X-Git-Url: http://erislabs.net/gitweb/?a=blobdiff_plain;f=lib%2Fstriconv.c;h=9bbed465a366770a55d7f3de3da9d0dea70f1363;hb=1bb384c44db25aef8f8b455cef22af32d8401abc;hp=35769bf757761acab609edf54760dbba0c0506b3;hpb=48079a8c38f504873d4ab2b4e892d4cf84099898;p=gnulib.git

diff --git a/lib/striconv.c b/lib/striconv.c
index 35769bf75..9bbed465a 100644
--- a/lib/striconv.c
+++ b/lib/striconv.c
@@ -1,5 +1,5 @@
 /* Charset conversion.
-   Copyright (C) 2001-2007 Free Software Foundation, Inc.
+   Copyright (C) 2001-2007, 2010-2011 Free Software Foundation, Inc.
    Written by Bruno Haible and Simon Josefsson.
 
    This program is free software; you can redistribute it and/or modify
@@ -31,7 +31,6 @@
 # include <limits.h>
 #endif
 
-#include "strdup.h"
 #include "c-strcase.h"
 
 #ifndef SIZE_MAX
@@ -43,7 +42,7 @@
 
 int
 mem_cd_iconv (const char *src, size_t srclen, iconv_t cd,
-	      char **resultp, size_t *lengthp)
+              char **resultp, size_t *lengthp)
 {
 # define tmpbufsize 4096
   size_t length;
@@ -51,7 +50,8 @@ mem_cd_iconv (const char *src, size_t srclen, iconv_t cd,
 
   /* Avoid glibc-2.1 bug and Solaris 2.7-2.9 bug.  */
 # if defined _LIBICONV_VERSION \
-    || !((__GLIBC__ - 0 == 2 && __GLIBC_MINOR__ - 0 <= 1) || defined __sun)
+     || !(((__GLIBC__ == 2 && __GLIBC_MINOR__ <= 1) && !defined __UCLIBC__) \
+          || defined __sun)
   /* Set to the initial state.  */
   iconv (cd, NULL, NULL, NULL, NULL);
 # endif
@@ -68,44 +68,45 @@ mem_cd_iconv (const char *src, size_t srclen, iconv_t cd,
 
     while (insize > 0)
       {
-	char *outptr = tmpbuf;
-	size_t outsize = tmpbufsize;
-	size_t res = iconv (cd,
-			    (ICONV_CONST char **) &inptr, &insize,
-			    &outptr, &outsize);
-
-	if (res == (size_t)(-1))
-	  {
-	    if (errno == E2BIG)
-	      ;
-	    else if (errno == EINVAL)
-	      break;
-	    else
-	      return -1;
-	  }
-# if !defined _LIBICONV_VERSION && !defined __GLIBC__
-	/* Irix iconv() inserts a NUL byte if it cannot convert.
-	   NetBSD iconv() inserts a question mark if it cannot convert.
-	   Only GNU libiconv and GNU libc are known to prefer to fail rather
-	   than doing a lossy conversion.  */
-	else if (res > 0)
-	  {
-	    errno = EILSEQ;
-	    return -1;
-	  }
+        char *outptr = tmpbuf;
+        size_t outsize = tmpbufsize;
+        size_t res = iconv (cd,
+                            (ICONV_CONST char **) &inptr, &insize,
+                            &outptr, &outsize);
+
+        if (res == (size_t)(-1))
+          {
+            if (errno == E2BIG)
+              ;
+            else if (errno == EINVAL)
+              break;
+            else
+              return -1;
+          }
+# if !defined _LIBICONV_VERSION && !(defined __GLIBC__ && !defined __UCLIBC__)
+        /* Irix iconv() inserts a NUL byte if it cannot convert.
+           NetBSD iconv() inserts a question mark if it cannot convert.
+           Only GNU libiconv and GNU libc are known to prefer to fail rather
+           than doing a lossy conversion.  */
+        else if (res > 0)
+          {
+            errno = EILSEQ;
+            return -1;
+          }
 # endif
-	count += outptr - tmpbuf;
+        count += outptr - tmpbuf;
       }
     /* Avoid glibc-2.1 bug and Solaris 2.7 bug.  */
 # if defined _LIBICONV_VERSION \
-    || !((__GLIBC__ - 0 == 2 && __GLIBC_MINOR__ - 0 <= 1) || defined __sun)
+     || !(((__GLIBC__ == 2 && __GLIBC_MINOR__ <= 1) && !defined __UCLIBC__) \
+          || defined __sun)
     {
       char *outptr = tmpbuf;
       size_t outsize = tmpbufsize;
       size_t res = iconv (cd, NULL, NULL, &outptr, &outsize);
 
       if (res == (size_t)(-1))
-	return -1;
+        return -1;
       count += outptr - tmpbuf;
     }
 # endif
@@ -118,19 +119,22 @@ mem_cd_iconv (const char *src, size_t srclen, iconv_t cd,
       *lengthp = 0;
       return 0;
     }
-  result =
-    (char *) (*resultp != NULL ? realloc (*resultp, length) : malloc (length));
-  if (result == NULL)
+  if (*resultp != NULL && *lengthp >= length)
+    result = *resultp;
+  else
     {
-      errno = ENOMEM;
-      return -1;
+      result = (char *) malloc (length);
+      if (result == NULL)
+        {
+          errno = ENOMEM;
+          return -1;
+        }
     }
-  *resultp = result;
-  *lengthp = length;
 
   /* Avoid glibc-2.1 bug and Solaris 2.7-2.9 bug.  */
 # if defined _LIBICONV_VERSION \
-    || !((__GLIBC__ - 0 == 2 && __GLIBC_MINOR__ - 0 <= 1) || defined __sun)
+     || !(((__GLIBC__ == 2 && __GLIBC_MINOR__ <= 1) && !defined __UCLIBC__) \
+          || defined __sun)
   /* Return to the initial state.  */
   iconv (cd, NULL, NULL, NULL, NULL);
 # endif
@@ -144,44 +148,59 @@ mem_cd_iconv (const char *src, size_t srclen, iconv_t cd,
 
     while (insize > 0)
       {
-	size_t res = iconv (cd,
-			    (ICONV_CONST char **) &inptr, &insize,
-			    &outptr, &outsize);
-
-	if (res == (size_t)(-1))
-	  {
-	    if (errno == EINVAL)
-	      break;
-	    else
-	      return -1;
-	  }
-# if !defined _LIBICONV_VERSION && !defined __GLIBC__
-	/* Irix iconv() inserts a NUL byte if it cannot convert.
-	   NetBSD iconv() inserts a question mark if it cannot convert.
-	   Only GNU libiconv and GNU libc are known to prefer to fail rather
-	   than doing a lossy conversion.  */
-	else if (res > 0)
-	  {
-	    errno = EILSEQ;
-	    return -1;
-	  }
+        size_t res = iconv (cd,
+                            (ICONV_CONST char **) &inptr, &insize,
+                            &outptr, &outsize);
+
+        if (res == (size_t)(-1))
+          {
+            if (errno == EINVAL)
+              break;
+            else
+              goto fail;
+          }
+# if !defined _LIBICONV_VERSION && !(defined __GLIBC__ && !defined __UCLIBC__)
+        /* Irix iconv() inserts a NUL byte if it cannot convert.
+           NetBSD iconv() inserts a question mark if it cannot convert.
+           Only GNU libiconv and GNU libc are known to prefer to fail rather
+           than doing a lossy conversion.  */
+        else if (res > 0)
+          {
+            errno = EILSEQ;
+            goto fail;
+          }
 # endif
       }
     /* Avoid glibc-2.1 bug and Solaris 2.7 bug.  */
 # if defined _LIBICONV_VERSION \
-    || !((__GLIBC__ - 0 == 2 && __GLIBC_MINOR__ - 0 <= 1) || defined __sun)
+     || !(((__GLIBC__ == 2 && __GLIBC_MINOR__ <= 1) && !defined __UCLIBC__) \
+          || defined __sun)
     {
       size_t res = iconv (cd, NULL, NULL, &outptr, &outsize);
 
       if (res == (size_t)(-1))
-	return -1;
+        goto fail;
     }
 # endif
     if (outsize != 0)
       abort ();
   }
 
+  *resultp = result;
+  *lengthp = length;
+
   return 0;
+
+ fail:
+  {
+    if (result != *resultp)
+      {
+        int saved_errno = errno;
+        free (result);
+        errno = saved_errno;
+      }
+    return -1;
+  }
 # undef tmpbufsize
 }
 
@@ -192,21 +211,24 @@ str_cd_iconv (const char *src, iconv_t cd)
      to a trailing NUL byte in the output.  But not for UTF-7.  So that this
      function is usable for UTF-7, we have to exclude the NUL byte from the
      conversion and add it by hand afterwards.  */
-# if PROBABLY_SLOWER
+# if !defined _LIBICONV_VERSION && !(defined __GLIBC__ && !defined __UCLIBC__)
+  /* Irix iconv() inserts a NUL byte if it cannot convert.
+     NetBSD iconv() inserts a question mark if it cannot convert.
+     Only GNU libiconv and GNU libc are known to prefer to fail rather
+     than doing a lossy conversion.  For other iconv() implementations,
+     we have to look at the number of irreversible conversions returned;
+     but this information is lost when iconv() returns for an E2BIG reason.
+     Therefore we cannot use the second, faster algorithm.  */
 
   char *result = NULL;
-  size_t length;
+  size_t length = 0;
   int retval = mem_cd_iconv (src, strlen (src), cd, &result, &length);
   char *final_result;
 
   if (retval < 0)
     {
       if (result != NULL)
-	{
-	  int saved_errno = errno;
-	  free (result);
-	  errno = saved_errno;
-	}
+        abort ();
       return NULL;
     }
 
@@ -215,8 +237,7 @@ str_cd_iconv (const char *src, iconv_t cd)
     (result != NULL ? realloc (result, length + 1) : malloc (length + 1));
   if (final_result == NULL)
     {
-      if (result != NULL)
-	free (result);
+      free (result);
       errno = ENOMEM;
       return NULL;
     }
@@ -225,7 +246,10 @@ str_cd_iconv (const char *src, iconv_t cd)
   return final_result;
 
 # else
-
+  /* This algorithm is likely faster than the one above.  But it may produce
+     iconv() returns for an E2BIG reason, when the output size guess is too
+     small.  Therefore it can only be used when we don't need the number of
+     irreversible conversions performed.  */
   char *result;
   size_t result_size;
   size_t length;
@@ -252,7 +276,8 @@ str_cd_iconv (const char *src, iconv_t cd)
 
   /* Avoid glibc-2.1 bug and Solaris 2.7-2.9 bug.  */
 # if defined _LIBICONV_VERSION \
-    || !((__GLIBC__ - 0 == 2 && __GLIBC_MINOR__ - 0 <= 1) || defined __sun)
+     || !(((__GLIBC__ == 2 && __GLIBC_MINOR__ <= 1) && !defined __UCLIBC__) \
+          || defined __sun)
   /* Set to the initial state.  */
   iconv (cd, NULL, NULL, NULL, NULL);
 # endif
@@ -264,92 +289,82 @@ str_cd_iconv (const char *src, iconv_t cd)
 
     for (;;)
       {
-	/* Here inptr + inbytes_remaining = src + strlen (src),
-		outptr + outbytes_remaining = result + result_size - 1.  */
-	size_t res = iconv (cd,
-			    (ICONV_CONST char **) &inptr, &inbytes_remaining,
-			    &outptr, &outbytes_remaining);
-
-	if (res == (size_t)(-1))
-	  {
-	    if (errno == EINVAL)
-	      break;
-	    else if (errno == E2BIG)
-	      {
-		size_t used = outptr - result;
-		size_t newsize = result_size * 2;
-		char *newresult;
-
-		if (!(newsize > result_size))
-		  {
-		    errno = ENOMEM;
-		    goto failed;
-		  }
-		newresult = (char *) realloc (result, newsize);
-		if (newresult == NULL)
-		  {
-		    errno = ENOMEM;
-		    goto failed;
-		  }
-		result = newresult;
-		result_size = newsize;
-		outptr = result + used;
-		outbytes_remaining = result_size - 1 - used;
-	      }
-	    else
-	      goto failed;
-	  }
-# if !defined _LIBICONV_VERSION && !defined __GLIBC__
-	/* Irix iconv() inserts a NUL byte if it cannot convert.
-	   NetBSD iconv() inserts a question mark if it cannot convert.
-	   Only GNU libiconv and GNU libc are known to prefer to fail rather
-	   than doing a lossy conversion.  */
-	else if (res > 0)
-	  {
-	    errno = EILSEQ;
-	    goto failed;
-	  }
-# endif
-	else
-	  break;
+        /* Here inptr + inbytes_remaining = src + strlen (src),
+                outptr + outbytes_remaining = result + result_size - 1.  */
+        size_t res = iconv (cd,
+                            (ICONV_CONST char **) &inptr, &inbytes_remaining,
+                            &outptr, &outbytes_remaining);
+
+        if (res == (size_t)(-1))
+          {
+            if (errno == EINVAL)
+              break;
+            else if (errno == E2BIG)
+              {
+                size_t used = outptr - result;
+                size_t newsize = result_size * 2;
+                char *newresult;
+
+                if (!(newsize > result_size))
+                  {
+                    errno = ENOMEM;
+                    goto failed;
+                  }
+                newresult = (char *) realloc (result, newsize);
+                if (newresult == NULL)
+                  {
+                    errno = ENOMEM;
+                    goto failed;
+                  }
+                result = newresult;
+                result_size = newsize;
+                outptr = result + used;
+                outbytes_remaining = result_size - 1 - used;
+              }
+            else
+              goto failed;
+          }
+        else
+          break;
       }
     /* Avoid glibc-2.1 bug and Solaris 2.7 bug.  */
 # if defined _LIBICONV_VERSION \
-    || !((__GLIBC__ - 0 == 2 && __GLIBC_MINOR__ - 0 <= 1) || defined __sun)
+     || !(((__GLIBC__ == 2 && __GLIBC_MINOR__ <= 1) && !defined __UCLIBC__) \
+          || defined __sun)
     for (;;)
       {
-	/* Here outptr + outbytes_remaining = result + result_size - 1.  */
-	size_t res = iconv (cd, NULL, NULL, &outptr, &outbytes_remaining);
-
-	if (res == (size_t)(-1))
-	  {
-	    if (errno == E2BIG)
-	      {
-		size_t used = outptr - result;
-		size_t newsize = result_size * 2;
-		char *newresult;
-
-		if (!(newsize > result_size))
-		  {
-		    errno = ENOMEM;
-		    goto failed;
-		  }
-		newresult = (char *) realloc (result, newsize);
-		if (newresult == NULL)
-		  {
-		    errno = ENOMEM;
-		    goto failed;
-		  }
-		result = newresult;
-		result_size = newsize;
-		outptr = result + used;
-		outbytes_remaining = result_size - 1 - used;
-	      }
-	    else
-	      goto failed;
-	  }
-	else
-	  break;
+        /* Here outptr + outbytes_remaining = result + result_size - 1.  */
+        size_t res = iconv (cd, NULL, NULL, &outptr, &outbytes_remaining);
+
+        if (res == (size_t)(-1))
+          {
+            if (errno == E2BIG)
+              {
+                size_t used = outptr - result;
+                size_t newsize = result_size * 2;
+                char *newresult;
+
+                if (!(newsize > result_size))
+                  {
+                    errno = ENOMEM;
+                    goto failed;
+                  }
+                newresult = (char *) realloc (result, newsize);
+                if (newresult == NULL)
+                  {
+                    errno = ENOMEM;
+                    goto failed;
+                  }
+                result = newresult;
+                result_size = newsize;
+                outptr = result + used;
+                outbytes_remaining = result_size - 1 - used;
+              }
+            else
+              goto failed;
+          }
+        else
+          break;
       }
 # endif
 
@@ -365,7 +380,7 @@ str_cd_iconv (const char *src, iconv_t cd)
       char *smaller_result = (char *) realloc (result, length);
 
       if (smaller_result != NULL)
-	result = smaller_result;
+        result = smaller_result;
     }
 
   return result;
@@ -386,8 +401,14 @@ str_cd_iconv (const char *src, iconv_t cd)
 char *
 str_iconv (const char *src, const char *from_codeset, const char *to_codeset)
 {
-  if (c_strcasecmp (from_codeset, to_codeset) == 0)
-    return strdup (src);
+  if (*src == '\0' || c_strcasecmp (from_codeset, to_codeset) == 0)
+    {
+      char *result = strdup (src);
+
+      if (result == NULL)
+        errno = ENOMEM;
+      return result;
+    }
   else
     {
 #if HAVE_ICONV
@@ -395,46 +416,47 @@ str_iconv (const char *src, const char *from_codeset, const char *to_codeset)
       char *result;
 
       /* Avoid glibc-2.1 bug with EUC-KR.  */
-# if (__GLIBC__ - 0 == 2 && __GLIBC_MINOR__ - 0 <= 1) && !defined _LIBICONV_VERSION
+# if ((__GLIBC__ == 2 && __GLIBC_MINOR__ <= 1) && !defined __UCLIBC__) \
+     && !defined _LIBICONV_VERSION
       if (c_strcasecmp (from_codeset, "EUC-KR") == 0
-	  || c_strcasecmp (to_codeset, "EUC-KR") == 0)
-	{
-	  errno = EINVAL;
-	  return NULL;
-	}
+          || c_strcasecmp (to_codeset, "EUC-KR") == 0)
+        {
+          errno = EINVAL;
+          return NULL;
+        }
 # endif
       cd = iconv_open (to_codeset, from_codeset);
       if (cd == (iconv_t) -1)
-	return NULL;
+        return NULL;
 
       result = str_cd_iconv (src, cd);
 
       if (result == NULL)
-	{
-	  /* Close cd, but preserve the errno from str_cd_iconv.  */
-	  int saved_errno = errno;
-	  iconv_close (cd);
-	  errno = saved_errno;
-	}
+        {
+          /* Close cd, but preserve the errno from str_cd_iconv.  */
+          int saved_errno = errno;
+          iconv_close (cd);
+          errno = saved_errno;
+        }
       else
-	{
-	  if (iconv_close (cd) < 0)
-	    {
-	      /* Return NULL, but free the allocated memory, and while doing
-		 that, preserve the errno from iconv_close.  */
-	      int saved_errno = errno;
-	      free (result);
-	      errno = saved_errno;
-	      return NULL;
-	    }
-	}
+        {
+          if (iconv_close (cd) < 0)
+            {
+              /* Return NULL, but free the allocated memory, and while doing
+                 that, preserve the errno from iconv_close.  */
+              int saved_errno = errno;
+              free (result);
+              errno = saved_errno;
+              return NULL;
+            }
+        }
       return result;
 #else
       /* This is a different error code than if iconv_open existed but didn't
-	 support from_codeset and to_codeset, so that the caller can emit
-	 an error message such as
-	   "iconv() is not supported. Installing GNU libiconv and
-	    then reinstalling this package would fix this."  */
+         support from_codeset and to_codeset, so that the caller can emit
+         an error message such as
+           "iconv() is not supported. Installing GNU libiconv and
+            then reinstalling this package would fix this."  */
       errno = ENOSYS;
       return NULL;
 #endif