D.J. Heap wrote:
> Something like this? It passes ra_local tests on Win32 and FC5.
Yes! Thank you.
(The next step would be to fix apr_file_attrs_set to do something
similar; _that_ implementation is my fault, too :( But this should work
correctly.)
[...]
> @@ -321,8 +322,10 @@
> *
> * @a path is the utf8-encoded path to the file. If @a enable_write
> * is @c TRUE, then make the file read-write. If @c FALSE, make it
> - * write-only. If @a ignore_enoent is @c TRUE, don't fail if the target
> + * read-only. If @a ignore_enoent is @c TRUE, don't fail if the target
> * file doesn't exist.
> + *
> + * @deprecated Provided for backward compatibility with the 1.1 API.
>
"for backward compatibility with the 1.4 API" (or 1.3, if this gets
back-ported to the 1.4 branch before the release).
[...]
> +/* This is a helper function for the svn_io_set_file_read* functions
> + that attempts to honor the users umask when dealing with
> + permission changes. */
> +static svn_error_t *
> +io_set_file_perms(const char *path,
> + svn_boolean_t change_readwrite,
> + svn_boolean_t enable_write,
> + svn_boolean_t change_executable,
> + svn_boolean_t executable,
> + svn_boolean_t ignore_enoent,
> + apr_pool_t *pool)
>
You could wrap this whole function into #ifndef WIN32, since it's not
going to be used on that platform.
[...]
-- Brane
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sun Jun 4 10:40:58 2006