Fix inadequately-sized output buffer in contrib/unaccent.

The output buffer size in unaccent_lexize() was calculated as input string
length times pg_database_encoding_max_length(), which effectively assumes
that replacement strings aren't more than one character.  While that was
all that we previously documented it to support, the code actually has
always allowed replacement strings of arbitrary length; so if you tried
to make use of longer strings, you were at risk of buffer overrun.  To fix,
use an expansible StringInfo buffer instead of trying to determine the
maximum space needed a-priori.

This would be a security issue if unaccent rules files could be installed
by unprivileged users; but fortunately they can't, so in the back branches
the problem can be labeled as improper configuration by a superuser.
Nonetheless, a memory stomp isn't a nice way of reacting to improper
configuration, so let's back-patch the fix.

diff --git a/contrib/unaccent/unaccent.c b/contrib/unaccent/unaccent.c
index 2770d895b2ef6e4b73585f6ea3a654d2ee4962fe..ab1a5d0bd36a22912453e8e0872b02785bb9a0b2 100644 (file)
--- a/contrib/unaccent/unaccent.c
+++ b/contrib/unaccent/unaccent.c
@@ -15,6 +15,7 @@
 
 #include "catalog/namespace.h"
 #include "commands/defrem.h"
+#include "lib/stringinfo.h"
 #include "tsearch/ts_cache.h"
 #include "tsearch/ts_locale.h"
 #include "tsearch/ts_public.h"
@@ -265,46 +266,48 @@ unaccent_lexize(PG_FUNCTION_ARGS)
    TrieChar   *rootTrie = (TrieChar *) PG_GETARG_POINTER(0);
    char       *srcchar = (char *) PG_GETARG_POINTER(1);
    int32       len = PG_GETARG_INT32(2);
-   char       *srcstart,
-              *trgchar = NULL;
-   int         charlen;
-   TSLexeme   *res = NULL;
-   TrieChar   *node;
+   char       *srcstart = srcchar;
+   TSLexeme   *res;
+   StringInfoData buf;
+
+   /* we allocate storage for the buffer only if needed */
+   buf.data = NULL;
 
-   srcstart = srcchar;
    while (srcchar - srcstart < len)
    {
+       TrieChar   *node;
+       int         charlen;
+
        charlen = pg_mblen(srcchar);
 
        node = findReplaceTo(rootTrie, (unsigned char *) srcchar, charlen);
        if (node && node->replaceTo)
        {
-           if (!res)
+           if (buf.data == NULL)
            {
-               /* allocate res only if it's needed */
-               res = palloc0(sizeof(TSLexeme) * 2);
-               res->lexeme = trgchar = palloc(len * pg_database_encoding_max_length() + 1 /* \0 */ );
-               res->flags = TSL_FILTER;
+               /* initialize buffer */
+               initStringInfo(&buf);
+               /* insert any data we already skipped over */
                if (srcchar != srcstart)
-               {
-                   memcpy(trgchar, srcstart, srcchar - srcstart);
-                   trgchar += (srcchar - srcstart);
-               }
+                   appendBinaryStringInfo(&buf, srcstart, srcchar - srcstart);
            }
-           memcpy(trgchar, node->replaceTo, node->replacelen);
-           trgchar += node->replacelen;
-       }
-       else if (res)
-       {
-           memcpy(trgchar, srcchar, charlen);
-           trgchar += charlen;
+           appendBinaryStringInfo(&buf, node->replaceTo, node->replacelen);
        }
+       else if (buf.data != NULL)
+           appendBinaryStringInfo(&buf, srcchar, charlen);
 
        srcchar += charlen;
    }
 
-   if (res)
-       *trgchar = '\0';
+   /* return a result only if we made at least one substitution */
+   if (buf.data != NULL)
+   {
+       res = (TSLexeme *) palloc0(sizeof(TSLexeme) * 2);
+       res->lexeme = buf.data;
+       res->flags = TSL_FILTER;
+   }
+   else
+       res = NULL;
 
    PG_RETURN_POINTER(res);
 }