]> 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:

Provide a stdin-filename to allow stdin to respect force-exclude rules (#1780)
[etc/vim.git] / src / black / __init__.py
index dce7a24445a668232b1250b41e0b9e9168d0a841..48690573810373ae94123839474c29ea772e9630 100644 (file)
@@ -68,6 +68,7 @@ DEFAULT_LINE_LENGTH = 88
 DEFAULT_EXCLUDES = r"/(\.direnv|\.eggs|\.git|\.hg|\.mypy_cache|\.nox|\.tox|\.venv|\.svn|_build|buck-out|build|dist)/"  # noqa: B950
 DEFAULT_INCLUDES = r"\.pyi?$"
 CACHE_DIR = Path(user_cache_dir("black", version=__version__))
+STDIN_PLACEHOLDER = "__BLACK_STDIN_FILENAME__"
 
 STRING_PREFIX_CHARS: Final = "furbFURB"  # All possible string prefix characters.
 
@@ -457,6 +458,15 @@ def target_version_option_callback(
         "excluded even when they are passed explicitly as arguments."
     ),
 )
+@click.option(
+    "--stdin-filename",
+    type=str,
+    help=(
+        "The name of the file when passing it through stdin. Useful to make "
+        "sure Black will respect --force-exclude option on some "
+        "editors that rely on using stdin."
+    ),
+)
 @click.option(
     "-q",
     "--quiet",
@@ -516,6 +526,7 @@ def main(
     include: str,
     exclude: str,
     force_exclude: Optional[str],
+    stdin_filename: Optional[str],
     src: Tuple[str, ...],
     config: Optional[str],
 ) -> None:
@@ -548,6 +559,7 @@ def main(
         exclude=exclude,
         force_exclude=force_exclude,
         report=report,
+        stdin_filename=stdin_filename,
     )
 
     path_empty(
@@ -587,6 +599,7 @@ def get_sources(
     exclude: str,
     force_exclude: Optional[str],
     report: "Report",
+    stdin_filename: Optional[str],
 ) -> Set[Path]:
     """Compute the set of files to be formatted."""
     try:
@@ -613,22 +626,14 @@ def get_sources(
     gitignore = get_gitignore(root)
 
     for s in src:
-        p = Path(s)
-        if p.is_dir():
-            sources.update(
-                gen_python_files(
-                    p.iterdir(),
-                    root,
-                    include_regex,
-                    exclude_regex,
-                    force_exclude_regex,
-                    report,
-                    gitignore,
-                )
-            )
-        elif s == "-":
-            sources.add(p)
-        elif p.is_file():
+        if s == "-" and stdin_filename:
+            p = Path(stdin_filename)
+            is_stdin = True
+        else:
+            p = Path(s)
+            is_stdin = False
+
+        if is_stdin or p.is_file():
             normalized_path = normalize_path_maybe_ignore(p, root, report)
             if normalized_path is None:
                 continue
@@ -643,6 +648,23 @@ def get_sources(
                 report.path_ignored(p, "matches the --force-exclude regular expression")
                 continue
 
+            if is_stdin:
+                p = Path(f"{STDIN_PLACEHOLDER}{str(p)}")
+
+            sources.add(p)
+        elif p.is_dir():
+            sources.update(
+                gen_python_files(
+                    p.iterdir(),
+                    root,
+                    include_regex,
+                    exclude_regex,
+                    force_exclude_regex,
+                    report,
+                    gitignore,
+                )
+            )
+        elif s == "-":
             sources.add(p)
         else:
             err(f"invalid path: {s}")
@@ -670,7 +692,18 @@ def reformat_one(
     """
     try:
         changed = Changed.NO
-        if not src.is_file() and str(src) == "-":
+
+        if str(src) == "-":
+            is_stdin = True
+        elif str(src).startswith(STDIN_PLACEHOLDER):
+            is_stdin = True
+            # Use the original name again in case we want to print something
+            # to the user
+            src = Path(str(src)[len(STDIN_PLACEHOLDER) :])
+        else:
+            is_stdin = False
+
+        if is_stdin:
             if format_stdin_to_stdout(fast=fast, write_back=write_back, mode=mode):
                 changed = Changed.YES
         else:
@@ -859,9 +892,9 @@ def color_diff(contents: str) -> str:
     for i, line in enumerate(lines):
         if line.startswith("+++") or line.startswith("---"):
             line = "\033[1;37m" + line + "\033[0m"  # bold white, reset
-        if line.startswith("@@"):
+        elif line.startswith("@@"):
             line = "\033[36m" + line + "\033[0m"  # cyan, reset
-        if line.startswith("+"):
+        elif line.startswith("+"):
             line = "\033[32m" + line + "\033[0m"  # green, reset
         elif line.startswith("-"):
             line = "\033[31m" + line + "\033[0m"  # red, reset
@@ -871,30 +904,22 @@ def color_diff(contents: str) -> str:
 
 def wrap_stream_for_windows(
     f: io.TextIOWrapper,
-) -> Union[io.TextIOWrapper, "colorama.AnsiToWin32.AnsiToWin32"]:
+) -> Union[io.TextIOWrapper, "colorama.AnsiToWin32"]:
     """
-    Wrap the stream in colorama's wrap_stream so colors are shown on Windows.
+    Wrap stream with colorama's wrap_stream so colors are shown on Windows.
 
-    If `colorama` is not found, then no change is made. If `colorama` does
-    exist, then it handles the logic to determine whether or not to change
-    things.
+    If `colorama` is unavailable, the original stream is returned unmodified.
+    Otherwise, the `wrap_stream()` function determines whether the stream needs
+    to be wrapped for a Windows environment and will accordingly either return
+    an `AnsiToWin32` wrapper or the original stream.
     """
     try:
-        from colorama import initialise
-
-        # We set `strip=False` so that we can don't have to modify
-        # test_express_diff_with_color.
-        f = initialise.wrap_stream(
-            f, convert=None, strip=False, autoreset=False, wrap=True
-        )
-
-        # wrap_stream returns a `colorama.AnsiToWin32.AnsiToWin32` object
-        # which does not have a `detach()` method. So we fake one.
-        f.detach = lambda *args, **kwargs: None  # type: ignore
+        from colorama.initialise import wrap_stream
     except ImportError:
-        pass
-
-    return f
+        return f
+    else:
+        # Set `strip=False` to avoid needing to modify test_express_diff_with_color.
+        return wrap_stream(f, convert=None, strip=False, autoreset=False, wrap=True)
 
 
 def format_stdin_to_stdout(
@@ -3598,7 +3623,8 @@ class StringSplitter(CustomSplitMapMixin, BaseStringSplitter):
         MIN_SUBSTR_SIZE characters.
 
         The string will ONLY be split on spaces (i.e. each new substring should
-        start with a space).
+        start with a space). Note that the string will NOT be split on a space
+        which is escaped with a backslash.
 
         If the string is an f-string, it will NOT be split in the middle of an
         f-expression (e.g. in f"FooBar: {foo() if x else bar()}", {foo() if x
@@ -3618,13 +3644,14 @@ class StringSplitter(CustomSplitMapMixin, BaseStringSplitter):
     MIN_SUBSTR_SIZE = 6
     # Matches an "f-expression" (e.g. {var}) that might be found in an f-string.
     RE_FEXPR = r"""
-    (?<!\{)\{
+    (?<!\{) (?:\{\{)* \{ (?!\{)
         (?:
             [^\{\}]
             | \{\{
             | \}\}
+            | (?R)
         )+?
-    (?<!\})(?:\}\})*\}(?!\})
+    (?<!\}) \} (?:\}\})* (?!\})
     """
 
     def do_splitter_match(self, line: Line) -> TMatchResult:
@@ -3938,11 +3965,23 @@ class StringSplitter(CustomSplitMapMixin, BaseStringSplitter):
                 section of this classes' docstring would be be met by returning @i.
             """
             is_space = string[i] == " "
+
+            is_not_escaped = True
+            j = i - 1
+            while is_valid_index(j) and string[j] == "\\":
+                is_not_escaped = not is_not_escaped
+                j -= 1
+
             is_big_enough = (
                 len(string[i:]) >= self.MIN_SUBSTR_SIZE
                 and len(string[:i]) >= self.MIN_SUBSTR_SIZE
             )
-            return is_space and is_big_enough and not breaks_fstring_expression(i)
+            return (
+                is_space
+                and is_not_escaped
+                and is_big_enough
+                and not breaks_fstring_expression(i)
+            )
 
         # First, we check all indices BELOW @max_break_idx.
         break_idx = max_break_idx
@@ -5186,31 +5225,52 @@ def normalize_numeric_literal(leaf: Leaf) -> None:
         # Leave octal and binary literals alone.
         pass
     elif text.startswith("0x"):
-        # Change hex literals to upper case.
-        before, after = text[:2], text[2:]
-        text = f"{before}{after.upper()}"
+        text = format_hex(text)
     elif "e" in text:
-        before, after = text.split("e")
-        sign = ""
-        if after.startswith("-"):
-            after = after[1:]
-            sign = "-"
-        elif after.startswith("+"):
-            after = after[1:]
-        before = format_float_or_int_string(before)
-        text = f"{before}e{sign}{after}"
+        text = format_scientific_notation(text)
     elif text.endswith(("j", "l")):
-        number = text[:-1]
-        suffix = text[-1]
-        # Capitalize in "2L" because "l" looks too similar to "1".
-        if suffix == "l":
-            suffix = "L"
-        text = f"{format_float_or_int_string(number)}{suffix}"
+        text = format_long_or_complex_number(text)
     else:
         text = format_float_or_int_string(text)
     leaf.value = text
 
 
+def format_hex(text: str) -> str:
+    """
+    Formats a hexadecimal string like "0x12b3"
+
+    Uses lowercase because of similarity between "B" and "8", which
+    can cause security issues.
+    see: https://github.com/psf/black/issues/1692
+    """
+
+    before, after = text[:2], text[2:]
+    return f"{before}{after.lower()}"
+
+
+def format_scientific_notation(text: str) -> str:
+    """Formats a numeric string utilizing scentific notation"""
+    before, after = text.split("e")
+    sign = ""
+    if after.startswith("-"):
+        after = after[1:]
+        sign = "-"
+    elif after.startswith("+"):
+        after = after[1:]
+    before = format_float_or_int_string(before)
+    return f"{before}e{sign}{after}"
+
+
+def format_long_or_complex_number(text: str) -> str:
+    """Formats a long or complex string like `10L` or `10j`"""
+    number = text[:-1]
+    suffix = text[-1]
+    # Capitalize in "2L" because "l" looks too similar to "1".
+    if suffix == "l":
+        suffix = "L"
+    return f"{format_float_or_int_string(number)}{suffix}"
+
+
 def format_float_or_int_string(text: str) -> str:
     """Formats a float string like "1.0"."""
     if "." not in text:
@@ -6738,13 +6798,33 @@ def is_docstring(leaf: Leaf) -> bool:
     return False
 
 
+def lines_with_leading_tabs_expanded(s: str) -> List[str]:
+    """
+    Splits string into lines and expands only leading tabs (following the normal
+    Python rules)
+    """
+    lines = []
+    for line in s.splitlines():
+        # Find the index of the first non-whitespace character after a string of
+        # whitespace that includes at least one tab
+        match = re.match(r"\s*\t+\s*(\S)", line)
+        if match:
+            first_non_whitespace_idx = match.start(1)
+
+            lines.append(
+                line[:first_non_whitespace_idx].expandtabs()
+                + line[first_non_whitespace_idx:]
+            )
+        else:
+            lines.append(line)
+    return lines
+
+
 def fix_docstring(docstring: str, prefix: str) -> str:
     # https://www.python.org/dev/peps/pep-0257/#handling-docstring-indentation
     if not docstring:
         return ""
-    # Convert tabs to spaces (following the normal Python rules)
-    # and split into a list of lines:
-    lines = docstring.expandtabs().splitlines()
+    lines = lines_with_leading_tabs_expanded(docstring)
     # Determine minimum indentation (first line doesn't count):
     indent = sys.maxsize
     for line in lines[1:]: