On 11/7/23 19:17, Timur Tabi wrote:
On Tue, 2023-11-07 at 19:11 +0100, Danilo Krummrich wrote:
I think the problem already starts with gf100_bar_new_() not setting its
pbar argument to NULL on failure, but this code assuming that.
Isn't it standard procedure for functions not to write to output parameters on failure?
Didn't want to say gf100_bar_new_() is wrong, but rather this function is
wrong obviously taking the wrong assumption.
Generally, I think it would be better if all those functions would return
an ERR_PTR on failure.
Agreed.
