feat: add --ignore-sic to skip misspellings marked with [sic]

README.rst

@@ -172,6 +172,23 @@ Words should be separated by a comma.
        def wrod(wrods):
            pass
 
+Ignoring misspellings marked with "[sic]"
+-----------------------------------------
+
+The ``--ignore-sic`` option tells codespell to skip a misspelling that is
+followed by the editorial ``[sic]`` marker (case-insensitive). Only the single
+occurrence preceding the marker is ignored, so other misspellings on the same
+line are still reported. A closing quote may sit between the word and the
+marker, which is the common case when documenting a corrected typo (for example
+in a changelog):
+
+.. code-block:: text
+
+    correct the "wrod" [sic] typo in a changelog entry
+
+Unlike ``codespell:ignore``, the marker is part of the prose itself and does not
+require naming the word in a tooling comment.
+
 Using a config file
 -------------------
 

codespell_lib/_codespell.py

@@ -63,6 +63,9 @@
 ignore_next_line_regex = re.compile(
     rf"[^\w\s]\s*{codespell_ignore_next_line_tag}\b(\s+(?P<words>[\w,]*))?"
 )
+# Editorial "[sic]" marker following a word, allowing an intervening closing
+# quote so a quoted typo like `"wrod" [sic]` is matched.
+sic_regex = re.compile(r"[\"'’”)\s]*\[sic\]", re.IGNORECASE)  # noqa: RUF001
 USAGE = """
 \t%prog [OPTIONS] [file1 file2 ... fileN]
 """
@@ -658,6 +661,13 @@ def convert_arg_line_to_args(self, arg_line: str) -> list[str]:
         metavar="LINES",
         help="print LINES of surrounding context",
     )
+    parser.add_argument(
+        "--ignore-sic",
+        action="store_true",
+        default=False,
+        help='ignore a misspelling immediately followed by the editorial "[sic]" '
+        "marker (optionally preceded by a closing quote).",
+    )
     parser.add_argument(
         "--stdin-single-line",
         action="store_true",
@@ -1049,6 +1059,11 @@ def parse_lines(
                 ):
                     continue
 
+                # An "[sic]" marker right after the word flags it as an
+                # intentional/quoted spelling, so leave it alone.
+                if options.ignore_sic and sic_regex.match(line, match.end()):
+                    continue
+
                 context_shown = False
                 fix = misspellings[lword].fix
                 fixword = fix_case(word, misspellings[lword].data)

codespell_lib/tests/test_basic.py

@@ -493,6 +493,46 @@ def test_inline_ignores(
     assert cs.main(d) == expected_error_count
 
 
+@pytest.mark.parametrize(
+    ("content", "expected_error_count"),
+    [
+        # marker right after the word (optional whitespace) excuses it
+        ("they wrod [sic] it\n", 0),
+        ("they wrod[sic] it\n", 0),
+        ("they wrod  [sic] it\n", 0),
+        # case-insensitive marker
+        ("they wrod [SIC] it\n", 0),
+        # quoted typo followed by the marker (changelog use case)
+        ('correct "wrod" [sic] typo\n', 0),
+        ('correct "wrod"[sic] typo\n', 0),
+        # only the immediately preceding occurrence is excused
+        ("wrod wrod [sic]\n", 1),
+        # a marker elsewhere on the line does not excuse the word
+        ("wrod it [sic] anyway abilty\n", 2),
+        # an intervening word breaks the association
+        ('wrod" abilty [sic]\n', 1),
+        # without a marker the misspelling is still reported
+        ("they wrod it\n", 1),
+        # not a real marker
+        ("they wrod (sic) it\n", 1),
+        ("they wrod [sick] it\n", 1),
+    ],
+)
+def test_ignore_sic(
+    tmpdir: pytest.TempPathFactory,
+    capsys: pytest.CaptureFixture[str],
+    content: str,
+    expected_error_count: int,
+) -> None:
+    d = str(tmpdir)
+    with open(op.join(d, "bad.txt"), "w", encoding="utf-8") as f:
+        f.write(content)
+    # off by default
+    assert cs.main(d) == content.count("wrod") + content.count("abilty")
+    # opt-in
+    assert cs.main("--ignore-sic", d) == expected_error_count
+
+
 @pytest.mark.parametrize(
     ("content", "expected_error_count"),
     [