]> git.madduck.net Git - etc/vim.git/blobdiff - src/black/__init__.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:

Disable universal newlines when reading TOML (#2408)
[etc/vim.git] / src / black / __init__.py
index ba4d3dea70ea862b0f19390ed40e0fd9ed94800a..d8b98196aa068e9be8c192d71358333dab9377bf 100644 (file)
@@ -10,7 +10,7 @@ from multiprocessing import Manager, freeze_support
 import os
 from pathlib import Path
 from pathspec.patterns.gitwildmatch import GitWildMatchPatternError
-import regex as re
+import re
 import signal
 import sys
 import tokenize
@@ -30,8 +30,9 @@ from typing import (
     Union,
 )
 
-from dataclasses import replace
 import click
+from dataclasses import replace
+from mypy_extensions import mypyc_attr
 
 from black.const import DEFAULT_LINE_LENGTH, DEFAULT_INCLUDES, DEFAULT_EXCLUDES
 from black.const import STDIN_PLACEHOLDER
@@ -39,7 +40,7 @@ from black.nodes import STARS, syms, is_simple_decorator_expression
 from black.lines import Line, EmptyLineTracker
 from black.linegen import transform_line, LineGenerator, LN
 from black.comments import normalize_fmt_off
-from black.mode import Mode, TargetVersion
+from black.mode import FUTURE_FLAG_TO_FEATURE, Mode, TargetVersion
 from black.mode import Feature, supports_feature, VERSION_TO_FEATURES
 from black.cache import read_cache, write_cache, get_cache_info, filter_cached, Cache
 from black.concurrency import cancel, shutdown, maybe_install_uvloop
@@ -56,6 +57,7 @@ from black.handle_ipynb_magics import (
     remove_trailing_semicolon,
     put_trailing_semicolon_back,
     TRANSFORMED_MAGICS,
+    PYTHON_CELL_MAGICS,
     jupyter_dependencies_are_installed,
 )
 
@@ -66,6 +68,8 @@ from blib2to3.pgen2 import token
 
 from _black_version import version as __version__
 
+COMPILED = Path(__file__).suffix in (".pyd", ".so")
+
 # types
 FileContent = str
 Encoding = str
@@ -173,11 +177,16 @@ def validate_regex(
 ) -> Optional[Pattern[str]]:
     try:
         return re_compile_maybe_verbose(value) if value is not None else None
-    except re.error:
-        raise click.BadParameter("Not a valid regular expression") from None
+    except re.error as e:
+        raise click.BadParameter(f"Not a valid regular expression: {e}") from None
 
 
-@click.command(context_settings=dict(help_option_names=["-h", "--help"]))
+@click.command(
+    context_settings={"help_option_names": ["-h", "--help"]},
+    # While Click does set this field automatically using the docstring, mypyc
+    # (annoyingly) strips 'em so we need to set it here too.
+    help="The uncompromising code formatter.",
+)
 @click.option("-c", "--code", type=str, help="Format the code passed in as a string.")
 @click.option(
     "-l",
@@ -346,7 +355,10 @@ def validate_regex(
         " due to exclusion patterns."
     ),
 )
-@click.version_option(version=__version__)
+@click.version_option(
+    version=__version__,
+    message=f"%(prog)s, %(version)s (compiled: {'yes' if COMPILED else 'no'})",
+)
 @click.argument(
     "src",
     nargs=-1,
@@ -387,7 +399,7 @@ def main(
     experimental_string_processing: bool,
     quiet: bool,
     verbose: bool,
-    required_version: str,
+    required_version: Optional[str],
     include: Pattern[str],
     exclude: Optional[Pattern[str]],
     extend_exclude: Optional[Pattern[str]],
@@ -655,6 +667,9 @@ def reformat_one(
         report.failed(src, str(exc))
 
 
+# diff-shades depends on being to monkeypatch this function to operate. I know it's
+# not ideal, but this shouldn't cause any issues ... hopefully. ~ichard26
+@mypyc_attr(patchable=True)
 def reformat_many(
     sources: Set[Path],
     fast: bool,
@@ -669,10 +684,11 @@ def reformat_many(
     worker_count = workers if workers is not None else DEFAULT_WORKERS
     if sys.platform == "win32":
         # Work around https://bugs.python.org/issue26903
+        assert worker_count is not None
         worker_count = min(worker_count, 60)
     try:
         executor = ProcessPoolExecutor(max_workers=worker_count)
-    except (ImportError, OSError):
+    except (ImportError, NotImplementedError, OSError):
         # we arrive here if the underlying system does not support multi-processing
         # like in AWS Lambda or Termux, in which case we gracefully fallback to
         # a ThreadPoolExecutor with just a single worker (more workers would not do us
@@ -928,7 +944,9 @@ def format_file_contents(src_contents: str, *, fast: bool, mode: Mode) -> FileCo
 
 
 def validate_cell(src: str) -> None:
-    """Check that cell does not already contain TransformerManager transformations.
+    """Check that cell does not already contain TransformerManager transformations,
+    or non-Python cell magics, which might cause tokenizer_rt to break because of
+    indentations.
 
     If a cell contains ``!ls``, then it'll be transformed to
     ``get_ipython().system('ls')``. However, if the cell originally contained
@@ -944,6 +962,8 @@ def validate_cell(src: str) -> None:
     """
     if any(transformed_magic in src for transformed_magic in TRANSFORMED_MAGICS):
         raise NothingChanged
+    if src[:2] == "%%" and src.split()[0][2:] not in PYTHON_CELL_MAGICS:
+        raise NothingChanged
 
 
 def format_cell(src: str, *, fast: bool, mode: Mode) -> str:
@@ -1060,7 +1080,7 @@ def format_str(src_contents: str, *, mode: Mode) -> FileContent:
     if mode.target_versions:
         versions = mode.target_versions
     else:
-        versions = detect_target_versions(src_node)
+        versions = detect_target_versions(src_node, future_imports=future_imports)
 
     # TODO: fully drop support and this code hopefully in January 2022 :D
     if TargetVersion.PY27 in mode.target_versions or versions == {TargetVersion.PY27}:
@@ -1112,7 +1132,9 @@ def decode_bytes(src: bytes) -> Tuple[FileContent, Encoding, NewLine]:
         return tiow.read(), encoding, newline
 
 
-def get_features_used(node: Node) -> Set[Feature]:  # noqa: C901
+def get_features_used(  # noqa: C901
+    node: Node, *, future_imports: Optional[Set[str]] = None
+) -> Set[Feature]:
     """Return a set of (relatively) new Python features used in this file.
 
     Currently looking for:
@@ -1122,9 +1144,17 @@ def get_features_used(node: Node) -> Set[Feature]:  # noqa: C901
     - positional only arguments in function signatures and lambdas;
     - assignment expression;
     - relaxed decorator syntax;
+    - usage of __future__ flags (annotations);
     - print / exec statements;
     """
     features: Set[Feature] = set()
+    if future_imports:
+        features |= {
+            FUTURE_FLAG_TO_FEATURE[future_import]
+            for future_import in future_imports
+            if future_import in FUTURE_FLAG_TO_FEATURE
+        }
+
     for n in node.pre_order():
         if n.type == token.STRING:
             value_head = n.value[:2]  # type: ignore
@@ -1132,8 +1162,17 @@ def get_features_used(node: Node) -> Set[Feature]:  # noqa: C901
                 features.add(Feature.F_STRINGS)
 
         elif n.type == token.NUMBER:
-            if "_" in n.value:  # type: ignore
+            assert isinstance(n, Leaf)
+            if "_" in n.value:
                 features.add(Feature.NUMERIC_UNDERSCORES)
+            elif n.value.endswith(("L", "l")):
+                # Python 2: 10L
+                features.add(Feature.LONG_INT_LITERAL)
+            elif len(n.value) >= 2 and n.value[0] == "0" and n.value[1].isdigit():
+                # Python 2: 0123; 00123; ...
+                if not all(char == "0" for char in n.value):
+                    # although we don't want to match 0000 or similar
+                    features.add(Feature.OCTAL_INT_LITERAL)
 
         elif n.type == token.SLASH:
             if n.parent and n.parent.type in {
@@ -1171,17 +1210,55 @@ def get_features_used(node: Node) -> Set[Feature]:  # noqa: C901
                         if argch.type in STARS:
                             features.add(feature)
 
-        elif n.type == token.PRINT_STMT:
+        elif (
+            n.type in {syms.return_stmt, syms.yield_expr}
+            and len(n.children) >= 2
+            and n.children[1].type == syms.testlist_star_expr
+            and any(child.type == syms.star_expr for child in n.children[1].children)
+        ):
+            features.add(Feature.UNPACKING_ON_FLOW)
+
+        elif (
+            n.type == syms.annassign
+            and len(n.children) >= 4
+            and n.children[3].type == syms.testlist_star_expr
+        ):
+            features.add(Feature.ANN_ASSIGN_EXTENDED_RHS)
+
+        # Python 2 only features (for its deprecation) except for integers, see above
+        elif n.type == syms.print_stmt:
             features.add(Feature.PRINT_STMT)
-        elif n.type == token.EXEC_STMT:
+        elif n.type == syms.exec_stmt:
             features.add(Feature.EXEC_STMT)
+        elif n.type == syms.tfpdef:
+            # def set_position((x, y), value):
+            #     ...
+            features.add(Feature.AUTOMATIC_PARAMETER_UNPACKING)
+        elif n.type == syms.except_clause:
+            # try:
+            #     ...
+            # except Exception, err:
+            #     ...
+            if len(n.children) >= 4:
+                if n.children[-2].type == token.COMMA:
+                    features.add(Feature.COMMA_STYLE_EXCEPT)
+        elif n.type == syms.raise_stmt:
+            # raise Exception, "msg"
+            if len(n.children) >= 4:
+                if n.children[-2].type == token.COMMA:
+                    features.add(Feature.COMMA_STYLE_RAISE)
+        elif n.type == token.BACKQUOTE:
+            # `i'm surprised this ever existed`
+            features.add(Feature.BACKQUOTE_REPR)
 
     return features
 
 
-def detect_target_versions(node: Node) -> Set[TargetVersion]:
+def detect_target_versions(
+    node: Node, *, future_imports: Optional[Set[str]] = None
+) -> Set[TargetVersion]:
     """Detect the version to target based on the nodes used."""
-    features = get_features_used(node)
+    features = get_features_used(node, future_imports=future_imports)
     return {
         version for version in TargetVersion if features <= VERSION_TO_FEATURES[version]
     }
@@ -1243,7 +1320,7 @@ def assert_equivalent(src: str, dst: str, *, pass_num: int = 1) -> None:
         src_ast = parse_ast(src)
     except Exception as exc:
         raise AssertionError(
-            "cannot use --safe with this file; failed to parse source file."
+            f"cannot use --safe with this file; failed to parse source file: {exc}"
         ) from exc
 
     try: