default=DEFAULT_INCLUDES,
help=(
"A regular expression that matches files and directories that should be "
- "included on recursive searches. On Windows, use forward slashes for "
- "directories."
+ "included on recursive searches. An empty value means all files are "
+ "included regardless of the name. Use forward slashes for directories on "
+ "all platforms (Windows, too). Exclusions are calculated first, inclusions "
+ "later."
),
show_default=True,
)
default=DEFAULT_EXCLUDES,
help=(
"A regular expression that matches files and directories that should be "
- "excluded on recursive searches. On Windows, use forward slashes for "
- "directories."
+ "excluded on recursive searches. An empty value means no paths are excluded. "
+ "Use forward slashes for directories on all platforms (Windows, too). "
+ "Exclusions are calculated first, inclusions later."
),
show_default=True,
)
except re.error:
err(f"Invalid regular expression for exclude given: {exclude!r}")
ctx.exit(2)
+ root = find_project_root(src)
for s in src:
p = Path(s)
if p.is_dir():
- sources.extend(gen_python_files_in_dir(p, include_regex, exclude_regex))
+ sources.extend(
+ gen_python_files_in_dir(p, root, include_regex, exclude_regex)
+ )
elif p.is_file():
# if a file was explicitly given, we don't care about its extension
sources.append(p)
cache: Cache = {}
if write_back != WriteBack.DIFF:
cache = read_cache(line_length, mode)
- src = src.resolve()
- if src in cache and cache[src] == get_cache_info(src):
+ res_src = src.resolve()
+ if res_src in cache and cache[res_src] == get_cache_info(res_src):
changed = Changed.CACHED
if changed is not Changed.CACHED and format_file_in_place(
src,
def gen_python_files_in_dir(
- path: Path, include: Pattern[str], exclude: Pattern[str]
+ path: Path, root: Path, include: Pattern[str], exclude: Pattern[str]
) -> Iterator[Path]:
"""Generate all files under `path` whose paths are not excluded by the
`exclude` regex, but are included by the `include` regex.
"""
-
+ assert root.is_absolute(), f"INTERNAL ERROR: `root` must be absolute but is {root}"
for child in path.iterdir():
- searchable_path = str(child.as_posix())
- if Path(child.parts[0]).is_dir():
- searchable_path = "/" + searchable_path
+ normalized_path = child.resolve().relative_to(root).as_posix()
if child.is_dir():
- searchable_path = searchable_path + "/"
- exclude_match = exclude.search(searchable_path)
- if exclude_match and len(exclude_match.group()) > 0:
- continue
+ normalized_path += "/"
+ exclude_match = exclude.search(normalized_path)
+ if exclude_match and exclude_match.group(0):
+ continue
- yield from gen_python_files_in_dir(child, include, exclude)
+ if child.is_dir():
+ yield from gen_python_files_in_dir(child, root, include, exclude)
- else:
- include_match = include.search(searchable_path)
- exclude_match = exclude.search(searchable_path)
- if (
- child.is_file()
- and include_match
- and len(include_match.group()) > 0
- and (not exclude_match or len(exclude_match.group()) == 0)
- ):
+ elif child.is_file():
+ include_match = include.search(normalized_path)
+ if include_match:
yield child
+def find_project_root(srcs: List[str]) -> Path:
+ """Return a directory containing .git, .hg, or pyproject.toml.
+
+ That directory can be one of the directories passed in `srcs` or their
+ common parent.
+
+ If no directory in the tree contains a marker that would specify it's the
+ project root, the root of the file system is returned.
+ """
+ if not srcs:
+ return Path("/").resolve()
+
+ common_base = min(Path(src).resolve() for src in srcs)
+ if common_base.is_dir():
+ # Append a fake file so `parents` below returns `common_base_dir`, too.
+ common_base /= "fake-file"
+ for directory in common_base.parents:
+ if (directory / ".git").is_dir():
+ return directory
+
+ if (directory / ".hg").is_dir():
+ return directory
+
+ if (directory / "pyproject.toml").is_file():
+ return directory
+
+ return directory
+
+
@dataclass
class Report:
"""Provides a reformatting counter. Can be rendered with `str(report)`."""