Re: msvcrt: Fixed (_w)fopen_s always fails.

[Austin English]

On Tue, Nov 9, 2010 at 9:38 PM, Eryk Wieliczko  wrote:
> (_w)fopen_s were always failing due to regression introduced in
> commit 00f07046c0be658d97c47f69e1b4c518bc06b3a2
>
> dlls/msvcrt/file.c |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/dlls/msvcrt/file.c b/dlls/msvcrt/file.c
> index 329cdeb..91cd448 100644
> --- a/dlls/msvcrt/file.c
> +++ b/dlls/msvcrt/file.c
> @@ -2742,7 +2742,7 @@ int CDECL MSVCRT_fopen_s(MSVCRT_FILE** pFile,
>         const char *filename, const char *mode)
>  {
>     if (!MSVCRT_CHECK_PMT(pFile != NULL) || !MSVCRT_CHECK_PMT(filename != NULL) ||
> - MSVCRT_CHECK_PMT(mode != NULL)) {
> - !MSVCRT_CHECK_PMT(mode != NULL)) {
>
> *MSVCRT__errno() = MSVCRT_EINVAL;
>         return MSVCRT_EINVAL;
>     }
> @@ -2769,7 +2769,7 @@ int CDECL MSVCRT__wfopen_s(MSVCRT_FILE** pFile, const MSVCRT_wchar_t *filename,
>         const MSVCRT_wchar_t *mode)
>  {
>     if (!MSVCRT_CHECK_PMT(pFile != NULL) || !MSVCRT_CHECK_PMT(filename != NULL) ||
> - MSVCRT_CHECK_PMT(mode != NULL)) {
> - !MSVCRT_CHECK_PMT(mode != NULL)) {
>
> *MSVCRT__errno() = MSVCRT_EINVAL;
>         return MSVCRT_EINVAL;
>     }
> --
> 1.7.0.4

Sounds like a good reason to add a unit test to prevent it from
happening again, then.