Hi Chris, David and Thomas,
I took a closer look now, too. Funny that the original change was contributed
by my colleagues because of coverity and that they didn't do it completely
right. 😉 As a code comment in our attachListener.hpp suggests, the '0'
termination to please coverity was added fa
On 6/03/2018 6:50 PM, Langer, Christoph wrote:
Hi Chris, David and Thomas,
I took a closer look now, too. Funny that the original change was contributed
by my colleagues because of coverity and that they didn't do it completely
right. 😉 As a code comment in our attachListener.hpp suggests, the
Thanks, David.
A colleague just told me that a guarantee would also quiesce Coverity. So that
could really be an option then. Let's wait for Chris' opinion...
> -Original Message-
> From: David Holmes [mailto:david.hol...@oracle.com]
> Sent: Dienstag, 6. März 2018 13:26
> To: Langer, Chr
On 3/6/18 4:26 AM, David Holmes wrote:
On 6/03/2018 6:50 PM, Langer, Christoph wrote:
Hi Chris, David and Thomas,
I took a closer look now, too. Funny that the original change was
contributed by my colleagues because of coverity and that they didn't
do it completely right. 😉 As a code comment
Hi Chris,
> > I don't know why strncpy would do zero padding?
> From the man page:
>
> The stpncpy() and strncpy() functions copy at most len characters
> from src into dst. If src is less than len
> characters long, the remainder of dst is filled with `\0'
> characters. Otherwise,
Maybe stupid question, any reason we could not just strdup() those strings?
And add an ~AttachOperation dtor to clean them up again?
Otherwise, I usually prefer using snprintf (or, jio_snprintf or the new
os::snprintf()) with "%s" as format string:
jio_snprintf(_name, sizeof(_name), "%s", inputst
