]> git.madduck.net Git - etc/vim.git/blobdiff - src/black/strings.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 peter-evans/create-or-update-comment from 3.0.1 to 3.0.2 (#3730)
[etc/vim.git] / src / black / strings.py
index 5b443ddebc9e7e662b469b2c2717052ff2dcbf99..ac18aef51ed7d7349f0f0bbb4e2dc3ce338c763b 100644 (file)
@@ -2,12 +2,34 @@
 Simple formatting on strings. Further string formatting code is in trans.py.
 """
 
-import regex as re
+import re
 import sys
-from typing import List, Pattern
-
-
-STRING_PREFIX_CHARS = "furbFURB"  # All possible string prefix characters.
+from functools import lru_cache
+from typing import List, Match, Pattern
+
+from blib2to3.pytree import Leaf
+
+if sys.version_info < (3, 8):
+    from typing_extensions import Final
+else:
+    from typing import Final
+
+from black._width_table import WIDTH_TABLE
+
+STRING_PREFIX_CHARS: Final = "furbFURB"  # All possible string prefix characters.
+STRING_PREFIX_RE: Final = re.compile(
+    r"^([" + STRING_PREFIX_CHARS + r"]*)(.*)$", re.DOTALL
+)
+FIRST_NON_WHITESPACE_RE: Final = re.compile(r"\s*\t+\s*(\S)")
+UNICODE_ESCAPE_RE: Final = re.compile(
+    r"(?P<backslashes>\\+)(?P<body>"
+    r"(u(?P<u>[a-fA-F0-9]{4}))"  # Character with 16-bit hex value xxxx
+    r"|(U(?P<U>[a-fA-F0-9]{8}))"  # Character with 32-bit hex value xxxxxxxx
+    r"|(x(?P<x>[a-fA-F0-9]{2}))"  # Character with hex value hh
+    r"|(N\{(?P<N>[a-zA-Z0-9 \-]{2,})\})"  # Character named name in the Unicode database
+    r")",
+    re.VERBOSE,
+)
 
 
 def sub_twice(regex: Pattern[str], replacement: str, original: str) -> str:
@@ -37,7 +59,7 @@ def lines_with_leading_tabs_expanded(s: str) -> List[str]:
     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)
+        match = FIRST_NON_WHITESPACE_RE.match(line)
         if match:
             first_non_whitespace_idx = match.start(1)
 
@@ -87,7 +109,7 @@ def get_string_prefix(string: str) -> str:
     prefix = ""
     prefix_idx = 0
     while string[prefix_idx] in STRING_PREFIX_CHARS:
-        prefix += string[prefix_idx].lower()
+        prefix += string[prefix_idx]
         prefix_idx += 1
 
     return prefix
@@ -128,20 +150,32 @@ def assert_is_leaf_string(string: str) -> None:
     ), f"{set(string[:quote_idx])} is NOT a subset of {set(STRING_PREFIX_CHARS)}."
 
 
-def normalize_string_prefix(s: str, remove_u_prefix: bool = False) -> str:
-    """Make all string prefixes lowercase.
-
-    If remove_u_prefix is given, also removes any u prefix from the string.
-    """
-    match = re.match(r"^([" + STRING_PREFIX_CHARS + r"]*)(.*)$", s, re.DOTALL)
+def normalize_string_prefix(s: str) -> str:
+    """Make all string prefixes lowercase."""
+    match = STRING_PREFIX_RE.match(s)
     assert match is not None, f"failed to match string {s!r}"
     orig_prefix = match.group(1)
-    new_prefix = orig_prefix.replace("F", "f").replace("B", "b").replace("U", "u")
-    if remove_u_prefix:
-        new_prefix = new_prefix.replace("u", "")
+    new_prefix = (
+        orig_prefix.replace("F", "f")
+        .replace("B", "b")
+        .replace("U", "")
+        .replace("u", "")
+    )
+
+    # Python syntax guarantees max 2 prefixes and that one of them is "r"
+    if len(new_prefix) == 2 and "r" != new_prefix[0].lower():
+        new_prefix = new_prefix[::-1]
     return f"{new_prefix}{match.group(2)}"
 
 
+# Re(gex) does actually cache patterns internally but this still improves
+# performance on a long list literal of strings by 5-9% since lru_cache's
+# caching overhead is much lower.
+@lru_cache(maxsize=64)
+def _cached_compile(pattern: str) -> Pattern[str]:
+    return re.compile(pattern)
+
+
 def normalize_string_quotes(s: str) -> str:
     """Prefer double quotes but only if it doesn't cause more escaping.
 
@@ -166,9 +200,9 @@ def normalize_string_quotes(s: str) -> str:
         return s  # There's an internal error
 
     prefix = s[:first_quote_pos]
-    unescaped_new_quote = re.compile(rf"(([^\\]|^)(\\\\)*){new_quote}")
-    escaped_new_quote = re.compile(rf"([^\\]|^)\\((?:\\\\)*){new_quote}")
-    escaped_orig_quote = re.compile(rf"([^\\]|^)\\((?:\\\\)*){orig_quote}")
+    unescaped_new_quote = _cached_compile(rf"(([^\\]|^)(\\\\)*){new_quote}")
+    escaped_new_quote = _cached_compile(rf"([^\\]|^)\\((?:\\\\)*){new_quote}")
+    escaped_orig_quote = _cached_compile(rf"([^\\]|^)\\((?:\\\\)*){orig_quote}")
     body = s[first_quote_pos + len(orig_quote) : -len(orig_quote)]
     if "r" in prefix.casefold():
         if unescaped_new_quote.search(body):
@@ -190,9 +224,9 @@ def normalize_string_quotes(s: str) -> str:
     if "f" in prefix.casefold():
         matches = re.findall(
             r"""
-            (?:[^{]|^)\{  # start of the string or a non-{ followed by a single {
+            (?:(?<!\{)|^)\{  # start of the string or a non-{ followed by a single {
                 ([^{].*?)  # contents of the brackets except if begins with {{
-            \}(?:[^}]|$)  # A } followed by end of the string or a non-}
+            \}(?:(?!\})|$)  # A } followed by end of the string or a non-}
             """,
             new_body,
             re.VERBOSE,
@@ -214,3 +248,88 @@ def normalize_string_quotes(s: str) -> str:
         return s  # Prefer double quotes
 
     return f"{prefix}{new_quote}{new_body}{new_quote}"
+
+
+def normalize_unicode_escape_sequences(leaf: Leaf) -> None:
+    """Replace hex codes in Unicode escape sequences with lowercase representation."""
+    text = leaf.value
+    prefix = get_string_prefix(text)
+    if "r" in prefix.lower():
+        return
+
+    def replace(m: Match[str]) -> str:
+        groups = m.groupdict()
+        back_slashes = groups["backslashes"]
+
+        if len(back_slashes) % 2 == 0:
+            return back_slashes + groups["body"]
+
+        if groups["u"]:
+            # \u
+            return back_slashes + "u" + groups["u"].lower()
+        elif groups["U"]:
+            # \U
+            return back_slashes + "U" + groups["U"].lower()
+        elif groups["x"]:
+            # \x
+            return back_slashes + "x" + groups["x"].lower()
+        else:
+            assert groups["N"], f"Unexpected match: {m}"
+            # \N{}
+            return back_slashes + "N{" + groups["N"].upper() + "}"
+
+    leaf.value = re.sub(UNICODE_ESCAPE_RE, replace, text)
+
+
+@lru_cache(maxsize=4096)
+def char_width(char: str) -> int:
+    """Return the width of a single character as it would be displayed in a
+    terminal or editor (which respects Unicode East Asian Width).
+
+    Full width characters are counted as 2, while half width characters are
+    counted as 1.  Also control characters are counted as 0.
+    """
+    table = WIDTH_TABLE
+    codepoint = ord(char)
+    highest = len(table) - 1
+    lowest = 0
+    idx = highest // 2
+    while True:
+        start_codepoint, end_codepoint, width = table[idx]
+        if codepoint < start_codepoint:
+            highest = idx - 1
+        elif codepoint > end_codepoint:
+            lowest = idx + 1
+        else:
+            return 0 if width < 0 else width
+        if highest < lowest:
+            break
+        idx = (highest + lowest) // 2
+    return 1
+
+
+def str_width(line_str: str) -> int:
+    """Return the width of `line_str` as it would be displayed in a terminal
+    or editor (which respects Unicode East Asian Width).
+
+    You could utilize this function to determine, for example, if a string
+    is too wide to display in a terminal or editor.
+    """
+    if line_str.isascii():
+        # Fast path for a line consisting of only ASCII characters
+        return len(line_str)
+    return sum(map(char_width, line_str))
+
+
+def count_chars_in_width(line_str: str, max_width: int) -> int:
+    """Count the number of characters in `line_str` that would fit in a
+    terminal or editor of `max_width` (which respects Unicode East Asian
+    Width).
+    """
+    total_width = 0
+    for i, char in enumerate(line_str):
+        width = char_width(char)
+        if width + total_width > max_width:
+            return i
+        total_width += width
+    return len(line_str)