[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [jdt-core-dev] Created "What is the JDT/Core bugs life cycle?" FAQ entry
|
No, JDT/Core never use the "CLOSED" state as it's definitely not clear what
should be the event to pass this resolution.
Some people think that bugs should be closed at each release delivery,
others (like me) think that this should be the reporter's responsibility to
change the bug to this state (i.e. the ultimate verification).
But whatever would be the event, we tacitly convinced that this final state
was not really necessary: we already have a double verification on the bugs
we have taken into account during a release, a third one would not add too
much quality value to our process.
By a lack of place in the diagram, I was not able to put all events and
dispatch all possible branches to make them as clear as I wanted to. So,
you're absolutely right, the top-left-most branch ends at the first
"VERIFIED" state when the bug is not backported to the "Y" stream. In fact
this diagram was done to show committers what are the important things to
do during the bug life cycle. It's supposed to help them to understand why
we create all the rules. We hope this can also help them to apply these
rules and make our verification process smoother and safer...
I'll surely add some explanations before the diagram of this FAQ input,
hopefully to make it a little bit clearer...
Cordialement/Regards,
Frédéric
Stephan Herrmann
<stephan@xxxxx-be
rlin.de> To
Sent by: "Eclipse JDT Core developers list."
jdt-core-dev-boun <jdt-core-dev@xxxxxxxxxxx>
ces@xxxxxxxxxxx cc
Subject
09/18/2008 01:22 Re: [jdt-core-dev] Created "What is
PM the JDT/Core bugs life cycle?"
FAQ entry
Please respond to
"Eclipse JDT Core
developers list."
<jdt-core-dev@ecl
ipse.org>
Hi,
Two questions from a curious outsider:
+ JDT/Core doesn't use the "CLOSED" state at all? (Why?)
+ The new picture looks like every bug needs to be verified against
two different targets before reaching a final state. Am I misreading the
chart, are some of the outgoing edges optional? (which?)
Take, e.g., the top-left-most "VERIFIED" state:
with target=X and verify X ok the story could be over, couldn't it?
Perhaps some of the VERIFIED states should be marked as (potentially)
final?
cheers,
Stephan
_______________________________________________
jdt-core-dev mailing list
jdt-core-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/jdt-core-dev
