import asyncio
-import pickle
from asyncio.base_events import BaseEventLoop
from concurrent.futures import Executor, ProcessPoolExecutor
+from datetime import datetime
from enum import Enum, Flag
from functools import partial, wraps
+import io
import keyword
import logging
from multiprocessing import Manager
import os
from pathlib import Path
+import pickle
import re
-import tokenize
import signal
import sys
+import tokenize
from typing import (
Any,
Callable,
# types
FileContent = str
Encoding = str
+NewLine = str
Depth = int
NodeType = int
LeafID = int
py36=py36, pyi=pyi, skip_string_normalization=skip_string_normalization
)
report = Report(check=check, quiet=quiet, verbose=verbose)
- sources: List[Path] = []
+ sources: Set[Path] = set()
try:
include_regex = re.compile(include)
except re.error:
for s in src:
p = Path(s)
if p.is_dir():
- sources.extend(
+ sources.update(
gen_python_files_in_dir(p, root, include_regex, exclude_regex, report)
)
elif p.is_file() or s == "-":
# if a file was explicitly given, we don't care about its extension
- sources.append(p)
+ sources.add(p)
else:
err(f"invalid path: {s}")
if len(sources) == 0:
- out("No paths given. Nothing to do 😴")
+ if verbose or not quiet:
+ out("No paths given. Nothing to do 😴")
ctx.exit(0)
return
elif len(sources) == 1:
reformat_one(
- src=sources[0],
+ src=sources.pop(),
line_length=line_length,
fast=fast,
write_back=write_back,
)
finally:
shutdown(loop)
- if not quiet:
- out("All done! ✨ 🍰 ✨")
- click.echo(str(report))
+ if verbose or not quiet:
+ out("All done! ✨ 🍰 ✨")
+ click.echo(str(report))
ctx.exit(report.return_code)
async def schedule_formatting(
- sources: List[Path],
+ sources: Set[Path],
line_length: int,
fast: bool,
write_back: WriteBack,
if write_back != WriteBack.DIFF:
cache = read_cache(line_length, mode)
sources, cached = filter_cached(cache, sources)
- for src in cached:
+ for src in sorted(cached):
report.done(src, Changed.CACHED)
cancelled = []
formatted = []
"""
if src.suffix == ".pyi":
mode |= FileMode.PYI
- with tokenize.open(src) as src_buffer:
- src_contents = src_buffer.read()
+
+ then = datetime.utcfromtimestamp(src.stat().st_mtime)
+ with open(src, "rb") as buf:
+ src_contents, encoding, newline = decode_bytes(buf.read())
try:
dst_contents = format_file_contents(
src_contents, line_length=line_length, fast=fast, mode=mode
return False
if write_back == write_back.YES:
- with open(src, "w", encoding=src_buffer.encoding) as f:
+ with open(src, "w", encoding=encoding, newline=newline) as f:
f.write(dst_contents)
elif write_back == write_back.DIFF:
- src_name = f"{src} (original)"
- dst_name = f"{src} (formatted)"
+ now = datetime.utcnow()
+ src_name = f"{src}\t{then} +0000"
+ dst_name = f"{src}\t{now} +0000"
diff_contents = diff(src_contents, dst_contents, src_name, dst_name)
if lock:
lock.acquire()
try:
- sys.stdout.write(diff_contents)
+ f = io.TextIOWrapper(
+ sys.stdout.buffer,
+ encoding=encoding,
+ newline=newline,
+ write_through=True,
+ )
+ f.write(diff_contents)
+ f.detach()
finally:
if lock:
lock.release()
`line_length`, `fast`, `is_pyi`, and `force_py36` arguments are passed to
:func:`format_file_contents`.
"""
- src = sys.stdin.read()
+ then = datetime.utcnow()
+ src, encoding, newline = decode_bytes(sys.stdin.buffer.read())
dst = src
try:
dst = format_file_contents(src, line_length=line_length, fast=fast, mode=mode)
return False
finally:
+ f = io.TextIOWrapper(
+ sys.stdout.buffer, encoding=encoding, newline=newline, write_through=True
+ )
if write_back == WriteBack.YES:
- sys.stdout.write(dst)
+ f.write(dst)
elif write_back == WriteBack.DIFF:
- src_name = "<stdin> (original)"
- dst_name = "<stdin> (formatted)"
- sys.stdout.write(diff(src, dst, src_name, dst_name))
+ now = datetime.utcnow()
+ src_name = f"STDIN\t{then} +0000"
+ dst_name = f"STDOUT\t{now} +0000"
+ f.write(diff(src, dst, src_name, dst_name))
+ f.detach()
def format_file_contents(
return dst_contents
+def decode_bytes(src: bytes) -> Tuple[FileContent, Encoding, NewLine]:
+ """Return a tuple of (decoded_contents, encoding, newline).
+
+ `newline` is either CRLF or LF but `decoded_contents` is decoded with
+ universal newlines (i.e. only contains LF).
+ """
+ srcbuf = io.BytesIO(src)
+ encoding, lines = tokenize.detect_encoding(srcbuf.readline)
+ if not lines:
+ return "", encoding, "\n"
+
+ newline = "\r\n" if b"\r\n" == lines[0][-2:] else "\n"
+ srcbuf.seek(0)
+ with io.TextIOWrapper(srcbuf, encoding) as tiow:
+ return tiow.read(), encoding, newline
+
+
GRAMMARS = [
pygram.python_grammar_no_print_statement_no_exec_statement,
pygram.python_grammar_no_print_statement,
def lib2to3_parse(src_txt: str) -> Node:
"""Given a string with source, return the lib2to3 Node."""
grammar = pygram.python_grammar_no_print_statement
- if src_txt[-1] != "\n":
- nl = "\r\n" if "\r\n" in src_txt[:1024] else "\n"
- src_txt += nl
+ if src_txt[-1:] != "\n":
+ src_txt += "\n"
for grammar in GRAMMARS:
drv = driver.Driver(grammar, pytree.convert)
try:
syms.dictsetmaker,
syms.listmaker,
syms.testlist_gexp,
+ syms.testlist_star_expr,
}
TEST_DESCENDANTS = {
syms.test,
result.append(leaf, preformatted=True)
for comment_after in line.comments_after(leaf):
result.append(comment_after, preformatted=True)
- bracket_split_succeeded_or_raise(head, body, tail)
assert opening_bracket and closing_bracket
+ body.should_explode = should_explode(body, opening_bracket)
+ bracket_split_succeeded_or_raise(head, body, tail)
if (
+ # the body shouldn't be exploded
+ not body.should_explode
# the opening bracket is an optional paren
- opening_bracket.type == token.LPAR
+ and opening_bracket.type == token.LPAR
and not opening_bracket.value
# the closing bracket is an optional paren
and closing_bracket.type == token.RPAR
yield from right_hand_split(line, line_length, py36=py36, omit=omit)
return
except CannotSplit:
- pass
+ if len(body.leaves) == 1 and not is_line_short_enough(
+ body, line_length=line_length
+ ):
+ raise CannotSplit(
+ "Splitting failed, body is still too long and can't be split."
+ )
ensure_visible(opening_bracket)
ensure_visible(closing_bracket)
- body.should_explode = should_explode(body, opening_bracket)
for result in (head, body, tail):
if result:
yield result
"""
assert root.is_absolute(), f"INTERNAL ERROR: `root` must be absolute but is {root}"
for child in path.iterdir():
- normalized_path = child.resolve().relative_to(root).as_posix()
+ normalized_path = "/" + child.resolve().relative_to(root).as_posix()
if child.is_dir():
normalized_path += "/"
exclude_match = exclude.search(normalized_path)
return stat.st_mtime, stat.st_size
-def filter_cached(
- cache: Cache, sources: Iterable[Path]
-) -> Tuple[List[Path], List[Path]]:
- """Split a list of paths into two.
+def filter_cached(cache: Cache, sources: Iterable[Path]) -> Tuple[Set[Path], Set[Path]]:
+ """Split an iterable of paths in `sources` into two sets.
- The first list contains paths of files that modified on disk or are not in the
- cache. The other list contains paths to non-modified files.
+ The first contains paths of files that modified on disk or are not in the
+ cache. The other contains paths to non-modified files.
"""
- todo, done = [], []
+ todo, done = set(), set()
for src in sources:
src = src.resolve()
if cache.get(src) != get_cache_info(src):
- todo.append(src)
+ todo.add(src)
else:
- done.append(src)
+ done.add(src)
return todo, done
def write_cache(
- cache: Cache, sources: List[Path], line_length: int, mode: FileMode
+ cache: Cache, sources: Iterable[Path], line_length: int, mode: FileMode
) -> None:
"""Update the cache file."""
cache_file = get_cache_file(line_length, mode)
projects / etc / vim.git / blobdiff