"Ryan Bloom" <[EMAIL PROTECTED]> writes:
> The docs may say that, but the code doesn't do anything about "." and
> "..". I just double-checked unix/dir.c. :-( I actually remember
> looking at this years ago when I wrote the original code, and thinking
> through the "." and ".." problem, which is why I said that APR will
> return those two values first, but I don't think I ever actually coded
> it, and I'm not sure how it would be done.
Heh! Pardon me while I feel smug :-)...
I also felt a bit suspicious about that promise, and therefore wrote
this code in Subversion a while back:
for (err = svn_io_dir_read (&finfo, flags, dir, pool);
err == SVN_NO_ERROR;
svn_pool_clear (subpool),
err = svn_io_dir_read (&finfo, flags, dir, pool))
{
const char *this_path, *this_edit_path;
if (finfo.filetype == APR_DIR)
{
/* Skip entries for this dir and its parent.
(APR promises that they'll come first, so technically
this guard could be moved outside the loop. But somehow
that feels iffy. */
if (finfo.name[0] == '.'
&& (finfo.name[1] == '\0'
|| (finfo.name[1] == '.' && finfo.name[2] == '\0')))
continue;
[...]
}
[...]
}
Anyway, will do something similar in apr_dir_check_empty().
-K
