diff options
Diffstat (limited to 'src/grep/tests/false-match-mb-non-utf8')
| -rwxr-xr-x | src/grep/tests/false-match-mb-non-utf8 | 38 |
1 files changed, 38 insertions, 0 deletions
diff --git a/src/grep/tests/false-match-mb-non-utf8 b/src/grep/tests/false-match-mb-non-utf8 new file mode 100755 index 0000000..e618996 --- /dev/null +++ b/src/grep/tests/false-match-mb-non-utf8 @@ -0,0 +1,38 @@ +#! /bin/sh +# Test for false matches in grep 2.19..2.26 in multibyte, non-UTF8 locales +# +# Copyright (C) 2016-2021 Free Software Foundation, Inc. +# +# Copying and distribution of this file, with or without modification, +# are permitted in any medium without royalty provided the copyright +# notice and this notice are preserved. + +. "${srcdir=.}/init.sh"; path_prepend_ ../src + +# Add "." to PATH for the use of get-mb-cur-max. +path_prepend_ . + +fail=0 + +loc=zh_CN.gb18030 +test "$(get-mb-cur-max $loc)" = 4 || skip_ "no support for the $loc locale" + +# This must not match: the input is a single character, \uC9 followed +# by a newline. But it just so happens that that character is made up +# of four bytes, the last of which is the digit, 7, and grep's DFA +# matcher would mistakenly report that ".*7" matches that input line. +printf '\2010\2077\n' > in || framework_failure_ +returns_ 1 env LC_ALL=$loc grep -E '.*7' in || fail=1 + +returns_ 1 env LC_ALL=$loc grep -E '.{0,1}7' in || fail=1 + +returns_ 1 env LC_ALL=$loc grep -E '.?7' in || fail=1 + +# Similar for the \ue9 code point, which ends in an "m" byte. +loc=zh_HK.big5hkscs +test "$(get-mb-cur-max $loc)" = 2 || skip_ "no support for the $loc locale" + +printf '\210m\n' > in || framework_failure_ +returns_ 1 env LC_ALL=$loc grep '.*m' in || fail=1 + +Exit $fail |