X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Flibpspp%2Fi18n.c;h=dca85db4f0ee48670a2519bd37bc68f27f58f536;hb=refs%2Fbuilds%2F20130102032118%2Fpspp;hp=754c9321f3c25b0638c62eb7f229fb2d1f217fe3;hpb=d8124a99ae0971ff04d47d31426192ed4bb753b7;p=pspp

diff --git a/src/libpspp/i18n.c b/src/libpspp/i18n.c
index 754c9321f3..dca85db4f0 100644
--- a/src/libpspp/i18n.c
+++ b/src/libpspp/i18n.c
@@ -27,6 +27,7 @@
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
+#include <unicase.h>
 #include <unigbrk.h>
 
 #include "libpspp/assertion.h"
@@ -39,6 +40,7 @@
 
 #include "gl/c-strcase.h"
 #include "gl/localcharset.h"
+#include "gl/minmax.h"
 #include "gl/xalloc.h"
 #include "gl/relocatable.h"
 #include "gl/xstrndup.h"
@@ -677,6 +679,109 @@ uc_name (ucs4_t uc, char buffer[16])
   return buffer;
 }
 
+/* UTF-8 functions that deal with uppercase/lowercase distinctions. */
+
+/* Returns a hash value for the N bytes of UTF-8 encoded data starting at S,
+   with lowercase and uppercase letters treated as equal, starting from
+   BASIS. */
+unsigned int
+utf8_hash_case_bytes (const char *s, size_t n, unsigned int basis)
+{
+  uint8_t folded_buf[2048];
+  size_t folded_len = sizeof folded_buf;
+  uint8_t *folded_s;
+  unsigned int hash;
+
+  folded_s = u8_casefold (CHAR_CAST (const uint8_t *, s), n,
+                          NULL, UNINORM_NFKD, folded_buf, &folded_len);
+  if (folded_s != NULL)
+    {
+      hash = hash_bytes (folded_s, folded_len, basis);
+      if (folded_s != folded_buf)
+        free (folded_s);
+    }
+  else
+    {
+      if (errno == ENOMEM)
+        xalloc_die ();
+      hash = hash_bytes (s, n, basis);
+    }
+
+  return hash;
+}
+
+/* Returns a hash value for null-terminated UTF-8 string S, with lowercase and
+   uppercase letters treated as equal, starting from BASIS. */
+unsigned int
+utf8_hash_case_string (const char *s, unsigned int basis)
+{
+  return utf8_hash_case_bytes (s, strlen (s), basis);
+}
+
+/* Compares UTF-8 strings A and B case-insensitively.
+   Returns a negative value if A < B, zero if A == B, positive if A > B. */
+int
+utf8_strcasecmp (const char *a, const char *b)
+{
+  return utf8_strncasecmp (a, strlen (a), b, strlen (b));
+}
+
+/* Compares UTF-8 strings A (with length AN) and B (with length BN)
+   case-insensitively.
+   Returns a negative value if A < B, zero if A == B, positive if A > B. */
+int
+utf8_strncasecmp (const char *a, size_t an, const char *b, size_t bn)
+{
+  int result;
+
+  if (u8_casecmp (CHAR_CAST (const uint8_t *, a), an,
+                  CHAR_CAST (const uint8_t *, b), bn,
+                  NULL, UNINORM_NFKD, &result))
+    {
+      if (errno == ENOMEM)
+        xalloc_die ();
+
+      result = memcmp (a, b, MIN (an, bn));
+      if (result == 0)
+        result = an < bn ? -1 : an > bn;
+    }
+
+  return result;
+}
+
+static char *
+utf8_casemap (const char *s,
+              uint8_t *(*f) (const uint8_t *, size_t, const char *, uninorm_t,
+                             uint8_t *, size_t *))
+{
+  char *result;
+  size_t size;
+
+  result = CHAR_CAST (char *,
+                      f (CHAR_CAST (const uint8_t *, s), strlen (s) + 1,
+                         NULL, NULL, NULL, &size));
+  if (result == NULL)
+    {
+      if (errno == ENOMEM)
+        xalloc_die ();
+
+      result = xstrdup (s);
+    }
+  return result;
+}
+
+char *
+utf8_to_upper (const char *s)
+{
+  return utf8_casemap (s, u8_toupper);
+}
+
+char *
+utf8_to_lower (const char *s)
+{
+  return utf8_casemap (s, u8_tolower);
+}
+
 bool
 get_encoding_info (struct encoding_info *e, const char *name)
 {