]> git.madduck.net Git - etc/vim.git/blobdiff - src/black/parsing.py

madduck's git repository

Every one of the projects in this repository is available at the canonical URL git://git.madduck.net/madduck/pub/<projectpath> — see each project's metadata for the exact URL.

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.

SSH access, as well as push access can be individually arranged.

If you use my repositories frequently, consider adding the following snippet to ~/.gitconfig and using the third clone URL listed for each project:

[url "git://git.madduck.net/madduck/"]
  insteadOf = madduck:

Use parentheses on method access on float and int literals (#2799)
[etc/vim.git] / src / black / parsing.py
index 504e20be0022d890759e2f3f081987b60131e349..db48ae4baf5cf181b7647609603c6f75d6b2d523 100644 (file)
@@ -17,17 +17,17 @@ 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
 
 ast3: Any
 
 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:
-    from typed_ast import ast3, ast27
+    from typed_ast import ast3
 except ImportError:
     # Either our python version is too low, or we're on pypy
     if sys.version_info < (3, 7) or (sys.version_info < (3, 8) and not _IS_PYPY):
 except ImportError:
     # Either our python version is too low, or we're on pypy
     if sys.version_info < (3, 7) or (sys.version_info < (3, 8) and not _IS_PYPY):
@@ -39,7 +39,10 @@ except ImportError:
         )
         sys.exit(1)
     else:
         )
         sys.exit(1)
     else:
-        ast3 = ast27 = ast
+        ast3 = ast
+
+
+PY2_HINT: Final = "Python 2 support was removed in version 22.0."
 
 
 class InvalidInput(ValueError):
 
 
 class InvalidInput(ValueError):
@@ -54,35 +57,26 @@ def get_grammars(target_versions: Set[TargetVersion]) -> List[Grammar]:
             pygram.python_grammar_no_print_statement_no_exec_statement_async_keywords,
             # Python 3.0-3.6
             pygram.python_grammar_no_print_statement_no_exec_statement,
             pygram.python_grammar_no_print_statement_no_exec_statement_async_keywords,
             # Python 3.0-3.6
             pygram.python_grammar_no_print_statement_no_exec_statement,
-            # Python 2.7 with future print_function import
-            pygram.python_grammar_no_print_statement,
-            # Python 2.7
-            pygram.python_grammar,
-        ]
-
-    if all(version.is_python2() for version in target_versions):
-        # Python 2-only code, so try Python 2 grammars.
-        return [
-            # Python 2.7 with future print_function import
-            pygram.python_grammar_no_print_statement,
-            # Python 2.7
-            pygram.python_grammar,
+            # Python 3.10+
+            pygram.python_grammar_soft_keywords,
         ]
 
         ]
 
-    # Python 3-compatible code, so only try Python 3 grammar.
     grammars = []
     grammars = []
-    if supports_feature(target_versions, Feature.PATTERN_MATCHING):
-        # 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 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
         )
     if not supports_feature(target_versions, Feature.ASYNC_KEYWORDS):
         # Python 3.0-3.6
         grammars.append(pygram.python_grammar_no_print_statement_no_exec_statement)
         grammars.append(
             pygram.python_grammar_no_print_statement_no_exec_statement_async_keywords
         )
     if not supports_feature(target_versions, Feature.ASYNC_KEYWORDS):
         # Python 3.0-3.6
         grammars.append(pygram.python_grammar_no_print_statement_no_exec_statement)
+    if supports_feature(target_versions, Feature.PATTERN_MATCHING):
+        # Python 3.10+
+        grammars.append(pygram.python_grammar_soft_keywords)
+
     # At least one of the above branches must have been taken, because every Python
     # version has exactly one of the two 'ASYNC_*' flags
     return grammars
     # At least one of the above branches must have been taken, because every Python
     # version has exactly one of the two 'ASYNC_*' flags
     return grammars
@@ -93,7 +87,9 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -
     if not src_txt.endswith("\n"):
         src_txt += "\n"
 
     if not src_txt.endswith("\n"):
         src_txt += "\n"
 
-    for grammar in get_grammars(set(target_versions)):
+    grammars = get_grammars(set(target_versions))
+    errors = {}
+    for grammar in grammars:
         drv = driver.Driver(grammar)
         try:
             result = drv.parse_string(src_txt, True)
         drv = driver.Driver(grammar)
         try:
             result = drv.parse_string(src_txt, True)
@@ -106,8 +102,29 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -
                 faulty_line = lines[lineno - 1]
             except IndexError:
                 faulty_line = "<line number missing in source>"
                 faulty_line = lines[lineno - 1]
             except IndexError:
                 faulty_line = "<line number missing in source>"
-            exc = InvalidInput(f"Cannot parse: {lineno}:{column}: {faulty_line}")
+            errors[grammar.version] = 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]
+            errors[grammar.version] = InvalidInput(
+                f"Cannot parse: {lineno}:{column}: {te.args[0]}"
+            )
+
     else:
     else:
+        # Choose the latest version when raising the actual parsing error.
+        assert len(errors) >= 1
+        exc = errors[max(errors)]
+
+        if matches_grammar(src_txt, pygram.python_grammar) or matches_grammar(
+            src_txt, pygram.python_grammar_no_print_statement
+        ):
+            original_msg = exc.args[0]
+            msg = f"{original_msg}\n{PY2_HINT}"
+            raise InvalidInput(msg) from None
+
         raise exc from None
 
     if isinstance(result, Leaf):
         raise exc from None
 
     if isinstance(result, Leaf):
@@ -115,6 +132,16 @@ def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -
     return result
 
 
     return result
 
 
+def matches_grammar(src_txt: str, grammar: Grammar) -> bool:
+    drv = driver.Driver(grammar)
+    try:
+        drv.parse_string(src_txt, True)
+    except (ParseError, TokenError, IndentationError):
+        return False
+    else:
+        return True
+
+
 def lib2to3_unparse(node: Node) -> str:
     """Given a lib2to3 node, return its string representation."""
     code = str(node)
 def lib2to3_unparse(node: Node) -> str:
     """Given a lib2to3 node, return its string representation."""
     code = str(node)
@@ -123,7 +150,7 @@ def lib2to3_unparse(node: Node) -> str:
 
 def parse_single_version(
     src: str, version: Tuple[int, int]
 
 def parse_single_version(
     src: str, version: Tuple[int, int]
-) -> Union[ast.AST, ast3.AST, ast27.AST]:
+) -> Union[ast.AST, ast3.AST]:
     filename = "<unknown>"
     # typed_ast is needed because of feature version limitations in the builtin ast
     if sys.version_info >= (3, 8) and version >= (3,):
     filename = "<unknown>"
     # typed_ast is needed because of feature version limitations in the builtin ast
     if sys.version_info >= (3, 8) and version >= (3,):
@@ -133,18 +160,13 @@ def parse_single_version(
             return ast3.parse(src, filename)
         else:
             return ast3.parse(src, filename, feature_version=version[1])
             return ast3.parse(src, filename)
         else:
             return ast3.parse(src, filename, feature_version=version[1])
-    elif version == (2, 7):
-        return ast27.parse(src)
     raise AssertionError("INTERNAL ERROR: Tried parsing unsupported Python version!")
 
 
     raise AssertionError("INTERNAL ERROR: Tried parsing unsupported Python version!")
 
 
-def parse_ast(src: str) -> Union[ast.AST, ast3.AST, ast27.AST]:
+def parse_ast(src: str) -> Union[ast.AST, ast3.AST]:
     # TODO: support Python 4+ ;)
     versions = [(3, minor) for minor in range(3, sys.version_info[1] + 1)]
 
     # TODO: support Python 4+ ;)
     versions = [(3, minor) for minor in range(3, sys.version_info[1] + 1)]
 
-    if ast27.__name__ != "ast":
-        versions.append((2, 7))
-
     first_error = ""
     for version in sorted(versions, reverse=True):
         try:
     first_error = ""
     for version in sorted(versions, reverse=True):
         try:
@@ -157,30 +179,38 @@ def parse_ast(src: str) -> Union[ast.AST, ast3.AST, ast27.AST]:
 
 
 ast3_AST: Final[Type[ast3.AST]] = ast3.AST
 
 
 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 _normalize(lineend: str, value: str) -> str:
+    # To normalize, we strip any leading and trailing space from
+    # each line...
+    stripped: List[str] = [i.strip() for i in value.splitlines()]
+    normalized = lineend.join(stripped)
+    # ...and remove any blank lines at the beginning and end of
+    # the whole string
+    return normalized.strip()
+
+
+def stringify_ast(node: Union[ast.AST, ast3.AST], depth: int = 0) -> Iterator[str]:
     """Simple visitor generating strings to compare ASTs by content."""
 
     node = fixup_ast_constants(node)
 
     yield f"{'  ' * depth}{node.__class__.__name__}("
 
     """Simple visitor generating strings to compare ASTs by content."""
 
     node = fixup_ast_constants(node)
 
     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)):
             # TypeIgnore has only one field 'lineno' which breaks this comparison
     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)):
             # TypeIgnore has only one field 'lineno' which breaks this comparison
-            type_ignore_classes = (ast3.TypeIgnore, ast27.TypeIgnore)
+            type_ignore_classes = (ast3.TypeIgnore,)
             if sys.version_info >= (3, 8):
                 type_ignore_classes += (ast.TypeIgnore,)
             if isinstance(node, type_ignore_classes):
                 break
 
         try:
             if sys.version_info >= (3, 8):
                 type_ignore_classes += (ast.TypeIgnore,)
             if isinstance(node, type_ignore_classes):
                 break
 
         try:
-            value = getattr(node, field)
+            value: object = getattr(node, field)
         except AttributeError:
             continue
 
         except AttributeError:
             continue
 
@@ -192,44 +222,35 @@ def stringify_ast(
                 # parentheses and they change the AST.
                 if (
                     field == "targets"
                 # parentheses and they change the AST.
                 if (
                     field == "targets"
-                    and isinstance(node, (ast.Delete, ast3.Delete, ast27.Delete))
-                    and isinstance(item, (ast.Tuple, ast3.Tuple, ast27.Tuple))
+                    and isinstance(node, (ast.Delete, ast3.Delete))
+                    and isinstance(item, (ast.Tuple, ast3.Tuple))
                 ):
                     for item in item.elts:
                         yield from stringify_ast(item, depth + 2)
 
                 ):
                     for item in item.elts:
                         yield from stringify_ast(item, depth + 2)
 
-                elif isinstance(item, (ast.AST, ast3.AST, ast27.AST)):
+                elif isinstance(item, (ast.AST, ast3.AST)):
                     yield from stringify_ast(item, depth + 2)
 
         # 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
                     yield from stringify_ast(item, depth + 2)
 
         # 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
+        # something to do with the ast3 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.
         # 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)):
+        elif isinstance(value, (ast.AST, ast3_AST)):
             yield from stringify_ast(value, depth + 2)
 
         else:
             yield from stringify_ast(value, depth + 2)
 
         else:
+            normalized: object
             # Constant strings may be indented across newlines, if they are
             # docstrings; fold spaces after newlines when comparing. Similarly,
             # trailing and leading space may be removed.
             # Constant strings may be indented across newlines, if they are
             # docstrings; fold spaces after newlines when comparing. Similarly,
             # trailing and leading space may be removed.
-            # Note that when formatting Python 2 code, at least with Windows
-            # line-endings, docstrings can end up here as bytes instead of
-            # str so make sure that we handle both cases.
             if (
                 isinstance(node, ast.Constant)
                 and field == "value"
             if (
                 isinstance(node, ast.Constant)
                 and field == "value"
-                and isinstance(value, (str, bytes))
+                and isinstance(value, str)
             ):
             ):
-                lineend = "\n" if isinstance(value, str) else b"\n"
-                # To normalize, we strip any leading and trailing space from
-                # each line...
-                stripped = [line.strip() for line in value.splitlines()]
-                normalized = lineend.join(stripped)  # type: ignore[attr-defined]
-                # ...and remove any blank lines at the beginning and end of
-                # the whole string
-                normalized = normalized.strip()
+                normalized = _normalize("\n", value)
             else:
                 normalized = value
             yield f"{'  ' * (depth+2)}{normalized!r},  # {value.__class__.__name__}"
             else:
                 normalized = value
             yield f"{'  ' * (depth+2)}{normalized!r},  # {value.__class__.__name__}"
@@ -237,14 +258,12 @@ def stringify_ast(
     yield f"{'  ' * depth})  # /{node.__class__.__name__}"
 
 
     yield f"{'  ' * depth})  # /{node.__class__.__name__}"
 
 
-def fixup_ast_constants(
-    node: Union[ast.AST, ast3.AST, ast27.AST]
-) -> Union[ast.AST, ast3.AST, ast27.AST]:
+def fixup_ast_constants(node: Union[ast.AST, ast3.AST]) -> Union[ast.AST, ast3.AST]:
     """Map ast nodes deprecated in 3.8 to Constant."""
     """Map ast nodes deprecated in 3.8 to Constant."""
-    if isinstance(node, (ast.Str, ast3.Str, ast27.Str, ast.Bytes, ast3.Bytes)):
+    if isinstance(node, (ast.Str, ast3.Str, ast.Bytes, ast3.Bytes)):
         return ast.Constant(value=node.s)
 
         return ast.Constant(value=node.s)
 
-    if isinstance(node, (ast.Num, ast3.Num, ast27.Num)):
+    if isinstance(node, (ast.Num, ast3.Num)):
         return ast.Constant(value=node.n)
 
     if isinstance(node, (ast.NameConstant, ast3.NameConstant)):
         return ast.Constant(value=node.n)
 
     if isinstance(node, (ast.NameConstant, ast3.NameConstant)):