This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
PATCH: fix portability for ext/stdio_filebuf/char/10063-2.cc
- From: Loren James Rittle <rittle at latour dot rsch dot comm dot mot dot com>
- To: libstdc++ at gcc dot gnu dot org
- Cc: pcarlini at suse dot de
- Date: Mon, 19 Jan 2004 19:02:51 -0600 (CST)
- Subject: PATCH: fix portability for ext/stdio_filebuf/char/10063-2.cc
- Reply-to: rittle at labs dot mot dot com
I wrote last week:
> FAIL: ext/stdio_filebuf/char/10063-2.cc execution test
[Explanation of assumption violation found, removed...]
Looking over the history of this extension, I think the test case just
needs to be slightly tweaked. Paolo, as the new expert in this area, OK?
* testsuite/ext/stdio_filebuf/char/10063-2.cc: Use
__gnu_cxx::stdio_sync_filebuf<char>.
Index: libstdc++-v3/testsuite/ext/stdio_filebuf/char/10063-2.cc
===================================================================
RCS file: /cvs/gcc/gcc/libstdc++-v3/testsuite/ext/stdio_filebuf/char/10063-2.cc,v
retrieving revision 1.2
diff -c -r1.2 10063-2.cc
*** libstdc++-v3/testsuite/ext/stdio_filebuf/char/10063-2.cc 29 Dec 2003 19:26:16 -0000 1.2
--- libstdc++-v3/testsuite/ext/stdio_filebuf/char/10063-2.cc 20 Jan 2004 00:57:41 -0000
***************
*** 22,28 ****
#include <cstdio>
#include <fstream>
! #include <ext/stdio_filebuf.h>
#include <testsuite_hooks.h>
void test2()
--- 22,28 ----
#include <cstdio>
#include <fstream>
! #include <ext/stdio_sync_filebuf.h>
#include <testsuite_hooks.h>
void test2()
***************
*** 41,47 ****
c = getc(file);
VERIFY(c == '1');
{
! __gnu_cxx::stdio_filebuf<char> sbuf(file, ios_base::in);
c = sbuf.sbumpc();
VERIFY(c == '2');
c = sbuf.sbumpc();
--- 41,47 ----
c = getc(file);
VERIFY(c == '1');
{
! __gnu_cxx::stdio_sync_filebuf<char> sbuf(file);
c = sbuf.sbumpc();
VERIFY(c == '2');
c = sbuf.sbumpc();