6:57 a.m.
- required by rhbz#873488
Signed-off-by: Jakub Filak <jfilak(a)redhat.com>
---
src/daemon/abrt-handle-event.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/src/daemon/abrt-handle-event.c b/src/daemon/abrt-handle-event.c
index 25dcfd7..29fbffd 100644
--- a/src/daemon/abrt-handle-event.c
+++ b/src/daemon/abrt-handle-event.c
@@ -287,13 +287,16 @@ int main(int argc, char **argv)
abrt_init(argv);
const char *program_usage_string = _(
- "& [-v] -e|--event EVENT DIR..."
+ "& [-v -i] -e|--event EVENT DIR..."
);
char *event_name = NULL;
+ bool interactive = false;
+
struct options program_options[] = {
OPT__VERBOSE(&g_verbose),
OPT_STRING('e', "event" , &event_name, "EVENT",
_("Run EVENT on DIR")),
+ OPT_BOOL('i', "interactive" , &interactive,
_("Communicate directly to the user")),
OPT_END()
};
@@ -323,6 +326,8 @@ int main(int argc, char **argv)
dd_close(dd);
struct run_event_state *run_state = new_run_event_state();
+ if (!interactive)
+ make_run_event_state_forwarding(run_state);
if (post_create)
run_state->post_run_callback = is_crash_a_dup;
run_state->logging_callback = do_log;
--
1.7.11.7
6:57 a.m.
New subject: [ABRT PATCH 2/3] check the correct return value of yesforever answer
- required by rhbz#873488
Signed-off-by: Jakub Filak <jfilak(a)redhat.com>
---
src/plugins/abrt-action-perform-ccpp-analysis.in | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/plugins/abrt-action-perform-ccpp-analysis.in
b/src/plugins/abrt-action-perform-ccpp-analysis.in
index 455388d..878cf95 100644
--- a/src/plugins/abrt-action-perform-ccpp-analysis.in
+++ b/src/plugins/abrt-action-perform-ccpp-analysis.in
@@ -30,7 +30,7 @@ def ask_yes_no_save_result(option_name, question):
sys.stdout.write("ASK_YES_NO_YESFOREVER {0} {1}\n"
.format(option_name, question))
sys.stdout.flush()
- return sys.stdin.readline() == "yes\n"
+ return sys.stdin.readline()[:-1] == _("y")
def handle_event(event_name, problem_dir):
"""Helper function handling a single event
--
1.7.11.7
6:57 a.m.
New subject: [ABRT PATCH 3/3] forward stdin, stdout and stderr to event handler
- closes rhbz#873488
Signed-off-by: Jakub Filak <jfilak(a)redhat.com>
---
src/plugins/abrt-action-perform-ccpp-analysis.in | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/plugins/abrt-action-perform-ccpp-analysis.in
b/src/plugins/abrt-action-perform-ccpp-analysis.in
index 878cf95..5266995 100644
--- a/src/plugins/abrt-action-perform-ccpp-analysis.in
+++ b/src/plugins/abrt-action-perform-ccpp-analysis.in
@@ -46,7 +46,7 @@ def handle_event(event_name, problem_dir):
executable = "@libexecdir@/abrt-handle-event"
try:
- proc = Popen([executable, "-e", event_name, "--",
problem_dir])
+ proc = Popen([executable, "-e", event_name, "--",
problem_dir], stdin=sys.stdin, stdout=sys.stdout, stderr=sys.stderr)
return proc.wait() == 0
except OSError as e:
sys.stderr.write("Can't run '{0}':
{1}\n".format(executable, e))
--
1.7.11.7
10:31 a.m.
New subject: [ABRT PATCH 3/3] forward stdin, stdout and stderr to event handler
Is this needed? If I understand the documentation [1] correctly, this is
the default behaviour.
"With the default settings of None, no redirection will occur; the
child’s file handles will be inherited from the parent."
[1] http://docs.python.org/2/library/subprocess.html#subprocess.Popen
On Tue, Nov 06, 2012 at 13:57:02 +0100, Jakub Filak wrote:
- closes rhbz#873488
Signed-off-by: Jakub Filak <jfilak(a)redhat.com>
---
src/plugins/abrt-action-perform-ccpp-analysis.in | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/plugins/abrt-action-perform-ccpp-analysis.in
b/src/plugins/abrt-action-perform-ccpp-analysis.in
index 878cf95..5266995 100644
--- a/src/plugins/abrt-action-perform-ccpp-analysis.in
+++ b/src/plugins/abrt-action-perform-ccpp-analysis.in
@@ -46,7 +46,7 @@ def handle_event(event_name, problem_dir):
executable = "@libexecdir@/abrt-handle-event"
try:
- proc = Popen([executable, "-e", event_name, "--",
problem_dir])
+ proc = Popen([executable, "-e", event_name, "--",
problem_dir], stdin=sys.stdin, stdout=sys.stdout, stderr=sys.stderr)
return proc.wait() == 0
except OSError as e:
sys.stderr.write("Can't run '{0}':
{1}\n".format(executable, e))
--
1.7.11.7
10:40 a.m.
New subject: [ABRT PATCH 3/3] forward stdin, stdout and stderr to event handler
Thanks! In the light of evidence you wrote in the last email, the last patch
is not needed. It was my first attempt to fix the hrbz#873488 and I forgot to
remove it.
I hope the rest is OK.
On Tuesday 06 of November 2012 17:31:33 Martin Milata wrote:
Is this needed? If I understand the documentation [1] correctly, this
is
the default behaviour.
"With the default settings of None, no redirection will occur; the
child’s file handles will be inherited from the parent."
[1] http://docs.python.org/2/library/subprocess.html#subprocess.Popen
On Tue, Nov 06, 2012 at 13:57:02 +0100, Jakub Filak wrote:
> - closes rhbz#873488
>
> Signed-off-by: Jakub Filak <jfilak(a)redhat.com>
> ---
>
> src/plugins/abrt-action-perform-ccpp-analysis.in | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/plugins/abrt-action-perform-ccpp-analysis.in
> b/src/plugins/abrt-action-perform-ccpp-analysis.in index 878cf95..5266995
> 100644
> --- a/src/plugins/abrt-action-perform-ccpp-analysis.in
> +++ b/src/plugins/abrt-action-perform-ccpp-analysis.in
>
> @@ -46,7 +46,7 @@ def handle_event(event_name, problem_dir):
> executable = "@libexecdir@/abrt-handle-event"
>
> try:
> - proc = Popen([executable, "-e", event_name, "--",
problem_dir])
> + proc = Popen([executable, "-e", event_name, "--",
problem_dir],
> stdin=sys.stdin, stdout=sys.stdout, stderr=sys.stderr)>
> return proc.wait() == 0
>
> except OSError as e:
> sys.stderr.write("Can't run '{0}':
{1}\n".format(executable, e))
3:38 a.m.
On 11/06/2012 01:57 PM, Jakub Filak wrote:
struct options program_options[] = {
OPT__VERBOSE(&g_verbose),
OPT_STRING('e', "event" , &event_name, "EVENT",
_("Run EVENT on DIR")),
+ OPT_BOOL('i', "interactive" , &interactive,
_("Communicate directly to the user")),
I'm afraid this help text will be not very clear to the reader :(
Maybe it's better to invert the option? Something like:
-m --machine Use machine-oriented message format
Human users won't even need to give any options
to get expected behavior, and help text is more clear.
--
vda
3:45 a.m.
On 11/07/2012 10:38 AM, Denys Vlasenko wrote:
On 11/06/2012 01:57 PM, Jakub Filak wrote:
> struct options program_options[] = {
> OPT__VERBOSE(&g_verbose),
> OPT_STRING('e', "event" , &event_name,
"EVENT", _("Run EVENT on DIR")),
> + OPT_BOOL('i', "interactive" , &interactive,
_("Communicate directly to the user")),
I'm afraid this help text will be not very clear to the reader :(
Maybe it's better to invert the option? Something like:
-m --machine Use machine-oriented message format
Human users won't even need to give any options
to get expected behavior, and help text is more clear.
- I would agree if it was an executable meant for humans, but in this
case it's a helper which is used from other programs in 99.9% of cases
and we would have to remember to add this option to every place where we
use it. So I'm for defaulting to machine-oriented format without option
and human readable only with option.
4249
days inactive
4250
days old
crash-catcher@lists.fedorahosted.org
6 comments
4 participants
participants (4)
-
Denys Vlasenko -
Jakub Filak -
Jiri Moskovcak -
Martin Milata