Update

2001-01-04  H.J. Lu  <hjl@gnu.org>

	* elf/dl-support.c (DL_FIND_AUXV): New.  Defined if not defined.
	(_dl_clktck): Declared.
	(non_dynamic_init): Take 3 arguments.
	Set _dl_pagesize, _dl_platform and _dl_clktck from AUX.

2001-01-01  Bruno Haible  <haible@clisp.cons.org>

	Finish implementation of plural form handling.
	* intl/dcigettext.c (known_translation_t): Rename 'domain' field to
	'domainname'.  Remove 'plindex' field. Add 'domain' and
	'translation_length' fields.
	(transcmp): Don't compare 'plindex' fields.
	(plural_lookup): New function.
	(DCIGETTEXT): Change cache handing in the plural case.  Don't call
	plural_eval before the translation and its catalog file have been
	found. Remove plindex from cache key.  Add 'translation_length' and
	'domain' to cache result.
	(_nl_find_msg): Remove index argument, return length of translation
	to the caller instead.  Weaken comparison of string lengths, to account
	for plural entries.  Call iconv() on the entire result string, not
	only on the portion needed so far.
	* intl/loadinfo.h (_nl_find_msg): Remove index argument, add lengthp
	argument.
	* intl/loadmsgcat.c (_nl_load_domain): Adapt to _nl_find_msg change.

1 files changed, 160 insertions, 141 deletions

diff --git a/intl/dcigettext.c b/intl/dcigettext.c
index 8712e4ba8c..c6540b9ab0 100644
--- a/intl/dcigettext.c
+++ b/intl/dcigettext.c
@@ -210,10 +210,7 @@ static void *mempcpy PARAMS ((void *dest, const void *src, size_t n));
 struct known_translation_t
 {
   /* Domain in which to search.  */
-  char *domain;
-
-  /* Plural index.  */
-  unsigned long int plindex;
+  char *domainname;
 
   /* The category.  */
   int category;
@@ -221,8 +218,12 @@ struct known_translation_t
   /* State of the catalog counter at the point the string was found.  */
   int counter;
 
+  /* Catalog where the string was found.  */
+  struct loaded_l10nfile *domain;
+
   /* And finally the translation.  */
   const char *translation;
+  size_t translation_length;
 
   /* Pointer to the string in question.  */
   char msgid[ZERO];
@@ -253,16 +254,12 @@ transcmp (const void *p1, const void *p2)
   result = strcmp (s1->msgid, s2->msgid);
   if (result == 0)
     {
-      result = strcmp (s1->domain, s2->domain);
+      result = strcmp (s1->domainname, s2->domainname);
       if (result == 0)
-	{
-	  result = s1->plindex - s2->plindex;
-	  if (result == 0)
-	    /* We compare the category last (though this is the cheapest
-	       operation) since it is hopefully always the same (namely
-	       LC_MESSAGES).  */
-	    result = s1->category - s2->category;
-	}
+	/* We compare the category last (though this is the cheapest
+	   operation) since it is hopefully always the same (namely
+	   LC_MESSAGES).  */
+	result = s1->category - s2->category;
     }
 
   return result;
@@ -284,8 +281,14 @@ const char _nl_default_dirname[] = GNULOCALEDIR;
 struct binding *_nl_domain_bindings;
 
 /* Prototypes for local functions.  */
-static unsigned long int plural_eval (struct expression *pexp,
-				      unsigned long int n) internal_function;
+static char *plural_lookup PARAMS ((struct loaded_l10nfile *domain,
+				    unsigned long int n,
+				    const char *translation,
+				    size_t translation_len))
+     internal_function;
+static unsigned long int plural_eval PARAMS ((struct expression *pexp,
+					      unsigned long int n))
+     internal_function;
 static const char *category_to_name PARAMS ((int category)) internal_function;
 static const char *guess_category_value PARAMS ((int category,
 						 const char *categoryname))
@@ -328,12 +331,17 @@ struct block_list
 #endif	/* have alloca */
 
 
+#ifdef _LIBC
 /* List of blocks allocated for translations.  */
-static struct transmem_list
+typedef struct transmem_list
 {
   struct transmem_list *next;
   char data[0];
-} *transmem_list;
+} transmem_block_t;
+static struct transmem_list *transmem_list;
+#else
+typedef char transmem_block_t;
+#endif
 
 
 /* Names for the libintl functions are a problem.  They must not clash
@@ -389,6 +397,7 @@ DCIGETTEXT (domainname, msgid1, msgid2, plural, n, category)
   char *dirname, *xdomainname;
   char *single_locale;
   char *retval;
+  size_t retlen;
   int saved_errno;
 #if defined HAVE_TSEARCH || defined _LIBC
   struct known_translation_t *search;
@@ -412,23 +421,26 @@ DCIGETTEXT (domainname, msgid1, msgid2, plural, n, category)
 #if defined HAVE_TSEARCH || defined _LIBC
   msgid_len = strlen (msgid1) + 1;
 
-  if (plural == 0)
+  /* Try to find the translation among those which we found at
+     some time.  */
+  search =
+    (struct known_translation_t *) alloca (sizeof (*search) + msgid_len);
+  memcpy (search->msgid, msgid1, msgid_len);
+  search->domainname = (char *) domainname;
+  search->category = category;
+
+  foundp = (struct known_translation_t **) tfind (search, &root, transcmp);
+  if (foundp != NULL && (*foundp)->counter == _nl_msg_cat_cntr)
     {
-      /* Try to find the translation among those which we found at
-	 some time.  */
-      search = (struct known_translation_t *) alloca (sizeof (*search)
-						      + msgid_len);
-      memcpy (search->msgid, msgid1, msgid_len);
-      search->domain = (char *) domainname;
-      search->plindex = 0;
-      search->category = category;
-
-      foundp = (struct known_translation_t **) tfind (search, &root, transcmp);
-      if (foundp != NULL && (*foundp)->counter == _nl_msg_cat_cntr)
-	{
-	  __libc_rwlock_unlock (_nl_state_lock);
-	  return (char *) (*foundp)->translation;
-	}
+      /* Now deal with plural.  */
+      if (plural)
+	retval = plural_lookup ((*foundp)->domain, n, (*foundp)->translation,
+				(*foundp)->translation_length);
+      else
+	retval = (char *) (*foundp)->translation;
+
+      __libc_rwlock_unlock (_nl_state_lock);
+      return retval;
     }
 #endif
 
@@ -563,39 +575,7 @@ DCIGETTEXT (domainname, msgid1, msgid2, plural, n, category)
 
       if (domain != NULL)
 	{
-	  unsigned long int index = 0;
-
-	  if (plural != 0)
-	    {
-	      const struct loaded_domain *domaindata =
-		(const struct loaded_domain *) domain->data;
-	      index = plural_eval (domaindata->plural, n);
-	      if (index >= domaindata->nplurals)
-		/* This should never happen.  It means the plural expression
-		   and the given maximum value do not match.  */
-		index = 0;
-
-#if defined HAVE_TSEARCH || defined _LIBC
-	      /* Try to find the translation among those which we
-		 found at some time.  */
-	      search = (struct known_translation_t *) alloca (sizeof (*search)
-							      + msgid_len);
-	      memcpy (search->msgid, msgid1, msgid_len);
-	      search->domain = (char *) domainname;
-	      search->plindex = index;
-	      search->category = category;
-
-	      foundp = (struct known_translation_t **) tfind (search, &root,
-							      transcmp);
-	      if (foundp != NULL && (*foundp)->counter == _nl_msg_cat_cntr)
-		{
-		  __libc_rwlock_unlock (_nl_state_lock);
-		  return (char *) (*foundp)->translation;
-		}
-#endif
-	    }
-
-	  retval = _nl_find_msg (domain, msgid1, index);
+	  retval = _nl_find_msg (domain, msgid1, &retlen);
 
 	  if (retval == NULL)
 	    {
@@ -604,15 +584,20 @@ DCIGETTEXT (domainname, msgid1, msgid2, plural, n, category)
 	      for (cnt = 0; domain->successor[cnt] != NULL; ++cnt)
 		{
 		  retval = _nl_find_msg (domain->successor[cnt], msgid1,
-					 index);
+					 &retlen);
 
 		  if (retval != NULL)
-		    break;
+		    {
+		      domain = domain->successor[cnt];
+		      break;
+		    }
 		}
 	    }
 
 	  if (retval != NULL)
 	    {
+	      /* Found the translation of MSGID1 in domain DOMAIN:
+		 starting at RETVAL, RETLEN bytes.  */
 	      FREE_BLOCKS (block_list);
 	      __set_errno (saved_errno);
 #if defined HAVE_TSEARCH || defined _LIBC
@@ -626,12 +611,14 @@ DCIGETTEXT (domainname, msgid1, msgid2, plural, n, category)
 			    + domainname_len + 1 - ZERO);
 		  if (newp != NULL)
 		    {
-		      newp->domain = mempcpy (newp->msgid, msgid1, msgid_len);
-		      memcpy (newp->domain, domainname, domainname_len + 1);
-		      newp->plindex = index;
+		      newp->domainname =
+			mempcpy (newp->msgid, msgid1, msgid_len);
+		      memcpy (newp->domainname, domainname, domainname_len + 1);
 		      newp->category = category;
 		      newp->counter = _nl_msg_cat_cntr;
+		      newp->domain = domain;
 		      newp->translation = retval;
+		      newp->translation_length = retlen;
 
 		      /* Insert the entry in the search tree.  */
 		      foundp = (struct known_translation_t **)
@@ -646,9 +633,15 @@ DCIGETTEXT (domainname, msgid1, msgid2, plural, n, category)
 		{
 		  /* We can update the existing entry.  */
 		  (*foundp)->counter = _nl_msg_cat_cntr;
+		  (*foundp)->domain = domain;
 		  (*foundp)->translation = retval;
+		  (*foundp)->translation_length = retlen;
 		}
 #endif
+	      /* Now deal with plural.  */
+	      if (plural)
+		retval = plural_lookup (domain, n, retval, retlen);
+
 	      __libc_rwlock_unlock (_nl_state_lock);
 	      return retval;
 	    }
@@ -660,14 +653,15 @@ DCIGETTEXT (domainname, msgid1, msgid2, plural, n, category)
 
 char *
 internal_function
-_nl_find_msg (domain_file, msgid, index)
+_nl_find_msg (domain_file, msgid, lengthp)
      struct loaded_l10nfile *domain_file;
      const char *msgid;
-     unsigned long int index;
+     size_t *lengthp;
 {
   const struct loaded_domain *domain;
   size_t act;
   char *result;
+  size_t resultlen;
 
   if (domain_file->decided == 0)
     _nl_load_domain (domain_file);
@@ -691,17 +685,21 @@ _nl_find_msg (domain_file, msgid, index)
 	/* Hash table entry is empty.  */
 	return NULL;
 
-      if (W (domain->must_swap, domain->orig_tab[nstr - 1].length) == len
-	  && strcmp (msgid,
-		     domain->data + W (domain->must_swap,
-				       domain->orig_tab[nstr - 1].offset)) == 0)
-	{
-	  act = nstr - 1;
-	  goto found;
-	}
-
       while (1)
 	{
+	  /* Compare msgid with the original string at index nstr-1.
+	     We compare the lengths with >=, not ==, because plural entries
+	     are represented by strings with an embedded NUL.  */
+	  if (W (domain->must_swap, domain->orig_tab[nstr - 1].length) >= len
+	      && (strcmp (msgid,
+			  domain->data + W (domain->must_swap,
+					    domain->orig_tab[nstr - 1].offset))
+		  == 0))
+	    {
+	      act = nstr - 1;
+	      goto found;
+	    }
+
 	  if (idx >= domain->hash_size - incr)
 	    idx -= domain->hash_size - incr;
 	  else
@@ -711,16 +709,6 @@ _nl_find_msg (domain_file, msgid, index)
 	  if (nstr == 0)
 	    /* Hash table entry is empty.  */
 	    return NULL;
-
-	  if (W (domain->must_swap, domain->orig_tab[nstr - 1].length) == len
-	      && (strcmp (msgid,
-			  domain->data + W (domain->must_swap,
-					    domain->orig_tab[nstr - 1].offset))
-		  == 0))
-	    {
-	      act = nstr - 1;
-	      goto found;
-	    }
 	}
       /* NOTREACHED */
     }
@@ -756,6 +744,7 @@ _nl_find_msg (domain_file, msgid, index)
      string to use a different character set, this is the time.  */
   result = ((char *) domain->data
 	    + W (domain->must_swap, domain->trans_tab[act].offset));
+  resultlen = W (domain->must_swap, domain->trans_tab[act].length) + 1;
 
 #if defined _LIBC || HAVE_ICONV
   if (
@@ -772,9 +761,10 @@ _nl_find_msg (domain_file, msgid, index)
 	 appropriate table with the same structure as the table
 	 of translations in the file, where we can put the pointers
 	 to the converted strings in.
-	 There is a slight complication with the INDEX: We don't know
-	 a priori which entries are plural entries. Therefore at any
-	 moment we can only translate the variants 0 .. INDEX.  */
+	 There is a slight complication with plural entries.  They
+	 are represented by consecutive NUL terminated strings.  We
+	 handle this case by converting RESULTLEN bytes, including
+	 NULs.  */
 
       if (domain->conv_tab == NULL
 	  && ((domain->conv_tab = (char **) calloc (domain->nstrings,
@@ -787,8 +777,7 @@ _nl_find_msg (domain_file, msgid, index)
 	/* Nothing we can do, no more memory.  */
 	goto converted;
 
-      if (domain->conv_tab[act] == NULL
-	  || *(nls_uint32 *) domain->conv_tab[act] < index)
+      if (domain->conv_tab[act] == NULL)
 	{
 	  /* We haven't used this string so far, so it is not
 	     translated yet.  Do this now.  */
@@ -796,46 +785,37 @@ _nl_find_msg (domain_file, msgid, index)
 	     We allocate always larger blocks which get used over
 	     time.  This is faster than many small allocations.   */
 	  __libc_lock_define_initialized (static, lock)
-#define INITIAL_BLOCK_SIZE	4080
+# define INITIAL_BLOCK_SIZE	4080
 	  static unsigned char *freemem;
 	  static size_t freemem_size;
 
-	  size_t resultlen;
 	  const unsigned char *inbuf;
 	  unsigned char *outbuf;
 	  int malloc_count;
-
-	  /* Note that we translate (index + 1) consecutive strings at
-	     once, including the final NUL byte.  */
-	  {
-	    unsigned long int i = index;
-	    char *p = result;
-	    do
-	      p += strlen (p) + 1;
-	    while (i-- > 0);
-	    resultlen = p - result;
-	  }
+# ifndef _LIBC
+	  transmem_block_t *transmem_list = NULL;
+# endif
 
 	  __libc_lock_lock (lock);
 
 	  inbuf = result;
-	  outbuf = freemem + sizeof (nls_uint32);
+	  outbuf = freemem + sizeof (size_t);
 
 	  malloc_count = 0;
 	  while (1)
 	    {
+	      transmem_block_t *newmem;
 # ifdef _LIBC
-	      struct transmem_list *newmem;
 	      size_t non_reversible;
 	      int res;
 
-	      if (freemem_size < 4)
+	      if (freemem_size < sizeof (size_t))
 		goto resize_freemem;
 
 	      res = __gconv (domain->conv,
 			     &inbuf, inbuf + resultlen,
 			     &outbuf,
-			     outbuf + freemem_size - sizeof (nls_uint32),
+			     outbuf + freemem_size - sizeof (size_t),
 			     &non_reversible);
 
 	      if (res == __GCONV_OK || res == __GCONV_EMPTY_INPUT)
@@ -850,16 +830,15 @@ _nl_find_msg (domain_file, msgid, index)
 	      inbuf = result;
 # else
 #  if HAVE_ICONV
-#   define transmem freemem
 	      const char *inptr = (const char *) inbuf;
 	      size_t inleft = resultlen;
 	      char *outptr = (char *) outbuf;
 	      size_t outleft;
 
-	      if (freemem_size < 4)
+	      if (freemem_size < sizeof (size_t))
 		goto resize_freemem;
 
-	      outleft = freemem_size - 4;
+	      outleft = freemem_size - sizeof (size_t);
 	      if (iconv (domain->conv, &inptr, &inleft, &outptr, &outleft)
 		  != (size_t) (-1))
 		{
@@ -871,30 +850,34 @@ _nl_find_msg (domain_file, msgid, index)
 		  __libc_lock_unlock (lock);
 		  goto converted;
 		}
-#  else
-#   define transmem freemem
 #  endif
 # endif
 
 	    resize_freemem:
-	      /* We must allocate a new buffer of resize the old one.  */
+	      /* We must allocate a new buffer or resize the old one.  */
 	      if (malloc_count > 0)
 		{
-		  struct transmem_list *next = transmem_list->next;
-
 		  ++malloc_count;
 		  freemem_size = malloc_count * INITIAL_BLOCK_SIZE;
-		  newmem = (struct transmem_list *) realloc (transmem_list,
-							     freemem_size);
-
+		  newmem = (transmem_block_t *) realloc (transmem_list,
+							 freemem_size);
+# ifdef _LIBC
 		  if (newmem != NULL)
-		    transmem_list = next;
+		    transmem_list = transmem_list->next;
+		  else
+		    {
+		      struct transmem_list *old = transmem_list;
+
+		      transmem_list = transmem_list->next;
+		      free (old);
+		    }
+# endif
 		}
 	      else
 		{
 		  malloc_count = 1;
 		  freemem_size = INITIAL_BLOCK_SIZE;
-		  newmem = (struct transmem_list *) malloc (freemem_size);
+		  newmem = (transmem_block_t *) malloc (freemem_size);
 		}
 	      if (__builtin_expect (newmem == NULL, 0))
 		{
@@ -912,27 +895,31 @@ _nl_find_msg (domain_file, msgid, index)
 
 	      freemem = newmem->data;
 	      freemem_size -= offsetof (struct transmem_list, data);
+# else
+	      transmem_list = newmem;
+	      freemem = newmem;
 # endif
 
-	      outbuf = freemem + sizeof (nls_uint32);
+	      outbuf = freemem + sizeof (size_t);
 	    }
 
 	  /* We have now in our buffer a converted string.  Put this
 	     into the table of conversions.  */
-	  *(nls_uint32 *) freemem = index;
+	  *(size_t *) freemem = outbuf - freemem - sizeof (size_t);
 	  domain->conv_tab[act] = freemem;
 	  /* Shrink freemem, but keep it aligned.  */
 	  freemem_size -= outbuf - freemem;
 	  freemem = outbuf;
-	  freemem += freemem_size & (alignof (nls_uint32) - 1);
-	  freemem_size = freemem_size & ~ (alignof (nls_uint32) - 1);
+	  freemem += freemem_size & (alignof (size_t) - 1);
+	  freemem_size = freemem_size & ~ (alignof (size_t) - 1);
 
 	  __libc_lock_unlock (lock);
 	}
 
-      /* Now domain->conv_tab[act] contains the translation of at least
-	 the variants 0 .. INDEX.  */
-      result = domain->conv_tab[act] + sizeof (nls_uint32);
+      /* Now domain->conv_tab[act] contains the translation of all
+	 the plural variants.  */
+      result = domain->conv_tab[act] + sizeof (size_t);
+      resultlen = *(size_t *) domain->conv_tab[act];
     }
 
  converted:
@@ -940,26 +927,58 @@ _nl_find_msg (domain_file, msgid, index)
 
 #endif /* _LIBC || HAVE_ICONV */
 
-  /* Now skip some strings.  How much depends on the index passed in.  */
+  *lengthp = resultlen;
+  return result;
+}
+
+
+/* Look up a plural variant.  */
+static char *
+internal_function
+plural_lookup (domain, n, translation, translation_len)
+     struct loaded_l10nfile *domain;
+     unsigned long int n;
+     const char *translation;
+     size_t translation_len;
+{
+  struct loaded_domain *domaindata = (struct loaded_domain *) domain->data;
+  unsigned long int index;
+  const char *p;
+
+  index = plural_eval (domaindata->plural, n);
+  if (index >= domaindata->nplurals)
+    /* This should never happen.  It means the plural expression and the
+       given maximum value do not match.  */
+    index = 0;
+
+  /* Skip INDEX strings at TRANSLATION.  */
+  p = translation;
   while (index-- > 0)
     {
 #ifdef _LIBC
-      result = __rawmemchr (result, '\0');
+      p = __rawmemchr (p, '\0');
 #else
-      result = strchr (result, '\0');
+      p = strchr (p, '\0');
 #endif
       /* And skip over the NUL byte.  */
-      ++result;
-    }
+      p++;
 
-  return result;
+      if (p >= translation + translation_len)
+	/* This should never happen.  It means the plural expression
+	   evaluated to a value larger than the number of variants
+	   available for MSGID1.  */
+	return (char *) translation;
+    }
+  return (char *) p;
 }
 
 
 /* Function to evaluate the plural expression and return an index value.  */
 static unsigned long int
 internal_function
-plural_eval (struct expression *pexp, unsigned long int n)
+plural_eval (pexp, n)
+     struct expression *pexp;
+     unsigned long int n;
 {
   switch (pexp->operation)
     {