]> 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:

Bump regex dependency to 2021.4.4 to fix import of Pattern class (#2621)
[etc/vim.git] / src / black / __init__.py
index 29fb244f8b751651238fd99fefb055a9ada26eb1..a5ddec9122105cae9393d126c92f3c3c5bfb3ec3 100644 (file)
@@ -9,6 +9,7 @@ import io
 from multiprocessing import Manager, freeze_support
 import os
 from pathlib import Path
+from pathspec.patterns.gitwildmatch import GitWildMatchPatternError
 import regex as re
 import signal
 import sys
@@ -29,8 +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
@@ -65,6 +67,8 @@ from blib2to3.pgen2 import token
 
 from _black_version import version as __version__
 
+COMPILED = Path(__file__).suffix in (".pyd", ".so")
+
 # types
 FileContent = str
 Encoding = str
@@ -94,6 +98,8 @@ class WriteBack(Enum):
 # Legacy name, left for integrations.
 FileMode = Mode
 
+DEFAULT_WORKERS = os.cpu_count()
+
 
 def read_pyproject_toml(
     ctx: click.Context, param: click.Parameter, value: Optional[str]
@@ -113,7 +119,7 @@ def read_pyproject_toml(
     except (OSError, ValueError) as e:
         raise click.FileError(
             filename=value, hint=f"Error reading configuration file: {e}"
-        )
+        ) from None
 
     if not config:
         return None
@@ -167,14 +173,19 @@ def validate_regex(
     ctx: click.Context,
     param: click.Parameter,
     value: Optional[str],
-) -> Optional[Pattern]:
+) -> 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")
+        raise click.BadParameter("Not a valid regular expression") from None
 
 
-@click.command(context_settings=dict(help_option_names=["-h", "--help"]))
+@click.command(
+    context_settings=dict(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",
@@ -317,6 +328,14 @@ def validate_regex(
         "editors that rely on using stdin."
     ),
 )
+@click.option(
+    "-W",
+    "--workers",
+    type=click.IntRange(min=1),
+    default=DEFAULT_WORKERS,
+    show_default=True,
+    help="Number of parallel workers",
+)
 @click.option(
     "-q",
     "--quiet",
@@ -335,7 +354,10 @@ 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,
@@ -376,12 +398,13 @@ def main(
     experimental_string_processing: bool,
     quiet: bool,
     verbose: bool,
-    required_version: str,
-    include: Pattern,
-    exclude: Optional[Pattern],
-    extend_exclude: Optional[Pattern],
-    force_exclude: Optional[Pattern],
+    required_version: Optional[str],
+    include: Pattern[str],
+    exclude: Optional[Pattern[str]],
+    extend_exclude: Optional[Pattern[str]],
+    force_exclude: Optional[Pattern[str]],
     stdin_filename: Optional[str],
+    workers: int,
     src: Tuple[str, ...],
     config: Optional[str],
 ) -> None:
@@ -428,18 +451,21 @@ def main(
             content=code, fast=fast, write_back=write_back, mode=mode, report=report
         )
     else:
-        sources = get_sources(
-            ctx=ctx,
-            src=src,
-            quiet=quiet,
-            verbose=verbose,
-            include=include,
-            exclude=exclude,
-            extend_exclude=extend_exclude,
-            force_exclude=force_exclude,
-            report=report,
-            stdin_filename=stdin_filename,
-        )
+        try:
+            sources = get_sources(
+                ctx=ctx,
+                src=src,
+                quiet=quiet,
+                verbose=verbose,
+                include=include,
+                exclude=exclude,
+                extend_exclude=extend_exclude,
+                force_exclude=force_exclude,
+                report=report,
+                stdin_filename=stdin_filename,
+            )
+        except GitWildMatchPatternError:
+            ctx.exit(1)
 
         path_empty(
             sources,
@@ -464,6 +490,7 @@ def main(
                 write_back=write_back,
                 mode=mode,
                 report=report,
+                workers=workers,
             )
 
     if verbose or not quiet:
@@ -639,15 +666,24 @@ 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, write_back: WriteBack, mode: Mode, report: "Report"
+    sources: Set[Path],
+    fast: bool,
+    write_back: WriteBack,
+    mode: Mode,
+    report: "Report",
+    workers: Optional[int],
 ) -> None:
     """Reformat multiple files using a ProcessPoolExecutor."""
     executor: Executor
     loop = asyncio.get_event_loop()
-    worker_count = os.cpu_count()
+    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)
@@ -742,7 +778,10 @@ async def schedule_formatting(
                     sources_to_cache.append(src)
                 report.done(src, changed)
     if cancelled:
-        await asyncio.gather(*cancelled, loop=loop, return_exceptions=True)
+        if sys.version_info >= (3, 7):
+            await asyncio.gather(*cancelled, return_exceptions=True)
+        else:
+            await asyncio.gather(*cancelled, loop=loop, return_exceptions=True)
     if sources_to_cache:
         write_cache(cache, sources_to_cache, mode)
 
@@ -773,7 +812,9 @@ def format_file_in_place(
     except NothingChanged:
         return False
     except JSONDecodeError:
-        raise ValueError(f"File '{src}' cannot be parsed as valid Jupyter notebook.")
+        raise ValueError(
+            f"File '{src}' cannot be parsed as valid Jupyter notebook."
+        ) from None
 
     if write_back == WriteBack.YES:
         with open(src, "w", encoding=encoding, newline=newline) as f:
@@ -943,7 +984,7 @@ def format_cell(src: str, *, fast: bool, mode: Mode) -> str:
     try:
         masked_src, replacements = mask_cell(src_without_trailing_semicolon)
     except SyntaxError:
-        raise NothingChanged
+        raise NothingChanged from None
     masked_dst = format_str(masked_src, mode=mode)
     if not fast:
         check_stability_and_equivalence(masked_src, masked_dst, mode=mode)
@@ -953,7 +994,7 @@ def format_cell(src: str, *, fast: bool, mode: Mode) -> str:
     )
     dst = dst.rstrip("\n")
     if dst == src:
-        raise NothingChanged
+        raise NothingChanged from None
     return dst
 
 
@@ -966,14 +1007,14 @@ def validate_metadata(nb: MutableMapping[str, Any]) -> None:
     """
     language = nb.get("metadata", {}).get("language_info", {}).get("name", None)
     if language is not None and language != "python":
-        raise NothingChanged
+        raise NothingChanged from None
 
 
 def format_ipynb_string(src_contents: str, *, fast: bool, mode: Mode) -> FileContent:
     """Format Jupyter notebook.
 
     Operate cell-by-cell, only on code cells, only for Python notebooks.
-    If the ``.ipynb`` originally had a trailing newline, it'll be preseved.
+    If the ``.ipynb`` originally had a trailing newline, it'll be preserved.
     """
     trailing_newline = src_contents[-1] == "\n"
     modified = False
@@ -1035,6 +1076,15 @@ def format_str(src_contents: str, *, mode: Mode) -> FileContent:
         versions = mode.target_versions
     else:
         versions = detect_target_versions(src_node)
+
+    # TODO: fully drop support and this code hopefully in January 2022 :D
+    if TargetVersion.PY27 in mode.target_versions or versions == {TargetVersion.PY27}:
+        msg = (
+            "DEPRECATION: Python 2 support will be removed in the first stable release "
+            "expected in January 2022."
+        )
+        err(msg, fg="yellow", bold=True)
+
     normalize_fmt_off(src_node)
     lines = LineGenerator(
         mode=mode,
@@ -1077,7 +1127,7 @@ def decode_bytes(src: bytes) -> Tuple[FileContent, Encoding, NewLine]:
         return tiow.read(), encoding, newline
 
 
-def get_features_used(node: Node) -> Set[Feature]:
+def get_features_used(node: Node) -> Set[Feature]:  # noqa: C901
     """Return a set of (relatively) new Python features used in this file.
 
     Currently looking for:
@@ -1087,6 +1137,7 @@ def get_features_used(node: Node) -> Set[Feature]:
     - positional only arguments in function signatures and lambdas;
     - assignment expression;
     - relaxed decorator syntax;
+    - print / exec statements;
     """
     features: Set[Feature] = set()
     for n in node.pre_order():
@@ -1096,11 +1147,24 @@ def get_features_used(node: Node) -> Set[Feature]:
                 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 {syms.typedargslist, syms.arglist}:
+            if n.parent and n.parent.type in {
+                syms.typedargslist,
+                syms.arglist,
+                syms.varargslist,
+            }:
                 features.add(Feature.POS_ONLY_ARGUMENTS)
 
         elif n.type == token.COLONEQUAL:
@@ -1131,6 +1195,32 @@ def get_features_used(node: Node) -> Set[Feature]:
                         if argch.type in STARS:
                             features.add(feature)
 
+        # 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 == 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
 
 
@@ -1198,9 +1288,8 @@ 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.  AST"
-            f" error message: {exc}"
-        )
+            "cannot use --safe with this file; failed to parse source file."
+        ) from exc
 
     try:
         dst_ast = parse_ast(dst)
@@ -1261,7 +1350,7 @@ def patch_click() -> None:
     """
     try:
         from click import core
-        from click import _unicodefun  # type: ignore
+        from click import _unicodefun
     except ModuleNotFoundError:
         return