summary refs log tree commit diff
path: root/libidn/toutf8.c
diff options
context:
space:
mode:
authorUlrich Drepper <drepper@redhat.com>2004-12-22 20:10:10 +0000
committerUlrich Drepper <drepper@redhat.com>2004-12-22 20:10:10 +0000
commita334319f6530564d22e775935d9c91663623a1b4 (patch)
treeb5877475619e4c938e98757d518bb1e9cbead751 /libidn/toutf8.c
parent0ecb606cb6cf65de1d9fc8a919bceb4be476c602 (diff)
downloadglibc-a334319f6530564d22e775935d9c91663623a1b4.tar.gz
glibc-a334319f6530564d22e775935d9c91663623a1b4.tar.xz
glibc-a334319f6530564d22e775935d9c91663623a1b4.zip
(CFLAGS-tst-align.c): Add -mpreferred-stack-boundary=4.
Diffstat (limited to 'libidn/toutf8.c')
-rw-r--r--libidn/toutf8.c168
1 files changed, 138 insertions, 30 deletions
diff --git a/libidn/toutf8.c b/libidn/toutf8.c
index dad47c9af0..262f252d5f 100644
--- a/libidn/toutf8.c
+++ b/libidn/toutf8.c
@@ -1,5 +1,5 @@
-/* toutf8.c --- Convert strings from system locale into UTF-8.
- * Copyright (C) 2002, 2003, 2004, 2005  Simon Josefsson
+/* toutf8.c	Convert strings from system locale into UTF-8.
+ * Copyright (C) 2002, 2003, 2004  Simon Josefsson
  *
  * This file is part of GNU Libidn.
  *
@@ -23,36 +23,33 @@
 # include "config.h"
 #endif
 
-/* Get prototypes. */
-#include "stringprep.h"
-
-/* Get fprintf. */
 #include <stdio.h>
-
-/* Get getenv. */
 #include <stdlib.h>
-
-/* Get strlen. */
 #include <string.h>
+#include <errno.h>
+#include <sys/param.h>
 
-/* Get iconv_string. */
-#include "iconvme.h"
+#include "stringprep.h"
 
 #ifdef _LIBC
 # define HAVE_ICONV 1
 # define LOCALE_WORKS 1
+# define ICONV_CONST
 #endif
 
-#if LOCALE_WORKS
-# include <langinfo.h>
-# include <locale.h>
-#endif
+#ifdef HAVE_ICONV
+# include <iconv.h>
 
-#ifdef _LIBC
-# define stringprep_locale_charset() nl_langinfo (CODESET)
-#else
+# if LOCALE_WORKS
+#  include <langinfo.h>
+#  include <locale.h>
+# endif
+
+# ifdef _LIBC
+#  define stringprep_locale_charset() nl_langinfo (CODESET)
+# else
 /**
- * stringprep_locale_charset - return charset used in current locale
+ * stringprep_locale_charset:
  *
  * Find out current locale charset.  The function respect the CHARSET
  * environment variable, but typically uses nl_langinfo(CODESET) when
@@ -79,19 +76,19 @@ stringprep_locale_charset (void)
   if (charset && *charset)
     return charset;
 
-# ifdef LOCALE_WORKS
+#  ifdef LOCALE_WORKS
   charset = nl_langinfo (CODESET);
 
   if (charset && *charset)
     return charset;
-# endif
+#  endif
 
   return "ASCII";
 }
-#endif
+# endif
 
 /**
- * stringprep_convert - encode string using new character set
+ * stringprep_convert:
  * @str: input zero-terminated string.
  * @to_codeset: name of destination character set.
  * @from_codeset: name of origin character set, as used by @str.
@@ -106,21 +103,132 @@ char *
 stringprep_convert (const char *str,
 		    const char *to_codeset, const char *from_codeset)
 {
-#if HAVE_ICONV
-  return iconv_string (str, from_codeset, to_codeset);
+  iconv_t cd;
+  char *dest;
+  char *outp;
+  ICONV_CONST char *p;
+  size_t inbytes_remaining;
+  size_t outbytes_remaining;
+  size_t err;
+  size_t outbuf_size;
+  int have_error = 0;
+
+  if (strcmp (to_codeset, from_codeset) == 0)
+    {
+#if defined HAVE_STRDUP || defined _LIBC
+      return strdup (str);
 #else
+      char *p;
+      p = malloc (strlen (str) + 1);
+      if (!p)
+	return NULL;
+      return strcpy (p, str);
+#endif
+    }
+
+  cd = iconv_open (to_codeset, from_codeset);
+
+  if (cd == (iconv_t) - 1)
+    return NULL;
+
+  p = (ICONV_CONST char *) str;
+
+  inbytes_remaining = strlen (p);
+  /* Guess the maximum length the output string can have.  */
+  outbuf_size = (inbytes_remaining + 1) * MAX (7, MB_CUR_MAX);
+
+  outp = dest = malloc (outbuf_size);
+  if (dest == NULL)
+    goto out;
+  outbytes_remaining = outbuf_size - 1;	/* -1 for NUL */
+
+again:
+
+  err = iconv (cd, (ICONV_CONST char **) &p, &inbytes_remaining,
+	       &outp, &outbytes_remaining);
+
+  if (err == (size_t) - 1)
+    {
+      switch (errno)
+	{
+	case EINVAL:
+	  /* Incomplete text, do not report an error */
+	  break;
+
+	case E2BIG:
+	  {
+	    size_t used = outp - dest;
+	    char *newdest;
+
+	    outbuf_size *= 2;
+	    newdest = realloc (dest, outbuf_size);
+	    if (newdest == NULL)
+	      {
+		have_error = 1;
+		goto out;
+	      }
+	    dest = newdest;
+
+	    outp = dest + used;
+	    outbytes_remaining = outbuf_size - used - 1; /* -1 for NUL */
+
+	    goto again;
+	  }
+	  break;
+
+	case EILSEQ:
+	  have_error = 1;
+	  break;
+
+	default:
+	  have_error = 1;
+	  break;
+	}
+    }
+
+  *outp = '\0';
+
+  if (*p != '\0')
+    have_error = 1;
+
+ out:
+  iconv_close (cd);
+
+  if (have_error)
+    {
+      free (dest);
+      dest = NULL;
+    }
+
+  return dest;
+}
+
+#else /* HAVE_ICONV */
+
+const char *
+stringprep_locale_charset ()
+{
+  return "ASCII";
+}
+
+char *
+stringprep_convert (const char *str,
+		    const char *to_codeset, const char *from_codeset)
+{
   char *p;
   fprintf (stderr, "libidn: warning: libiconv not installed, cannot "
 	   "convert data to UTF-8\n");
   p = malloc (strlen (str) + 1);
   if (!p)
     return NULL;
-  return strcpy (p, str);
-#endif
+  strcpy (p, str);
+  return p;
 }
 
+#endif /* HAVE_ICONV */
+
 /**
- * stringprep_locale_to_utf8 - convert locale encoded string to UTF-8
+ * stringprep_locale_to_utf8:
  * @str: input zero terminated string.
  *
  * Convert string encoded in the locale's character set into UTF-8 by
@@ -136,7 +244,7 @@ stringprep_locale_to_utf8 (const char *str)
 }
 
 /**
- * stringprep_utf8_to_locale - encode UTF-8 string to locale encoding
+ * stringprep_utf8_to_locale:
  * @str: input zero terminated string.
  *
  * Convert string encoded in UTF-8 into the locale's character set by