maint: ensure that new "make syntax-check"-run sc_codespell passes

* cfg.mk (codespell_ignore_words_list): Ignore false-positives.
(exclude_file_name_regexp--sc_codespell): Skip some file names.
* sed/debug.c: Fix typo(s).
* sed/mbcs.c: Likewise.
* sed/sed.c: Likewise.
* testsuite/badenc.sh: Likewise.
* testsuite/compile-errors.sh: Likewise.
* testsuite/distrib.inp: Likewise.
* testsuite/distrib.sh: Likewise.
* testsuite/execute-tests.sh: Likewise.
* testsuite/mb-bad-delim.sh: Likewise.
* testsuite/mb-charclass-non-utf8.sh: Likewise.
* testsuite/regex-max-int.sh: Likewise.
* testsuite/subst-options.sh: Likewise.
* testsuite/test-mbrtowc.c: Likewise.
* testsuite/utf8-ru.sh: Likewise.

cfg.mk

@@ -408,3 +408,7 @@ build-ubsan:
 	    { echo "./configure script not found" >&2; exit 1; }
 	./configure CFLAGS="$(UBSAN_CFLAGS)" LDFLAGS="$(UBSAN_LDFLAGS)"
 	make
+
+exclude_file_name_regexp--sc_codespell = \
+  ^(THANKS\.in|testsuite/(8bit|mac-mf)\..+)$$
+codespell_ignore_words_list = ket,tre,debbugs,fo,2Rd,jaques,hel,te,bu,readin

sed/debug.c

@@ -424,7 +424,7 @@ debug_print_command (const struct vector *program, const struct sed_cmd *sc)
   debug_print_addr (sc->a2);
 
   addr_bang = sc->addr_bang;
-  /* Implmentation detail: GNU Sed implements beginning of block
+  /* Implementation detail: GNU Sed implements beginning of block
      by negating the matched address and jumping if there's no match.  */
   if (sc->cmd == '{')
     addr_bang = !addr_bang;

sed/mbcs.c

@@ -29,7 +29,7 @@ bool is_utf8;
 
    Return zero in all other cases:
     CH is a valid single-byte character (e.g. 0x01-0x7F in UTF-8 locales);
-    CH is an invalid byte in a multibyte sequence for the currentl locale,
+    CH is an invalid byte in a multibyte sequence for the current locale,
     CH is the NUL byte.
 
    Reset CUR_STAT in the case of an invalid byte.

sed/sed.c

@@ -59,7 +59,7 @@ bool separate_files = false;
 /* If set, follow symlinks when processing in place */
 bool follow_symlinks = false;
 
-/* If set, opearate in 'sandbox' mode */
+/* If set, operate in 'sandbox' mode */
 bool sandbox = false;
 
 /* if set, print debugging information */

testsuite/badenc.sh

@@ -27,7 +27,7 @@ printf "abc\nde\320f\nghi\njkl\nmno\npqr\nstu\nvwx\nyz\n" > badenc-inp \
     || framework_failure_
 
 
-# The progarm: using 'z' to clear the pattern-space even
+# The program: using 'z' to clear the pattern-space even
 # if it contains invalid multibyte sequences.
 # Using 's/.*//' would not be able to clear the pattern-space.
 cat << \EOF > badenc.sed || framework_failure_

testsuite/compile-errors.sh

@@ -259,7 +259,7 @@ returns_ 1 sed 'y/a/a' </dev/null 2>err-unterm-y5 || fail=1
 compare_ exp-unterm-y err-unterm-y5 || fail=1
 
 #
-# Y command with bad legth (Y_CMD_LEN)
+# Y command with bad length (Y_CMD_LEN)
 # TODO: check with multibyte strings.
 cat <<\EOF >exp-bad-y-len || framework_failure_
 sed: -e expression #1, char 7: 'y' command strings have different lengths

testsuite/distrib.inp

@@ -20,7 +20,7 @@ X-Mailer: ELM [version 2.4 PL23]
            shine of the sea:
     Breath and bloom, shade and shine, -- wonder,
         wealth, and -- how far above them --
-          Truth, thats brighter than gem,
+          Truth, that's brighter than gem,
           Trust, that's purer than pearl, --
     Brightest truth, purest trust in the universe --
               all were for me

testsuite/distrib.sh

@@ -103,7 +103,7 @@ Approved: alt.buddha.short.fat.guy@cygnus.com
            shine of the sea:
     Breath and bloom, shade and shine, -- wonder,
         wealth, and -- how far above them --
-          Truth, thats brighter than gem,
+          Truth, that's brighter than gem,
           Trust, that's purer than pearl, --
     Brightest truth, purest trust in the universe --
               all were for me

testsuite/execute-tests.sh

@@ -40,17 +40,17 @@ echo "printf a" | sed '1e' > out3 || fail=1
 compare_ exp3 out3 || fail=1
 
 #
-# plain 'e' with a command that returns delimted '\n' output
+# plain 'e' with a command that returns delimited '\n' output
 # (implementation note: the delimiter is first chomp'd)
 printf "a\n" > exp4 || framework_failure_
 echo "echo a" | sed '1e' > out4 || fail=1
 compare_ exp4 out4 || fail=1
 
 #
-# e with a command that returns delimted '\0' output
+# e with a command that returns delimited '\0' output
 #
 printf "b\0" > exp5 || framework_failure_
-# This input file contains the shell command to be excuted:
+# This input file contains the shell command to be executed:
 printf 'cat exp5' > in5 || framework_failure_
 sed -z '1e' <in5 > out5 || fail=1
 compare_ exp5 out5 || fail=1

testsuite/mb-bad-delim.sh

@@ -30,7 +30,7 @@ require_en_utf8_locale_
 # Using the first octet alone (\316) causes various multibyte related functions
 # to return '-2' (incomplete multibyte sequence).
 # using the second octet alone (\246) causess same functions to return '-1'
-# (ivalid multibyte sequence).
+# (invalid multibyte sequence).
 
 
 # Reject a valid multibyte delimiter (instead of slash).
@@ -66,7 +66,7 @@ compare_ exp2 out2 || fail=1
 
 
 # An invalid multibyte sequence is treated as a valid single byte,
-# thus accepted as a delimter (instead of slash).
+# thus accepted as a delimiter (instead of slash).
 # This is an implmentation-specific behavior.
 printf 's\246a\246b\246' > prog3 || framework_failure_
 echo a > in3 || framework_failure_

testsuite/mb-charclass-non-utf8.sh

@@ -71,11 +71,11 @@ require_valid_ja_shiftjis_locale_ "$LOCALE_JA_SJIS"
 
 
 #
-# Tests 1,2: Test y/// command with multibyte, non-utf8 seqeunce.
-# Implmenetation notes: str_append() has special code path for non-utf8 cases.
+# Tests 1,2: Test y/// command with multibyte, non-utf8 sequence.
+# Implementation notes: str_append() has special code path for non-utf8 cases.
 #
 
-# Test 1: valid multibyte seqeunce
+# Test 1: valid multibyte sequence
 printf 'y/a/\203\133/' > p1 || framework_failure_
 echo Xa > in1 || framework_failure_
 printf 'X\203\133\n' > exp1 || framework_failure_
@@ -83,7 +83,7 @@ printf 'X\203\133\n' > exp1 || framework_failure_
 LC_ALL="$LOCALE_JA_SJIS" sed -f p1 <in1 >out1 || fail=1
 compare_ exp1 out1 || fail=1
 
-# Test 2: invalid multibyte seqeunce, treated as two single-byte characters.
+# Test 2: invalid multibyte sequence, treated as two single-byte characters.
 printf 'y/aa/\203\060/' > p2 || framework_failure_
 LC_ALL="$LOCALE_JA_SJIS" sed -f p2 </dev/null 2>out2 || fail=1
 compare_ /dev/null out2 || fail=1

testsuite/regex-max-int.sh

@@ -39,7 +39,7 @@ sed: regex input buffer length larger than INT_MAX
 EOF
 
 
-# Before sed-4.5, this was silently a no-op: would not perform the subsitution
+# Before sed-4.5, this was silently a no-op: would not perform the substitution
 # but would not indicate any error either (https://bugs.gnu.org/30520).
 # Exit code 4 is "panic".
 returns_ 4 sed 's/a/b/g' input >/dev/null 2>err1 || fail=1

testsuite/subst-options.sh

@@ -91,7 +91,7 @@ compare_ subst-exp2 subst-out2 || fail=1
 #
 # Multiline modifier: s///m
 # ('N' will read and concatenate the second line
-#  into the patten space, making it "foo\nbar".
+#  into the pattern space, making it "foo\nbar".
 #  s// will then operate on it as one string).
 printf "foo\nbar\n" > subst-in3 || fail=1
 printf "Xoo\nXar\n" > subst-exp3 || fail=1

testsuite/test-mbrtowc.c

@@ -15,7 +15,7 @@
    along with this program; If not, see <https://www.gnu.org/licenses/>. */
 
 /* Test the operating-system's native mbrtowc(3) function,
-   by feeding it multibyte seqeunces one byte at a time,
+   by feeding it multibyte sequences one byte at a time,
    and reporting the result.
 
    The program prints the following values after each mbrtowc invocation,
@@ -83,7 +83,7 @@
     $ printf '\203\133' | LC_ALL=ja_JP.SJIS test-mbrtowc
     -2,1
 
-  The follwing is an INVALID multibyte sequence in SHIFT-JIS
+  The following is an INVALID multibyte sequence in SHIFT-JIS
   (The byte ':' is not valid as a second octet).
   Buggy implementations will accept this as a valid multibyte sequence:
 
@@ -102,7 +102,7 @@
   ---------------------------------------------
 
   In GNU Sed some tests purposely use invalid multibyte sequences
-  to test sed's behaviour. A buggy implemetation of mbrtowc
+  to test sed's behaviour. A buggy implementation of mbrtowc
   would result in false-alarm failures.
 
   The following are expected results in correct implementations:

testsuite/utf8-ru.sh

@@ -58,11 +58,11 @@ cp utf8-1-exp utf8-2-exp || framework_failure_
 
 
 # Test 3: Capitalize only the next character (\u)
-# Only the first "DE" should be capitilized.
+# Only the first "DE" should be capitalized.
 #       s/д.*/\u&/g
 printf 's/\320\264.*/\\u&/g' > utf8-3.sed || framework_failure_
 
-# Test 3: Expected output - First DE capitilized, second DE not.
+# Test 3: Expected output - First DE capitalized, second DE not.
 #       Д       а        д
 printf '\320\224\320\260 \320\264\n' > utf8-3-exp || framework_failure_