Hi,
Pete Batard wrote:
> Technically we could have a non NULL last_p_stat
This would be clearer if the test would be in the reuse case of
/* Reuse multiextent p_stat if not NULL */
if (!p_stat) {
p_stat = calloc(1, stat_len);
first_extent = true;
} else {
+ /* Ignore Rock Ridge Deep Directory RE entries */
+ if (p_stat->rr.u_su_fields & ISO_ROCK_SUF_RE)
+ goto fail;
first_extent = false;
}
(A trigger for "goto fail" would be a violation of ECMA-119 6.3:
"Each directory shall be recorded as a file in a single Extent, [...]"
But out-of-specs robustness cannot harm here.
)
---------------------------------------------------------------------------
The rest of the changes looks good to me.
I will next make an own ISO image with deep directory relocation and look
how it behaves with the usual programs.
Have a nice day :)
Thomas
