Change the length of strings created from fixed-size buffers.

Use the usual C rule for NUL-terminated strings instead of discarding all the
trailing NUL characters.

This was unexpected (as buffers in C code are not necessarily always padded
with NULs to their full length) and also inconsistent among languages as this
was only done for those of them using typemaps/strings.swg but not for C# or
Java, for example, which terminated the string at the first NUL even before
this change.

Notice that this patch couldn't use strlen() or wcslen() with possibly not
NUL-terminated strings, so we had to add [our own equivalents of] strnlen()
and wcsnlen() and use them instead. This required adding yet another parameter
to string typemap macros, so update the example using them accordingly too.

CHANGES.current

@@ -5,6 +5,14 @@ See the RELEASENOTES file for a summary of changes in each release.
 Version 3.0.0 (in progress)
 ============================
 
+2013-12-23: vadz
+            [Octave, Perl, Python, R, Ruby, Tcl] Change the length of strings created from fixed-size char
+            buffers in C code.
+
+            This is a potential backwards compatibility break: a "char buf[5]" containing "ho\0la" was
+            returned as a string of length 5 before, but is returned as a string of length 2 now. Apply
+            "char FIXSIZE[ANY]" typemaps to explicitly choose the old behaviour.
+
 2013-12-23: talby
             [Perl] Add support for directors.
 

Examples/perl5/xmlstring/xmlstring.i

@@ -93,6 +93,17 @@ SWIG_FromXMLChPtrAndSize(const XMLCh* input, size_t size)
 }
 }
 
+%fragment("SWIG_XMLStringNLen","header") {
+size_t
+SWIG_XMLStringNLen(const XMLCh* s, size_t maxlen)
+{
+  const XMLCh *p;
+  for (p = s; maxlen-- && *p; p++)
+    ;
+  return p - s;
+}
+}
+
 %init {
   if (!SWIG_UTF8Transcoder()) {
     croak("ERROR: XML::Xerces: INIT: Could not create UTF-8 transcoder");
@@ -106,6 +117,7 @@ SWIG_FromXMLChPtrAndSize(const XMLCh* input, size_t size)
 		 SWIG_AsXMLChPtrAndSize, 
 		 SWIG_FromXMLChPtrAndSize,
 		 XERCES_CPP_NAMESPACE::XMLString::stringLen,
+		 SWIG_XMLStringNLen,
 		 "<XMLCh.h>", INT_MIN, INT_MAX);
 
 

Examples/test-suite/perl5/primitive_types_runme.pl

@@ -166,7 +166,7 @@ $t->{var_paramd} = $primitive_types::sct_paramd;
 $t->{var_paramc} = $primitive_types::sct_paramc;
 ok($t->v_check(), 'v_check');
 
-is($primitive_types::def_namet, "ho\0la", "namet");
+is($primitive_types::def_namet, "hola", "namet");
 $t->{var_namet} = $primitive_types::def_namet;
 is($t->{var_namet}, $primitive_types::def_namet, "namet");
 

Examples/test-suite/primitive_types.i

@@ -185,7 +185,7 @@
   char* const def_pchar = (char *const)"hello";
   const char* const def_pcharc = "hija";
 
-  const namet def_namet = {'h','o',0, 'l','a'};
+  const namet def_namet = {'h','o','l','a', 0};
 
   extern namet gbl_namet;
 

Examples/test-suite/python/primitive_types_runme.py

@@ -165,7 +165,7 @@ t.var_paramc = sct_paramc
 t.v_check()
 
 # this value contains a '0' char!
-if def_namet != 'ho\0la':
+if def_namet != 'hola':
   print "bad namet", def_namet
   raise RuntimeError
 

Lib/typemaps/string.swg

@@ -17,8 +17,19 @@ SWIG_pchar_descriptor(void)
 }
 }
 
+%fragment("SWIG_strnlen","header",fragment="SWIG_FromCharPtrAndSize") {
+size_t
+SWIG_strnlen(const char* s, size_t maxlen)
+{
+  const char *p;
+  for (p = s; maxlen-- && *p; p++)
+    ;
+  return p - s;
+}
+}
 
 %include <typemaps/strings.swg>
 %typemaps_string(%checkcode(STRING), %checkcode(CHAR),
-		 char, Char, SWIG_AsCharPtrAndSize, SWIG_FromCharPtrAndSize, strlen,
+		 char, Char, SWIG_AsCharPtrAndSize, SWIG_FromCharPtrAndSize,
+		 strlen, SWIG_strnlen,
 		"<limits.h>", CHAR_MIN, CHAR_MAX)

Lib/typemaps/strings.swg

@@ -22,6 +22,7 @@
 			 SWIG_AsCharPtrAndSize,
 			 SWIG_FromCharPtrAndSize,
 			 SWIG_CharPtrLen,
+			 SWIG_CharBufLen,
 			 SWIG_AsCharPtr,
 			 SWIG_FromCharPtr,
 			 SWIG_AsCharArray,
@@ -276,12 +277,13 @@
 }
 %typemap(freearg) const Char (&)[ANY] "";
 
-%typemap(out,fragment=#SWIG_FromCharPtrAndSize)
+%typemap(out,fragment=#SWIG_FromCharPtrAndSize,fragment=#SWIG_CharBufLen)
   Char [ANY], const Char[ANY] 
 {
-  size_t size = $1_dim0;
 %#ifndef SWIG_PRESERVE_CARRAY_SIZE
-  while (size && ($1[size - 1] == '\0')) --size;
+  size_t size = SWIG_CharBufLen($1, $1_dim0);
+%#else
+  size_t size = $1_dim0;
 %#endif
    %set_output(SWIG_FromCharPtrAndSize($1, size));
 }
@@ -298,23 +300,25 @@
 
 /* varout */
 
-%typemap(varout,noblock=1,fragment=#SWIG_FromCharPtrAndSize)
+%typemap(varout,noblock=1,fragment=#SWIG_CharBufLen)
   Char [ANY], const Char [ANY] {
-  size_t size = $1_dim0;
 %#ifndef SWIG_PRESERVE_CARRAY_SIZE
-  while (size && ($1[size - 1] == '\0')) --size;
+  size_t size = SWIG_CharBufLen($1, $1_dim0);
+%#else
+  size_t size = $1_dim0;
 %#endif
   %set_varoutput(SWIG_FromCharPtrAndSize($1, size));
 }
 
 /* constant */
 
-%typemap(constcode,fragment=#SWIG_FromCharPtrAndSize)
+%typemap(constcode,fragment=#SWIG_CharBufLen)
   Char [ANY], const Char [ANY]
 {
-  size_t size = $value_dim0;
 %#ifndef SWIG_PRESERVE_CARRAY_SIZE
-  while (size && ($value[size - 1] == '\0')) --size;
+  size_t size = SWIG_CharBufLen($1, $1_dim0);
+%#else
+  size_t size = $value_dim0;
 %#endif
   %set_constant("$symname", SWIG_FromCharPtrAndSize($value,size));
 }
@@ -323,12 +327,13 @@
 #if defined(SWIG_DIRECTOR_TYPEMAPS)
 
 /* directorin */
-%typemap(directorin,fragment=#SWIG_FromCharPtrAndSize)
+%typemap(directorin,fragment=#SWIG_CharBufLen)
   Char [ANY], const Char [ANY] 
 {
-  size_t size = $1_dim0;
 %#ifndef SWIG_PRESERVE_CARRAY_SIZE
-  while (size && ($1[size - 1] == '\0')) --size;
+  size_t size = SWIG_CharBufLen($1, $1_dim0);
+%#else
+  size_t size = $1_dim0;
 %#endif
   $input = SWIG_FromCharPtrAndSize($1, size);
 }
@@ -360,12 +365,13 @@
 
 /* throws */
 
-%typemap(throws,fragment=#SWIG_FromCharPtrAndSize)
+%typemap(throws,fragment=#SWIG_CharBufLen)
   Char [ANY], const Char[ANY]
 {
-  size_t size = $1_dim0;
 %#ifndef SWIG_PRESERVE_CARRAY_SIZE
-  while (size && ($1[size - 1] == '\0')) --size;
+  size_t size = SWIG_CharBufLen($1, $1_dim0);
+%#else
+  size_t size = $1_dim0;
 %#endif
   %raise(SWIG_FromCharPtrAndSize($1, size), "$type", 0); 
 }
@@ -499,6 +505,7 @@
 			 SWIG_AsCharPtrAndSize,
 			 SWIG_FromCharPtrAndSize,
 			 SWIG_CharPtrLen,
+			 SWIG_CharBufLen,
 			 SWIG_NewCopyCharArray,
 			 SWIG_DeleteCharArray,
 			 FragLimits, CHAR_MIN, CHAR_MAX)
@@ -586,6 +593,7 @@ SWIG_AsVal_dec(Char)(SWIG_Object obj, Char *val)
 		 SWIG_AsCharPtrAndSize,
 		 SWIG_FromCharPtrAndSize,
 		 SWIG_CharPtrLen,
+		 SWIG_CharBufLen,
 		 SWIG_As##CharName##Ptr,
 		 SWIG_From##CharName##Ptr,
 		 SWIG_As##CharName##Array,
@@ -604,12 +612,14 @@ SWIG_AsVal_dec(Char)(SWIG_Object obj, Char *val)
 			 SWIG_AsCharPtrAndSize,
 			 SWIG_FromCharPtrAndSize,
 			 SWIG_CharPtrLen,
+			 SWIG_CharBufLen,
 			 FragLimits, CHAR_MIN, CHAR_MAX)
 %_typemap2_string(StringCode, CharCode,
 		  Char, CharName,
 		  SWIG_AsCharPtrAndSize,
 		  SWIG_FromCharPtrAndSize,
 		  SWIG_CharPtrLen,
+		  SWIG_CharBufLen,
 		  %new_copy_array,
 		  %delete_array,
 		  FragLimits, CHAR_MIN, CHAR_MAX)
@@ -624,6 +634,7 @@ SWIG_AsVal_dec(Char)(SWIG_Object obj, Char *val)
 			       SWIG_AsCharPtrAndSize,
 			       SWIG_FromCharPtrAndSize,
 			       SWIG_CharPtrLen,
+			       SWIG_CharBufLen,
 			       SWIG_NewCopyCharArray,
 			       SWIG_DeleteCharArray,
 			       FragLimits, CHAR_MIN, CHAR_MAX)
@@ -632,6 +643,7 @@ SWIG_AsVal_dec(Char)(SWIG_Object obj, Char *val)
 		  SWIG_AsCharPtrAndSize,
 		  SWIG_FromCharPtrAndSize,
 		  SWIG_CharPtrLen,
+		  SWIG_CharBufLen,
 		  SWIG_NewCopyCharArray,
 		  SWIG_DeleteCharArray,
 		  FragLimits, CHAR_MIN, CHAR_MAX)

Lib/typemaps/wstring.swg

@@ -18,8 +18,20 @@ SWIG_pwchar_descriptor()
 }
 }
 
+%fragment("SWIG_wcsnlen","header",fragment="SWIG_FromWCharPtrAndSize") {
+size_t
+SWIG_wcsnlen(const wchar_t* s, size_t maxlen)
+{
+  const wchar_t *p;
+  for (p = s; maxlen-- && *p; p++)
+    ;
+  return p - s;
+}
+}
+
 %include <typemaps/strings.swg>
 %typemaps_string(%checkcode(UNISTRING), %checkcode(UNICHAR),
-		 wchar_t, WChar, SWIG_AsWCharPtrAndSize, SWIG_FromWCharPtrAndSize, wcslen,
+		 wchar_t, WChar, SWIG_AsWCharPtrAndSize, SWIG_FromWCharPtrAndSize,
+		 wcslen, SWIG_wcsnlen,
 		"<wchar.h>", WCHAR_MIN, WCHAR_MAX)