parsing.py

"""
Parse Python code and perform AST validation.
"""
import ast
import platform
import sys
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

from black.mode import VERSION_TO_FEATURES, Feature, TargetVersion, supports_feature
from black.nodes import syms
from blib2to3 import pygram
from blib2to3.pgen2 import driver
from blib2to3.pgen2.grammar import Grammar
from blib2to3.pgen2.parse import ParseError
from blib2to3.pgen2.tokenize import TokenError
from blib2to3.pytree import Leaf, Node

ast3: Any

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

try:
    from typed_ast import ast3
except ImportError:
    if sys.version_info < (3, 8) and not _IS_PYPY:
        print(
            "The typed_ast package is required but not installed.\n"
            "You can upgrade to Python 3.8+ or install typed_ast with\n"
            "`python3 -m pip install typed-ast`.",
            file=sys.stderr,
        )
        sys.exit(1)
    else:
        ast3 = ast


PY2_HINT: Final = "Python 2 support was removed in version 22.0."


class InvalidInput(ValueError):
    """Raised when input source code fails all parse attempts."""


def get_grammars(target_versions: Set[TargetVersion]) -> List[Grammar]:
    if not target_versions:
        # No target_version specified, so try all grammars.
        return [
            # Python 3.7-3.9
            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 3.10+
            pygram.python_grammar_soft_keywords,
        ]

    grammars = []
    # If we have to parse both, try to parse async as a keyword first
    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)
    if any(Feature.PATTERN_MATCHING in VERSION_TO_FEATURES[v] for v in target_versions):
        # 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


def lib2to3_parse(src_txt: str, target_versions: Iterable[TargetVersion] = ()) -> Node:
    """Given a string with source, return the lib2to3 Node."""
    if not src_txt.endswith("\n"):
        src_txt += "\n"

    grammars = get_grammars(set(target_versions))
    errors = {}
    for grammar in grammars:
        drv = driver.Driver(grammar)
        try:
            result = drv.parse_string(src_txt, True)
            break

        except ParseError as pe:
            lineno, column = pe.context[1]
            lines = src_txt.splitlines()
            try:
                faulty_line = lines[lineno - 1]
            except IndexError:
                faulty_line = "<line number missing in source>"
            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:
        # 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):
        result = Node(syms.file_input, [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)
    return code


def parse_single_version(
    src: str, version: Tuple[int, int], *, type_comments: bool
) -> Union[ast.AST, ast3.AST]:
    filename = "<unknown>"
    # typed-ast is needed because of feature version limitations in the builtin ast 3.8>
    if sys.version_info >= (3, 8) and version >= (3,):
        return ast.parse(
            src, filename, feature_version=version, type_comments=type_comments
        )

    if _IS_PYPY:
        # PyPy 3.7 doesn't support type comment tracking which is not ideal, but there's
        # not much we can do as typed-ast won't work either.
        if sys.version_info >= (3, 8):
            return ast3.parse(src, filename, type_comments=type_comments)
        else:
            return ast3.parse(src, filename)
    else:
        if type_comments:
            # Typed-ast is guaranteed to be used here and automatically tracks type
            # comments separately.
            return ast3.parse(src, filename, feature_version=version[1])
        else:
            return ast.parse(src, filename)


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)]

    first_error = ""
    for version in sorted(versions, reverse=True):
        try:
            return parse_single_version(src, version, type_comments=True)
        except SyntaxError as e:
            if not first_error:
                first_error = str(e)

    # Try to parse without type comments
    for version in sorted(versions, reverse=True):
        try:
            return parse_single_version(src, version, type_comments=False)
        except SyntaxError:
            pass

    raise SyntaxError(first_error)


ast3_AST: Final[Type[ast3.AST]] = ast3.AST


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__}("

    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
            type_ignore_classes = (ast3.TypeIgnore,)
            if sys.version_info >= (3, 8):
                type_ignore_classes += (ast.TypeIgnore,)
            if isinstance(node, type_ignore_classes):
                break

        try:
            value: object = getattr(node, field)
        except AttributeError:
            continue

        yield f"{'  ' * (depth+1)}{field}="

        if isinstance(value, list):
            for item in value:
                # Ignore nested tuples within del statements, because we may insert
                # parentheses and they change the AST.
                if (
                    field == "targets"
                    and isinstance(node, (ast.Delete, ast3.Delete))
                    and isinstance(item, (ast.Tuple, ast3.Tuple))
                ):
                    for elt in item.elts:
                        yield from stringify_ast(elt, depth + 2)

                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
        # 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.
        elif isinstance(value, (ast.AST, ast3_AST)):
            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.
            if (
                isinstance(node, ast.Constant)
                and field == "value"
                and isinstance(value, str)
            ):
                normalized = _normalize("\n", value)
            else:
                normalized = value
            yield f"{'  ' * (depth+2)}{normalized!r},  # {value.__class__.__name__}"

    yield f"{'  ' * depth})  # /{node.__class__.__name__}"


def fixup_ast_constants(node: Union[ast.AST, ast3.AST]) -> Union[ast.AST, ast3.AST]:
    """Map ast nodes deprecated in 3.8 to Constant."""
    if isinstance(node, (ast.Str, ast3.Str, ast.Bytes, ast3.Bytes)):
        return ast.Constant(value=node.s)

    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.value)

    return node