| 1 | #! /bin/sh
|
|---|
| 2 | # Ensure that, with -P, Unicode \p{} symbols are correctly matched.
|
|---|
| 3 | #
|
|---|
| 4 | # Copyright (C) 2012-2021 Free Software Foundation, Inc.
|
|---|
| 5 | #
|
|---|
| 6 | # Copying and distribution of this file, with or without modification,
|
|---|
| 7 | # are permitted in any medium without royalty provided the copyright
|
|---|
| 8 | # notice and this notice are preserved.
|
|---|
| 9 |
|
|---|
| 10 | . "${srcdir=.}/init.sh"; path_prepend_ ../src
|
|---|
| 11 | require_en_utf8_locale_
|
|---|
| 12 | LC_ALL=en_US.UTF-8 require_pcre_
|
|---|
| 13 |
|
|---|
| 14 | fail=0
|
|---|
| 15 |
|
|---|
| 16 | echo '$' | LC_ALL=en_US.UTF-8 grep -qP '\p{S}' \
|
|---|
| 17 | || skip_ 'PCRE support is compiled out, or it does not support properties'
|
|---|
| 18 |
|
|---|
| 19 | euro='\342\202\254 euro'
|
|---|
| 20 | printf "$euro\\n" > in || framework_failure_
|
|---|
| 21 |
|
|---|
| 22 | # The euro sign has the unicode "Symbol" property, so this must match:
|
|---|
| 23 | LC_ALL=en_US.UTF-8 grep -P '^\p{S}' in > out || fail=1
|
|---|
| 24 | compare in out || fail=1
|
|---|
| 25 |
|
|---|
| 26 | # This RE must *not* match in the C locale, because the first
|
|---|
| 27 | # byte is not a "Symbol".
|
|---|
| 28 | LC_ALL=C grep -P '^\p{S}' in > out && fail=1
|
|---|
| 29 | compare /dev/null out || fail=1
|
|---|
| 30 |
|
|---|
| 31 | LC_ALL=en_US.UTF-8 grep -P '^. euro$' in > out2 || fail=1
|
|---|
| 32 | compare in out2 || fail=1
|
|---|
| 33 |
|
|---|
| 34 | LC_ALL=en_US.UTF-8 grep -oP '. euro' in > out3 || fail=1
|
|---|
| 35 | compare in out3 || fail=1
|
|---|
| 36 |
|
|---|
| 37 | LC_ALL=en_US.UTF-8 grep -P '^\P{S}' in > out4
|
|---|
| 38 | compare /dev/null out4 || fail=1
|
|---|
| 39 |
|
|---|
| 40 | Exit $fail
|
|---|
