Re: [PATCH v2] win32: explicitly set errno to enoent if err is not error_directory

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Am 19.12.22 um 17:20 schrieb Rose via GitGitGadget:
> From: Seija Kijin <doremylover123@xxxxxxxxx>
> 
> At this point, the only two possible errors are
> ERROR_DIRECTORY or ERROR_BAD_PATHNAME.
> 
> This code clarifies this and also saves a call to
> err_win_to_posix.
> 
> Signed-off-by: Seija Kijin <doremylover123@xxxxxxxxx>
> ---

> diff --git a/compat/win32/dirent.c b/compat/win32/dirent.c
> index 52420ec7d4d..8f94a5ab6db 100644
> --- a/compat/win32/dirent.c
> +++ b/compat/win32/dirent.c
> @@ -34,13 +34,13 @@ DIR *opendir(const char *name)
>  	if (len && !is_dir_sep(pattern[len - 1]))
>  		pattern[len++] = '/';
>  	pattern[len++] = '*';
> -	pattern[len] = 0;
> +	pattern[len] = '\0';
>  
>  	/* open find handle */
>  	h = FindFirstFileW(pattern, &fdata);
>  	if (h == INVALID_HANDLE_VALUE) {
>  		DWORD err = GetLastError();
> -		errno = (err == ERROR_DIRECTORY) ? ENOTDIR : err_win_to_posix(err);
> +		errno = (err == ERROR_DIRECTORY) ? ENOTDIR : ENOENT;

This change does not look correct. The documentation of
FindFirstFileW()[*] does not give any guarantees that ERROR_DIRECTORY
and ERROR_BAD_PATHNAME are the only possible error values. It is
certainly not unthinkable that a hardware error is reported as
ERROR_IO_DEVICE and should be translated to EIO.

[*]
https://learn.microsoft.com/en-us/windows/win32/api/fileapi/nf-fileapi-findfirstfilew

-- Hannes




[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux