib2to3 fork
 from blib2to3.pytree import Node, Leaf

-from blib2to3 import pygram, pytree
+from blib2to3 import pygram

 from blib2to3.pgen2 import driver
 from blib2to3.pgen2.grammar import Grammar
 from blib2to3.pgen2.parse import ParseError
 from blib2to3.pgen2 import driver
 from blib2to3.pgen2.grammar import Grammar
 from blib2to3.pgen2.parse import ParseError

+from blib2to3.pgen2.tokenize import TokenError

 
 from black.mode import TargetVersion, Feature, supports_feature
 from black.nodes import syms
 
 
 from black.mode import TargetVersion, Feature, supports_feature
 from black.nodes import syms
 

+ast3: Any
+ast27: Any
+

 _IS_PYPY = platform.python_implementation() == "PyPy"
 
 try:
 _IS_PYPY = platform.python_implementation() == "PyPy"
 
 try:


@@ -67,8 +76,10 @@ def get_grammars(target_versions: Set[TargetVersion]) -> List[Grammar]:
         # Python 3.10+
         grammars.append(pygram.python_grammar_soft_keywords)
     # If we have to parse both, try to parse async as a keyword first
         # Python 3.10+
         grammars.append(pygram.python_grammar_soft_keywords)
     # If we have to parse both, try to parse async as a keyword first

-    if not supports_feature(target_versions, Feature.ASYNC_IDENTIFIERS):
-        # Python 3.7+
+    if not supports_feature(
+        target_versions, Feature.ASYNC_IDENTIFIERS
+    ) and not supports_feature(target_versions, Feature.PATTERN_MATCHING):
+        # Python 3.7-3.9

         grammars.append(
             pygram.python_grammar_no_print_statement_no_exec_statement_async_keywords
         )
         grammars.append(
             pygram.python_grammar_no_print_statement_no_exec_statement_async_keywords
         )


@@ -86,7 +97,7 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -
         src_txt += "\n"
 
     for grammar in get_grammars(set(target_versions)):
         src_txt += "\n"
 
     for grammar in get_grammars(set(target_versions)):

-        drv = driver.Driver(grammar, pytree.convert)
+        drv = driver.Driver(grammar)

         try:
             result = drv.parse_string(src_txt, True)
             break
         try:
             result = drv.parse_string(src_txt, True)
             break


@@ -99,6 +110,12 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -
             except IndexError:
                 faulty_line = "<line number missing in source>"
             exc = InvalidInput(f"Cannot parse: {lineno}:{column}: {faulty_line}")
             except IndexError:
                 faulty_line = "<line number missing in source>"
             exc = InvalidInput(f"Cannot parse: {lineno}:{column}: {faulty_line}")

+
+        except TokenError as te:
+            # In edge cases these are raised; and typically don't have a "faulty_line".
+            lineno, column = te.args[1]
+            exc = InvalidInput(f"Cannot parse: {lineno}:{column}: {te.args[0]}")
+

     else:
         raise exc from None
 
     else:
         raise exc from None
 


@@ -148,6 +165,10 @@ def parse_ast(src: str) -> Union[ast.AST, ast3.AST, ast27.AST]:
     raise SyntaxError(first_error)
 
 
     raise SyntaxError(first_error)
 
 

+ast3_AST: Final[Type[ast3.AST]] = ast3.AST
+ast27_AST: Final[Type[ast27.AST]] = ast27.AST
+
+

 def stringify_ast(
     node: Union[ast.AST, ast3.AST, ast27.AST], depth: int = 0
 ) -> Iterator[str]:
 def stringify_ast(
     node: Union[ast.AST, ast3.AST, ast27.AST], depth: int = 0
 ) -> Iterator[str]:


@@ -157,6 +178,7 @@ def stringify_ast(
 
     yield f"{'  ' * depth}{node.__class__.__name__}("
 
 
     yield f"{'  ' * depth}{node.__class__.__name__}("
 

+    type_ignore_classes: Tuple[Type[Any], ...]

     for field in sorted(node._fields):  # noqa: F402
         # TypeIgnore will not be present using pypy < 3.8, so need for this
         if not (_IS_PYPY and sys.version_info < (3, 8)):
     for field in sorted(node._fields):  # noqa: F402
         # TypeIgnore will not be present using pypy < 3.8, so need for this
         if not (_IS_PYPY and sys.version_info < (3, 8)):


@@ -189,7 +211,13 @@ def stringify_ast(
                 elif isinstance(item, (ast.AST, ast3.AST, ast27.AST)):
                     yield from stringify_ast(item, depth + 2)
 
                 elif isinstance(item, (ast.AST, ast3.AST, ast27.AST)):
                     yield from stringify_ast(item, depth + 2)
 

-        elif isinstance(value, (ast.AST, ast3.AST, ast27.AST)):
+        # Note that we are referencing the typed-ast ASTs via global variables and not
+        # direct module attribute accesses because that breaks mypyc. It's probably
+        # something to do with the ast3 / ast27 variables being marked as Any leading
+        # mypy to think this branch is always taken, leaving the rest of the code
+        # unanalyzed. Tighting up the types for the typed-ast AST types avoids the
+        # mypyc crash.
+        elif isinstance(value, (ast.AST, ast3_AST, ast27_AST)):

             yield from stringify_ast(value, depth + 2)
 
         else:
             yield from stringify_ast(value, depth + 2)
 
         else: