Define is_name_token (and friends) to resolve some `type: ignore`s (GH-2714)

[etc/vim.git] / src / black / strings.py

diff --git a/src/black/strings.py b/src/black/strings.py
index 5b443ddebc9e7e662b469b2c2717052ff2dcbf99..06a5da01f0cf71f786f3878480e71282731f01c3 100644 (file)
--- a/src/black/strings.py
+++ b/src/black/strings.py
@@ -2,12 +2,22 @@
 Simple formatting on strings. Further string formatting code is in trans.py.
 """
 
 Simple formatting on strings. Further string formatting code is in trans.py.
 """
 

-import regex as re
+import re

 import sys
 import sys

+from functools import lru_cache

 from typing import List, Pattern
 
 from typing import List, Pattern
 

+if sys.version_info < (3, 8):
+    from typing_extensions import Final
+else:
+    from typing import Final

 
 

-STRING_PREFIX_CHARS = "furbFURB"  # All possible string prefix characters.
+
+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)")

 
 
 def sub_twice(regex: Pattern[str], replacement: str, original: str) -> str:
 
 
 def sub_twice(regex: Pattern[str], replacement: str, original: str) -> str:


@@ -37,7 +47,7 @@ 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
     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)
 
         if match:
             first_non_whitespace_idx = match.start(1)
 


@@ -87,7 +97,7 @@ def get_string_prefix(string: str) -> str:
     prefix = ""
     prefix_idx = 0
     while string[prefix_idx] in STRING_PREFIX_CHARS:
     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
         prefix_idx += 1
 
     return prefix


@@ -133,7 +143,7 @@ def normalize_string_prefix(s: str, remove_u_prefix: bool = False) -> str:
 
     If remove_u_prefix is given, also removes any u prefix from the string.
     """
 
     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)
+    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")
     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")


@@ -142,6 +152,14 @@ def normalize_string_prefix(s: str, remove_u_prefix: bool = False) -> str:
     return f"{new_prefix}{match.group(2)}"
 
 
     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.
 
 def normalize_string_quotes(s: str) -> str:
     """Prefer double quotes but only if it doesn't cause more escaping.
 


@@ -166,9 +184,9 @@ def normalize_string_quotes(s: str) -> str:
         return s  # There's an internal error
 
     prefix = s[:first_quote_pos]
         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):
     body = s[first_quote_pos + len(orig_quote) : -len(orig_quote)]
     if "r" in prefix.casefold():
         if unescaped_new_quote.search(body):


@@ -190,9 +208,9 @@ def normalize_string_quotes(s: str) -> str:
     if "f" in prefix.casefold():
         matches = re.findall(
             r"""
     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 {{
                 ([^{].*?)  # 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,
             """,
             new_body,
             re.VERBOSE,