Jiri Moskovcak <jmoskovc(a)redhat.com> writes:
On 02/20/2012 09:57 AM, Nikola Pajkovsky wrote:
> Signed-off-by: Nikola Pajkovsky<npajkovs(a)redhat.com>
> ---
> src/plugins/rhbz.c | 6 +++++-
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
> diff --git a/src/plugins/rhbz.c b/src/plugins/rhbz.c
> index 991b1cb..32b61dc 100644
> --- a/src/plugins/rhbz.c
> +++ b/src/plugins/rhbz.c
> @@ -420,7 +420,11 @@ int rhbz_new_bug(struct abrt_xmlrpc *ax, problem_data_t
*problem_data,
> }
> else
> {
> - strbuf_append_strf(buf_summary, "[abrt] %s", package);
> + if (analyzer&& !strcmp(analyzer, "Kerneloops"))
> + strbuf_append_str(buf_summary, "[abrt]");
> + else
> + strbuf_append_strf(buf_summary, "[abrt] %s", package);
> +
> if (function&& strlen(function)< 30)
> strbuf_append_strf(buf_summary, ": %s", function);
>
- is it necessary to hardcode this for kerneloops? Can't we just drop
it for everything? I don't remember if there was a special need for
this...
I was thinking to remove package name from summary for every crash. Then
I realize it doesn't make sense.
w/ pkg name
[abrt] libfoo: Process /usr/bin/lpstat.cups was killed by signal 11
(SIGSEGV)
w/o
[abrt] Process /usr/bin/lpstat.cups was killed by signal 11 (SIGSEGV)
or
[abrt] Preferences.py:827:_on_extension_changed:TypeError: 'NoneType'
object does not support item assignment
I'd rather see, that normal pkgs have pkg-name in summary. If anyone
have any idea how to reorganize or rephrase summary I'm all ears.
--
Nikola