Change the 'len' parameter type to 'size_t'.

19 files changed, 66 insertions, 49 deletions

diff --git a/ChangeLog b/ChangeLog
index 9eb3de2..072a0ca 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,19 @@
 2016-11-26  Bruno Haible  <bruno@clisp.org>
 
+	Change the 'len' parameter type to 'size_t'.
+	* src/output.cc (output_upperlower_strncmp, output_upperlower_memcmp,
+	Output::output_hash_function, Output::output_lookup_function,
+	Output::output): Emit declarations with 'size_t len' instead of
+	'unsigned int len'.
+	* tests/*.exp: Update.
+	* doc/gperf.texi (Output Format): Change declarations of 'hash' and
+	'in_word_set'.
+	* NEWS: Mention the change.
+	Reported by Pedro Giffuni <pfg@FreeBSD.org>
+	at <http://lists.gnu.org/archive/html/bug-gperf/2014-08/msg00000.html>.
+
+2016-11-26  Bruno Haible  <bruno@clisp.org>
+
 	Use proper 'ar' program when cross-compiling or when AR is specified.
 	* lib/configure.ac (AR): Determine through AC_CHECK_TOOL.
 	* lib/Makefile.in (AR): Use value determined by 'configure'.
diff --git a/NEWS b/NEWS
index bcac2f3..f3abd0e 100644
--- a/NEWS
+++ b/NEWS
@@ -2,6 +2,9 @@ New in 3.1:
 * The generated C code is now in ANSI-C by default. If you want to support
   pre-ANSI-C compilers, you need to provide the option --language=C on the
   command line or %language=C in the source file.
+* The 'len' parameter of the hash function and of the lookup function is now
+  of type 'size_t' instead of 'unsigned int'. This makes it safe to call these
+  functions with strings of length > 4 GB, on 64-bit machines.
 * Added option --constants-prefix.
 * Added declaration %define constants-prefix.
 
diff --git a/doc/gperf.texi b/doc/gperf.texi
index 9c0fd7d..64f08ba 100644
--- a/doc/gperf.texi
+++ b/doc/gperf.texi
@@ -17,7 +17,7 @@
 
 @c some day we should @include version.texi instead of defining
 @c these values at hand.
-@set UPDATED 11 January 2011
+@set UPDATED 26 November 2016
 @set EDITION 3.1
 @set VERSION 3.1
 @c ---------------------
@@ -40,7 +40,7 @@
 This file documents the features of the GNU Perfect Hash Function
 Generator @value{VERSION}.
 
-Copyright @copyright{} 1989-2011, 2014 Free Software Foundation, Inc.
+Copyright @copyright{} 1989-2016 Free Software Foundation, Inc.
 
 Permission is granted to make and distribute verbatim copies of this
 manual provided the copyright notice and this permission notice are
@@ -78,7 +78,7 @@ Software Foundation instead of in the original English.
 
 @page
 @vskip 0pt plus 1filll
-Copyright @copyright{} 1989-2011 Free Software Foundation, Inc.
+Copyright @copyright{} 1989-2016 Free Software Foundation, Inc.
 
 
 Permission is granted to make and distribute verbatim copies of
@@ -804,7 +804,7 @@ option.  Both functions require two arguments, a string, @code{char *}
 @var{str}, and a length parameter, @code{int} @var{len}.  Their default 
 function prototypes are as follows:
 
-@deftypefun {unsigned int} hash (const char * @var{str}, unsigned int @var{len})
+@deftypefun {unsigned int} hash (const char * @var{str}, size_t @var{len})
 By default, the generated @code{hash} function returns an integer value
 created by adding @var{len} to several user-specified @var{str} byte
 positions indexed into an @dfn{associated values} table stored in a
@@ -815,7 +815,7 @@ into @var{str}) are specified via the @samp{-k} option when running
 @code{gperf}, as detailed in the @emph{Options} section below (@pxref{Options}).
 @end deftypefun
 
-@deftypefun {} in_word_set (const char * @var{str}, unsigned int @var{len})
+@deftypefun {} in_word_set (const char * @var{str}, size_t @var{len})
 If @var{str} is in the keyword set, returns a pointer to that
 keyword.  More exactly, if the option @samp{-t} (or, equivalently, the
 @samp{%struct-type} declaration) was given, it returns
diff --git a/src/output.cc b/src/output.cc
index 1ec5d39..abb9bb5 100644
--- a/src/output.cc
+++ b/src/output.cc
@@ -369,14 +369,14 @@ output_upperlower_strncmp ()
                "(s1, s2, n)\n"
           "     %schar *s1;\n"
           "     %schar *s2;\n"
-          "     %sunsigned int n;\n" :
+          "     %ssize_t n;\n" :
           option[C] ?
                "(s1, s2, n)\n"
           "     %sconst char *s1;\n"
           "     %sconst char *s2;\n"
-          "     %sunsigned int n;\n" :
+          "     %ssize_t n;\n" :
           option[ANSIC] | option[CPLUSPLUS] ?
-               "(%sconst char *s1, %sconst char *s2, %sunsigned int n)\n" :
+               "(%sconst char *s1, %sconst char *s2, %ssize_t n)\n" :
           "",
           register_scs, register_scs, register_scs);
   #if USE_DOWNCASE_TABLE
@@ -430,14 +430,14 @@ output_upperlower_memcmp ()
                "(s1, s2, n)\n"
           "     %schar *s1;\n"
           "     %schar *s2;\n"
-          "     %sunsigned int n;\n" :
+          "     %ssize_t n;\n" :
           option[C] ?
                "(s1, s2, n)\n"
           "     %sconst char *s1;\n"
           "     %sconst char *s2;\n"
-          "     %sunsigned int n;\n" :
+          "     %ssize_t n;\n" :
           option[ANSIC] | option[CPLUSPLUS] ?
-               "(%sconst char *s1, %sconst char *s2, %sunsigned int n)\n" :
+               "(%sconst char *s1, %sconst char *s2, %ssize_t n)\n" :
           "",
           register_scs, register_scs, register_scs);
   #if USE_DOWNCASE_TABLE
@@ -797,7 +797,7 @@ Output::output_asso_values_ref (int pos) const
 /* Generates C code for the hash function that returns the
    proper encoding for each keyword.
    The hash function has the signature
-     unsigned int <hash> (const char *str, unsigned int len).  */
+     unsigned int <hash> (const char *str, size_t len).  */
 
 void
 Output::output_hash_function () const
@@ -832,13 +832,13 @@ Output::output_hash_function () const
   printf (option[KRC] ?
                  "(str, len)\n"
             "     %schar *str;\n"
-            "     %sunsigned int len;\n" :
+            "     %ssize_t len;\n" :
           option[C] ?
                  "(str, len)\n"
             "     %sconst char *str;\n"
-            "     %sunsigned int len;\n" :
+            "     %ssize_t len;\n" :
           option[ANSIC] | option[CPLUSPLUS] ?
-                 "(%sconst char *str, %sunsigned int len)\n" :
+                 "(%sconst char *str, %ssize_t len)\n" :
           "",
           register_scs, register_scs);
 
@@ -1999,13 +1999,13 @@ Output::output_lookup_function () const
   printf (option[KRC] ?
                  "(str, len)\n"
             "     %schar *str;\n"
-            "     %sunsigned int len;\n" :
+            "     %ssize_t len;\n" :
           option[C] ?
                  "(str, len)\n"
             "     %sconst char *str;\n"
-            "     %sunsigned int len;\n" :
+            "     %ssize_t len;\n" :
           option[ANSIC] | option[CPLUSPLUS] ?
-                 "(%sconst char *str, %sunsigned int len)\n" :
+                 "(%sconst char *str, %ssize_t len)\n" :
           "",
           register_scs, register_scs);
 
@@ -2178,9 +2178,9 @@ Output::output ()
     printf ("class %s\n"
             "{\n"
             "private:\n"
-            "  static inline unsigned int %s (const char *str, unsigned int len);\n"
+            "  static inline unsigned int %s (const char *str, size_t len);\n"
             "public:\n"
-            "  static %s%s%s (const char *str, unsigned int len);\n"
+            "  static %s%s%s (const char *str, size_t len);\n"
             "};\n"
             "\n",
             option.get_class_name (), option.get_hash_name (),
diff --git a/tests/c-parse.exp b/tests/c-parse.exp
index e526710..613d820 100644
--- a/tests/c-parse.exp
+++ b/tests/c-parse.exp
@@ -49,7 +49,7 @@ inline
 static unsigned int
 hash (str, len)
      register const char *str;
-     register unsigned int len;
+     register size_t len;
 {
   static unsigned char asso_values[] =
     {
@@ -206,7 +206,7 @@ __attribute__ ((__gnu_inline__))
 struct resword *
 is_reserved_word (str, len)
      register const char *str;
-     register unsigned int len;
+     register size_t len;
 {
   if (len <= MAX_WORD_LENGTH && len >= MIN_WORD_LENGTH)
     {
diff --git a/tests/charsets.exp b/tests/charsets.exp
index 7e2b988..2f008bd 100644
--- a/tests/charsets.exp
+++ b/tests/charsets.exp
@@ -56,7 +56,7 @@ inline
 #endif
 #endif
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static const unsigned short asso_values[] =
     {
@@ -1808,7 +1808,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 const struct charset *
-in_word_set (register const char *str, register unsigned int len)
+in_word_set (register const char *str, register size_t len)
 {
   if (len <= MAX_WORD_LENGTH && len >= MIN_WORD_LENGTH)
     {
diff --git a/tests/chill.exp b/tests/chill.exp
index 8fa7879..18fb2aa 100644
--- a/tests/chill.exp
+++ b/tests/chill.exp
@@ -47,7 +47,7 @@ inline
 static unsigned int
 hash (str, len)
      register const char *str;
-     register unsigned int len;
+     register size_t len;
 {
   static unsigned short asso_values[] =
     {
@@ -185,7 +185,7 @@ __attribute__ ((__gnu_inline__))
 struct resword *
 in_word_set (str, len)
      register const char *str;
-     register unsigned int len;
+     register size_t len;
 {
   enum
     {
diff --git a/tests/cplusplus.exp b/tests/cplusplus.exp
index 69c2fd0..5f28621 100644
--- a/tests/cplusplus.exp
+++ b/tests/cplusplus.exp
@@ -49,7 +49,7 @@ inline
 static unsigned int
 hash (str, len)
      register const char *str;
-     register unsigned int len;
+     register size_t len;
 {
   static unsigned char asso_values[] =
     {
@@ -110,7 +110,7 @@ __attribute__ ((__gnu_inline__))
 struct resword *
 is_reserved_word (str, len)
      register const char *str;
-     register unsigned int len;
+     register size_t len;
 {
   static struct resword wordlist[] =
     {
diff --git a/tests/gpc.exp b/tests/gpc.exp
index 59b5a18..b01f40a 100644
--- a/tests/gpc.exp
+++ b/tests/gpc.exp
@@ -56,7 +56,7 @@ inline
 #endif
 #endif
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static unsigned char asso_values[] =
     {
@@ -97,7 +97,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 struct resword *
-is_reserved_word (register const char *str, register unsigned int len)
+is_reserved_word (register const char *str, register size_t len)
 {
   static struct resword wordlist[] =
     {
diff --git a/tests/incomplete.exp b/tests/incomplete.exp
index dfb0b57..65e8ea5 100644
--- a/tests/incomplete.exp
+++ b/tests/incomplete.exp
@@ -46,7 +46,7 @@ inline
 #endif
 #endif
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static unsigned char asso_values[] =
     {
@@ -87,7 +87,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 struct month *
-in_word_set (register const char *str, register unsigned int len)
+in_word_set (register const char *str, register size_t len)
 {
   static struct month wordlist[] =
     {
diff --git a/tests/java.exp b/tests/java.exp
index 465d6d9..f9e49f2 100644
--- a/tests/java.exp
+++ b/tests/java.exp
@@ -73,7 +73,7 @@ inline
 static unsigned int
 hash (str, len)
      register const char *str;
-     register unsigned int len;
+     register size_t len;
 {
   static unsigned char asso_values[] =
     {
@@ -128,7 +128,7 @@ __attribute__ ((__gnu_inline__))
 struct java_keyword *
 java_keyword (str, len)
      register const char *str;
-     register unsigned int len;
+     register size_t len;
 {
   static struct java_keyword wordlist[] =
     {
diff --git a/tests/languages.exp b/tests/languages.exp
index 051c8e0..8e801d6 100644
--- a/tests/languages.exp
+++ b/tests/languages.exp
@@ -59,7 +59,7 @@ inline
 #endif
 #endif
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static const unsigned short asso_values[] =
     {
@@ -1465,7 +1465,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 const struct language *
-in_word_set (register const char *str, register unsigned int len)
+in_word_set (register const char *str, register size_t len)
 {
   if (len <= MAX_WORD_LENGTH && len >= MIN_WORD_LENGTH)
     {
diff --git a/tests/modula2.exp b/tests/modula2.exp
index 3c6451d..c58bc6e 100644
--- a/tests/modula2.exp
+++ b/tests/modula2.exp
@@ -44,7 +44,7 @@ inline
 #endif
 #endif
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static unsigned char asso_values[] =
     {
@@ -114,7 +114,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 const char *
-in_word_set (register const char *str, register unsigned int len)
+in_word_set (register const char *str, register size_t len)
 {
   static unsigned char lengthtable[] =
     {
diff --git a/tests/objc.exp b/tests/objc.exp
index 78680c5..7bd0612 100644
--- a/tests/objc.exp
+++ b/tests/objc.exp
@@ -47,7 +47,7 @@ inline
 #endif
 #endif
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static unsigned char asso_values[] =
     {
@@ -100,7 +100,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 struct resword *
-is_reserved_word (register const char *str, register unsigned int len)
+is_reserved_word (register const char *str, register size_t len)
 {
   static struct resword wordlist[] =
     {
diff --git a/tests/permut2.exp b/tests/permut2.exp
index 5d29a91..9d8e224 100644
--- a/tests/permut2.exp
+++ b/tests/permut2.exp
@@ -46,7 +46,7 @@ inline
 #endif
 /*ARGSUSED*/
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static unsigned char asso_values[] =
     {
@@ -87,7 +87,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 const char *
-in_word_set (register const char *str, register unsigned int len)
+in_word_set (register const char *str, register size_t len)
 {
   static const char * wordlist[] =
     {
diff --git a/tests/permut3.exp b/tests/permut3.exp
index 2612088..cfae620 100644
--- a/tests/permut3.exp
+++ b/tests/permut3.exp
@@ -46,7 +46,7 @@ inline
 #endif
 /*ARGSUSED*/
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static unsigned char asso_values[] =
     {
@@ -87,7 +87,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 const char *
-in_word_set (register const char *str, register unsigned int len)
+in_word_set (register const char *str, register size_t len)
 {
   static const char * wordlist[] =
     {
diff --git a/tests/permutc2.exp b/tests/permutc2.exp
index 157b049..083e790 100644
--- a/tests/permutc2.exp
+++ b/tests/permutc2.exp
@@ -91,7 +91,7 @@ inline
 #endif
 /*ARGSUSED*/
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static unsigned char asso_values[] =
     {
@@ -132,7 +132,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 const char *
-in_word_set (register const char *str, register unsigned int len)
+in_word_set (register const char *str, register size_t len)
 {
   static const char * wordlist[] =
     {
diff --git a/tests/smtp.gperf b/tests/smtp.gperf
index 5d26f5e..da6ec7d 100644
--- a/tests/smtp.gperf
+++ b/tests/smtp.gperf
@@ -148,7 +148,7 @@ main (argc, argv)
      char *argv[];
 {
   int i, j, k, n, exitcode;
-  unsigned int len;
+  size_t len;
   const struct header_state *hs;
 
   n = 1;
diff --git a/tests/test-4.exp b/tests/test-4.exp
index e860053..c53b0d7 100644
--- a/tests/test-4.exp
+++ b/tests/test-4.exp
@@ -47,7 +47,7 @@ inline
 #endif
 #endif
 static unsigned int
-hash (register const char *str, register unsigned int len)
+hash (register const char *str, register size_t len)
 {
   static unsigned char asso_values[] =
     {
@@ -88,7 +88,7 @@ __attribute__ ((__gnu_inline__))
 #endif
 #endif
 struct resword *
-in_word_set (register const char *str, register unsigned int len)
+in_word_set (register const char *str, register size_t len)
 {
   static struct resword wordlist[] =
     {