Re: gpbs, syslog() and ebuf variable

[Fred Kiefer]

Hi Ivan,

last night I cleaned up this code by copying over the version from gdnc, but I did not get around to commit it or even write you. I don't know how we did get the current useless code and wanted to find out about this before commiting the clean up. I should have mailed.Sorry for that. The way forward will now bw to revert your change and commit mine. The reason why the code is needed is to have some debug information when gpbs fails completely.

Fred

On the road

Am 06.05.2014 um 04:05 schrieb Ivan Vučica <address@hidden>:

I have removed this code in r37842.

If anyone needs this code or sees a reason why it should be/should have been there, please revert and let me know.

On Mon, May 5, 2014 at 3:18 AM, Ivan Vučica <address@hidden> wrote:

Cheers,

GCC is (rightly) dying as follows:

gpbs.m: In function ‘gpbs_log’:

gpbs.m:67:7: error: format not a string literal and no format arguments [-Werror=format-security]

       syslog (log_priority | prio, ebuf);

ebuf seems to be a static char[2048], but I can't see where it's being set in gpbs.m. It looks like it's only being accessed.

Even gpbs_log() seems to be called just once during init().

If I'm reading the code correctly, it should be safe to get rid of both gpbs_log() and of ebuf -- hence even of syslog(). I'd like to do that.

Anything I'm missing? Any further insight?

--

Ivan Vučica

address@hidden

--

Ivan Vučica

address@hidden

_______________________________________________
Gnustep-dev mailing list
address@hidden
https://lists.gnu.org/mailman/listinfo/gnustep-dev