Re: [PATCH 01/12] fsmonitor: refactor filesystem checks to common interface

["Ævar Arnfjörð Bjarmason" <avarab@gmail.com>]

On Sun, Oct 09 2022, Eric DeCosta via GitGitGadget wrote:

> From: Eric DeCosta <edecosta@mathworks.com>
> [...]
> +int fsmonitor__get_fs_info(const char *path, struct fs_info *fs_info)
> +{
> +	struct statfs fs;
> +	if (statfs(path, &fs) == -1) {
> +		int saved_errno = errno;
> +		trace_printf_key(&trace_fsmonitor, "statfs('%s') failed: %s",
> +				 path, strerror(saved_errno));
> +		errno = saved_errno;
> +		return -1;
> +	}
> +
> +	trace_printf_key(&trace_fsmonitor,
> +			 "statfs('%s') [type 0x%08x][flags 0x%08x] '%s'",
> +			 path, fs.f_type, fs.f_flags, fs.f_fstypename);
> +
> +	if (!(fs.f_flags & MNT_LOCAL))
> +		fs_info->is_remote = 1;
> +	else
> +		fs_info->is_remote = 0;

Odd to go through the trouble of inverting the flag check like this just
to assign to it manually. I'd think:

	if (flags & LOCAL)
		is_remote = 0;
	else
		is_remote = 1;

Would make more sense, or actually if you do invert it:

	fs_info->is_remote = !(fs.f_flags & MNT_LOCAL);

No?

> +	fs_info->typename = xstrdup(fs.f_fstypename);

Instead of xstrdup()-ing this...
> +
> +	trace_printf_key(&trace_fsmonitor,
> +				"'%s' is_remote: %d",
> +				path, fs_info->is_remote);
> +	return 0;
> +}
> +
> +int fsmonitor__is_fs_remote(const char *path)
> +{
> +	struct fs_info fs;
> +	if (fsmonitor__get_fs_info(path, &fs))
> +		return -1;
> +
> +	free(fs.typename);

...just to have some callers free() it again, maybe it makes sense to
say that return a const char *, and if you need it past the call to
fsmonitor__get_fs_info() you should do the xstrdup() yourself.

Skimming the end-state the OSX one gives you a copy of your own, as it's
stuck into the struct you provided.

For Linux we don't get a copy, but it's filled per getmntent() call.

Anyway, small potatoes...

> +	if (h == INVALID_HANDLE_VALUE) {
> +		error(_("[GLE %ld] unable to open for read '%ls'"),
> +		      GetLastError(), wpath);
> +		return -1;
> +	}

Do away with the braces and just "return error(..."

> +
> +	if (!GetFileInformationByHandleEx(h, FileRemoteProtocolInfo,
> +		&proto_info, sizeof(proto_info))) {
> +		error(_("[GLE %ld] unable to get protocol information for '%ls'"),
> +		      GetLastError(), wpath);
> +		CloseHandle(h);
> +		return -1;
> +	}
> +
> +	CloseHandle(h);

And this duplication you can avoid with an earlier:

	int ret = 0;


Then
		ret = error(...)
		goto cleanup;

and then at the end:

cleanup:
	CloseHandle(h)
	return ret;

> +	/*
> +	 * Do everything in wide chars because the drive letter might be
> +	 * a multi-byte sequence.  See win32_has_dos_drive_prefix().
> +	 */
> +	if (xutftowcs_path(wpath, path) < 0) {
> +		return -1;
> +	}

Don't need the braces.
> +
> +	/*
> +	 * GetDriveTypeW() requires a final slash.  We assume that the
> +	 * worktree pathname points to an actual directory.
> +	 */
> +	wlen = wcslen(wpath);
> +	if (wpath[wlen - 1] != L'\\' && wpath[wlen - 1] != L'/') {
> +		wpath[wlen++] = L'\\';
> +		wpath[wlen] = 0;
> +	}
> +
> +	/*
> +	 * Normalize the path.  If nothing else, this converts forward
> +	 * slashes to backslashes.  This is essential to get GetDriveTypeW()
> +	 * correctly handle some UNC "\\server\share\..." paths.
> +	 */
> +	if (!GetFullPathNameW(wpath, MAX_PATH, wfullpath, NULL)) {
> +		return -1;
> +	}
> +
> +	driveType = GetDriveTypeW(wfullpath);

Can't you just do away with this driveType variable and assign directly
to fs_info->is_remote if it's == DRIVE_REMOTE?...

> +	trace_printf_key(&trace_fsmonitor,
> +			 "DriveType '%s' L'%ls' (%u)",
> +			 path, wfullpath, driveType);

...well, not due to this, but do we really need to log the raw value &
is that meaningful? A user of the trace2 logs would need to reverse map
that, doesn't the API provide some "what is it then?" function?

> +
> +	if (driveType == DRIVE_REMOTE) {
> +		fs_info->is_remote = 1;

Here we do that assignment...

> +		if (check_remote_protocol(wfullpath) < 0)
> +			return -1;
> +	} else {
> +		fs_info->is_remote = 0;
> +	}
> +
> +	trace_printf_key(&trace_fsmonitor,
> +				"'%s' is_remote: %d",
> +				path, fs_info->is_remote);

ditto just "is remote?"