-
Kevin Wolf authored
So far we relied on job->ret and strerror() to produce an error message for failed jobs. Not surprisingly, this tends to result in completely useless messages. This adds a Job.error field that can contain an error string for a failing job, and a parameter to job_completed() that sets the field. As a default, if NULL is passed, we continue to use strerror(job->ret). All existing callers are changed to pass NULL. They can be improved in separate patches. Signed-off-by:
Kevin Wolf <kwolf@redhat.com> Reviewed-by:
Max Reitz <mreitz@redhat.com> Reviewed-by:
Jeff Cody <jcody@redhat.com>
Kevin Wolf authoredSo far we relied on job->ret and strerror() to produce an error message for failed jobs. Not surprisingly, this tends to result in completely useless messages. This adds a Job.error field that can contain an error string for a failing job, and a parameter to job_completed() that sets the field. As a default, if NULL is passed, we continue to use strerror(job->ret). All existing callers are changed to pass NULL. They can be improved in separate patches. Signed-off-by:
Kevin Wolf <kwolf@redhat.com> Reviewed-by:
Max Reitz <mreitz@redhat.com> Reviewed-by:
Jeff Cody <jcody@redhat.com>