Do not allow for a word to start or end with punctuation symbols

codespell_lib/_codespell.py

@@ -53,7 +53,9 @@
     __version__ as VERSION,  # noqa: N812
 )
 
-word_regex_def = r"[\w\-'’]+"  # noqa: RUF001
+# Words could be surrounded in quotes, so we allow for that, but no nested quotes.
+# Cannot have leading but can have trailing hyphens or apostrophes.
+word_regex_def = r"(?<=')[\w\-’]+(?=')|(?<=’)[\w\-']+(?=’)|\w[\w\-'’]*"  # noqa: RUF001
 # While we want to treat characters like ( or " as okay for a starting break,
 # these may occur unescaped in URIs, and so we are more restrictive on the
 # endpoint.  Emails are more restrictive, so the endpoint remains flexible.

codespell_lib/tests/test_basic.py

@@ -51,6 +51,7 @@ def main(
         capsys = frame.f_locals["capsys"]
         stdout, stderr = capsys.readouterr()
         assert code in (EX_OK, EX_USAGE, EX_DATAERR, EX_CONFIG)
+        # print(f"stderr: {stderr}")
         if code == EX_DATAERR:  # have some misspellings
             code = int(stderr.split("\n")[-2])
         elif code == EX_OK and count:
@@ -117,7 +118,13 @@ def test_basic(
     assert cs.main("--builtin", "clear,rare,names,informal", fname) == 4
     with fname.open("w") as f:  # overwrite the file
         f.write("var = 'nwe must check codespell likes escapes nin strings'\n")
-    assert cs.main(fname) == 1, "checking our string escape test word is bad"
+    assert cs.main(fname) == 2, "checking our string escape test word is bad"
+    with fname.open("w") as f:  # overwrite the file
+        f.write("fully 'nwe' quoted\n")
+    assert cs.main(fname) == 1, "fully quoted"
+    with fname.open("w") as f:  # overwrite the file
+        f.write("only end nwe' quoted\n")
+    assert cs.main(fname) == 0, "only end quoted should be ok since we have werent'"
     # the first one is missed because the apostrophe means its not currently
     # treated as a word on its own
     with fname.open("w") as f:  # overwrite the file