grep: fix bug with . and some Hangul Syllables

* NEWS: Mention the fix, which comes from the recent Gnulib update.
* tests/hangul-syllable: New file.
* tests/Makefile.am (TESTS): Add it.

3 files changed, 96 insertions, 0 deletions

diff --git a/NEWS b/NEWS
index 86c82ed8..fb0e4cf1 100644
--- a/NEWS
+++ b/NEWS
@@ -13,6 +13,13 @@ GNU grep NEWS                                    -*- outline -*-
 
 ** Bug fixes
 
+  In locales using UTF-8 encoding, the regular expression '.' no
+  longer sometimes fails to match Unicode characters U+D400 through
+  U+D7FF (some Hangul Syllables, and Hangul Jamo Extended-B) and
+  Unicode characters U+108000 through U+10FFFF (half of Supplemental
+  Private Use Area plane B).
+  [bug introduced in grep 3.4]
+
   The -s option no longer suppresses "binary file matches" messages.
   [Bug#51860 introduced in grep 3.5]
 
diff --git a/tests/Makefile.am b/tests/Makefile.am
index 708980df..d72637f7 100644
--- a/tests/Makefile.am
+++ b/tests/Makefile.am
@@ -110,6 +110,7 @@ TESTS =						\
   grep-dev-null					\
   grep-dev-null-out				\
   grep-dir					\
+  hangul-syllable				\
   hash-collision-perf				\
   help-version					\
   high-bit-range				\
diff --git a/tests/hangul-syllable b/tests/hangul-syllable
new file mode 100755
index 00000000..9f94d2eb
--- /dev/null
+++ b/tests/hangul-syllable
@@ -0,0 +1,88 @@
+#!/bin/sh
+# grep 3.4 through 3.7 mishandled matching '.' against the valid UTF-8
+# sequences (ED)(90-9F)(80-BF) corresponding to U+D400 through U+D7FF,
+# which are some Hangul Syllables and Hangul Jamo Extended-B.  They
+# also mishandled (F4)(88-8F)(80-BF)(80-BF) which correspond to
+# U+108000 through U+10FFFF (Supplemental Private Use Area plane B).
+
+. "${srcdir=.}/init.sh"; path_prepend_ ../src
+
+require_en_utf8_locale_
+
+LC_ALL=en_US.UTF-8
+export LC_ALL
+
+check_char ()
+{
+  printf "$1\\n" >in || framewmork_failure_
+
+  grep $2 '^.$' in >out || fail=1
+  cmp in out || fail=1
+}
+
+fail=0
+
+# "." should match U+D45C HANGUL SYLLABLE PYO.
+check_char '\355\221\234'
+
+# Check boundary-condition characters
+# while we are at it.
+
+check_char '\0' -a
+check_char '\177'
+
+for i in 302 337; do
+  for j in 200 277; do
+    check_char "\\$i\\$j"
+  done
+done
+for i in 340; do
+  for j in 240 277; do
+    for k in 200 277; do
+      check_char "\\$i\\$j\\$k"
+    done
+  done
+done
+for i in 341 354 356 357; do
+  for j in 200 277; do
+    for k in 200 277; do
+      check_char "\\$i\\$j\\$k"
+    done
+  done
+done
+for i in 355; do
+  for j in 200 237; do
+    for k in 200 277; do
+      check_char "\\$i\\$j\\$k"
+    done
+  done
+done
+for i in 360; do
+  for j in 220 277; do
+    for k in 200 277; do
+      for l in 200 277; do
+        check_char "\\$i\\$j\\$k\\$l"
+      done
+    done
+  done
+done
+for i in 361 363; do
+  for j in 200 277; do
+    for k in 200 277; do
+      for l in 200 277; do
+        check_char "\\$i\\$j\\$k\\$l"
+      done
+    done
+  done
+done
+for i in 364; do
+  for j in 200 217; do
+    for k in 200 277; do
+      for l in 200 277; do
+        check_char "\\$i\\$j\\$k\\$l"
+      done
+    done
+  done
+done
+
+Exit $fail