Barath Aron wrote:
> What system implements a realloc() which does not work
> if ptr is null pointer?
None, according to our findings [1].
> I looked into the fchdir.c ...
new_dirs =
(dirs != NULL
? (dir_info_t *) realloc (dirs, new_allocated * sizeof *dirs)
: (dir_info_t *) malloc (new_allocated * sizeof *dirs));
> To me, it looks the ternary is unnecessary there, and only a single
> realloc() is fine.
This is correct.
However, I prefer to keep the code as it is. Given that everyone knows
what malloc() does, whereas
- For realloc, there are three cases,
- You have to look up the documentation to know exactly how realloc()
behaves in each of the cases,
the code is simpler to understand as is.
> Or this is really a bug?
It's not a bug. It's a question of style.
Bruno
[1] https://www.gnu.org/software/gnulib/manual/html_node/realloc.html
