libunistring: update to Unicode 9.0.0ueno/unicode-9.0.0

* lib/gen-uni-tables.c (fill_properties): Recognize Sentence_Terminal
and Prepended_Concatenation_Mark.
(is_property_default_ignorable_code_point): Exclude U+08E2.
(fill_arabicshaping): Allow missing whitespace when parsing;
recognize "AFRICAN FEH", "AFRICAN QAF", and "AFRICAN MOON".
(output_blocks): Increase the element size of the level1 table to
accommodate more blocks.
(get_lbp): Recognize ZWJ, E_Base, and E_Modifier characters;
Update each class according to the standard.
(get_wbp): Recognize ZWJ, E_Base, E_Modifier, Glue_After_Zwj, and
E_Base_GAZ characters.
(output_gbp_table): Recognize ZWJ, E_Base, E_Modifier, Glue_After_Zwj,
and E_Base_GAZ characters.
* lib/unictype.in.h (UC_JOINING_GROUP_AFRICAN_FEH,
UC_JOINING_GROUP_AFRICAN_QAF, UC_JOINING_GROUP_AFRICAN_MOON): New enum value.
* lib/unilbrk/lbrktables.h (LBP_ZWJ, LBP_EB, LBP_EM): New enum value.
* lib/unilbrk/lbrktables.c (unilbrk_table): Extend the table with
LBP_ZWJ, LBP_EB, and LBP_EM.
* lib/uniwbrk.in.h (WBP_ZWJ, WBP_EB, WBP_EM, WBP_GAZ, WBP_EBG): New
enum value.
* lib/uniwbrk/u-wordbreaks.h: Implement WB3c, WB15, and WB16.
* lib/uniwbrk/wbrktable.h (uniwbrk_prop_index): New variable declaration.
* lib/uniwbrk/wbrktable.c (uniwbrk_prop_index): New variable.
(uniwbrk_table): Implement WB14.
* tests/uniwbrk/test-uc-wordbreaks.c (wordbreakproperty_to_string):
Check WBP_ZWJ, WBP_EB, WBP_EM, WBP_GAZ, and WBP_EBG.
* modules/unigbrk/u{32,16,8}-grapheme-breaks: No longer depend on
uc-is-grapheme-break.
* modules/unigbrk/uc-grapheme-breaks: New module.
* modules/unigbrk/uc-grapheme-breaks-tests: New module.
* lib/unigbrk.in.h (GBP_ZWJ, GBP_EB, GBP_EM, GBP_GAZ, GBP_EBG): New
enum value.
(uc_grapheme_breaks): New function, replacing uc_is_grapheme_break.
* lib/unigbrk/u-grapheme-breaks.h: New file.
* lib/unigbrk/u{32,16,8}-grapheme-breaks.c: Rewrite using
u-grapheme-breaks.h instead of uc_is_grapheme_break.
* lib/unigbrk/uc-grapheme-breaks.c: New file.
* lib/unigbrk/uc-is-grapheme-break.c: Partially update to TR29 rev 29.
* tests/unigbrk/test-uc-gbrk-prop.c (graphemebreakproperty_to_string):
Check GBP_ZWJ, GBP_EB, GBP_EM, GBP_GAZ, and GBP_EBG.
* tests/unigbrk/test-uc-grapheme-breaks.c: New test.
* tests/unigbrk/test-uc-is-grapheme-break.c (graphemebreakproperty_to_string):
Check GBP_ZWJ, GBP_EB, GBP_EM, GBP_GAZ, and GBP_EBG.
(main): Skip unsupported rules involving 3 or more characters, namely
GB10, GB12, and GB13.
* lib/uniwidth/width.c (nonspacing_table_data): Update.

1 files changed, 191 insertions, 0 deletions

diff --git a/tests/unigbrk/test-uc-grapheme-breaks.c b/tests/unigbrk/test-uc-grapheme-breaks.c
new file mode 100644
index 0000000000..7ee02b9c45
--- /dev/null
+++ b/tests/unigbrk/test-uc-grapheme-breaks.c
@@ -0,0 +1,191 @@
+/* Grapheme cluster break function test.
+   Copyright (C) 2010-2017 Free Software Foundation, Inc.
+
+   This program is free software: you can redistribute it and/or modify it
+   under the terms of the GNU Lesser General Public License as published
+   by the Free Software Foundation; either version 3 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+   Lesser General Public License for more details.
+
+   You should have received a copy of the GNU Lesser General Public License
+   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
+
+/* Written by Ben Pfaff <blp@cs.stanford.edu>, 2010. */
+
+#include <config.h>
+
+/* Specification. */
+#include <unigbrk.h>
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+
+#include "macros.h"
+
+static const char *
+graphemebreakproperty_to_string (int gbp)
+{
+  printf ("%d\n", gbp);
+  switch (gbp)
+    {
+#define CASE(VALUE) case GBP_##VALUE: return #VALUE;
+      CASE(OTHER)
+      CASE(CR)
+      CASE(LF)
+      CASE(CONTROL)
+      CASE(EXTEND)
+      CASE(PREPEND)
+      CASE(SPACINGMARK)
+      CASE(L)
+      CASE(V)
+      CASE(T)
+      CASE(LV)
+      CASE(LVT)
+      CASE(RI)
+      CASE(ZWJ)
+      CASE(EB)
+      CASE(EM)
+      CASE(GAZ)
+      CASE(EBG)
+    }
+  abort ();
+}
+
+static void
+test_uc_grapheme_breaks (const char *expected, ucs4_t *s, size_t n,
+                          const char *filename, int lineno)
+{
+  char breaks[16];
+  size_t i;
+
+  ASSERT (n <= 16);
+
+  uc_grapheme_breaks (s, n, breaks);
+  for (i = 0; i < n; i++)
+    if (breaks[i] != (expected[i] == '#'))
+      {
+        size_t j;
+
+        fprintf (stderr, "wrong grapheme breaks:\n");
+
+        fprintf (stderr, "   input:");
+        for (j = 0; j < n; j++)
+          fprintf (stderr, " %02x", s[j]);
+        putc ('\n', stderr);
+
+        fprintf (stderr, "expected:");
+        for (j = 0; j < n; j++)
+          fprintf (stderr, "  %d", expected[j] == '#');
+        putc ('\n', stderr);
+
+        fprintf (stderr, "  actual:");
+        for (j = 0; j < n; j++)
+          fprintf (stderr, "  %d", breaks[j]);
+        putc ('\n', stderr);
+
+        abort ();
+      }
+}
+
+int
+main (int argc, char *argv[])
+{
+  const char *filename;
+  char line[1024];
+  int exit_code;
+  FILE *stream;
+  int lineno;
+
+  if (argc != 2)
+    {
+      fprintf (stderr, "usage: %s FILENAME\n"
+               "where FILENAME is the location of the GraphemeBreakTest.txt\n"
+               "test file.\n", argv[0]);
+      exit (1);
+    }
+
+  filename = argv[1];
+  stream = fopen (filename, "r");
+  if (stream == NULL)
+    {
+      fprintf (stderr, "error during fopen of '%s'\n", filename);
+      exit (1);
+    }
+
+  exit_code = 0;
+  lineno = 0;
+  while (fgets (line, sizeof line, stream))
+    {
+      char *comment;
+      const char *p;
+      ucs4_t s[16];
+      char breaks[16];
+      size_t i = 0;
+
+      lineno++;
+
+      comment = strchr (line, '#');
+      if (comment != NULL)
+        *comment = '\0';
+      if (line[strspn (line, " \t\r\n")] == '\0')
+        continue;
+
+      s[0] = 0;
+      p = line;
+      do
+        {
+          ucs4_t next;
+
+          p += strspn (p, " \t\r\n");
+          if (!strncmp (p, "\303\267" /* ÷ */, 2))
+            {
+              breaks[i] = '#';
+              p += 2;
+            }
+          else if (!strncmp (p, "\303\227" /* × */, 2))
+            {
+              breaks[i] = '_';
+              p += 2;
+            }
+          else
+            {
+              fprintf (stderr, "%s:%d.%d: syntax error expecting '÷' or '×'\n",
+                       filename, lineno, (int) (p - line + 1));
+              exit (1);
+            }
+
+          p += strspn (p, " \t\r\n");
+          if (*p == '\0')
+            s[i] = 0;
+          else
+            {
+              unsigned int next_int;
+              int n;
+
+              if (sscanf (p, "%x%n", &next_int, &n) != 1)
+                {
+                  fprintf (stderr, "%s:%d.%d: syntax error at '%s' "
+                           "expecting hexadecimal Unicode code point number\n",
+                           filename, lineno, (int) (p - line + 1), p);
+                  exit (1);
+                }
+              p += n;
+
+              s[i] = next_int;
+            }
+          p += strspn (p, " \t\r\n");
+          i++;
+        }
+      while (*p != '\0');
+
+      if (i > 0)
+        test_uc_grapheme_breaks (breaks, s, i, filename, lineno);
+    }
+
+  return exit_code;
+}