[vtk-developers] Accidental gerrit merge

[Marcus D. Hanwell]

Agreed, but at that point it is also approved for merging. If you
don't want it to be merged don't give it a +2, a +1 is a good way to
say I think it is good but really want someone else to view it. The
submit change button also becomes available on the main topic page
after a +2, which will not ask for confirmation either.

On Tue, Sep 11, 2012 at 4:53 PM, David Cole <david.cole at kitware.com> wrote:
> Well, *anybody* with approval rights
>
> On Tue, Sep 11, 2012 at 4:53 PM, David Cole <david.cole at kitware.com> wrote:
>> The button is available if *anybody* gives it a +2
>>
>>
>> On Tue, Sep 11, 2012 at 4:13 PM, Bill Lorensen <bill.lorensen at gmail.com> wrote:
>>> I don't recall doiing a +2, but I'll be careful in the future...
>>>
>>>
>>>
>>> On Tue, Sep 11, 2012 at 3:47 PM, Marcus D. Hanwell
>>> <marcus.hanwell at kitware.com> wrote:
>>>>
>>>> On Tue, Sep 11, 2012 at 2:57 PM, Bill Lorensen <bill.lorensen at gmail.com>
>>>> wrote:
>>>> > Folks,
>>>> >
>>>> > I recently accidentally submitted a gerrit patch. This was in spite of
>>>> > the
>>>> > fact that I had given a -1 to the topic.
>>>> >
>>>> > When you review a topic (not just a single change) you are presented
>>>> > with
>>>> > two buttons: "Publish Comments" and "Publish and Submit". I must have
>>>> > clicked the latter.
>>>> >
>>>> > I suggest we remove that second button to prevent accidental merges.
>>>> >
>>>> Hi Bill,
>>>>
>>>> I think the button is useful. I just verified that even if I
>>>> accidentally press the "Publish and Submit" button it will not merge
>>>> it unless I set the review to +2 for the topic. This means you need to
>>>> accidentally set review to +2 and hit the publish and submit button. I
>>>> like the convenience of being able to review and merge a topic in one
>>>> commit.
>>>>
>>>> The reviews on the individual commits in a topic are effectively
>>>> there to aid the review, when using topic reviews the top-level topic
>>>> score is the only one that matters. I would like to enhance that so
>>>> that a -2 review of a commit would block it. Right now we don't have
>>>> many (or any) cycles available to patch Gerrit, and would like to
>>>> rebase to the latest upstream before making any additional changes.
>>>>
>>>> I am updating some of the wiki pages, and can add a warning about this
>>>> but it doesn't look too dangerous to me as you must review a commit as
>>>> good to merge and then accidentally hit the wrong button, at which
>>>> stage you had already stated you felt the commit was good to merge.
>>>>
>>>> Marcus
>>>
>>>
>>>
>>>
>>> --
>>> Unpaid intern in BillsBasement at noware dot com
>>>
>>>
>>> _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://www.vtk.org/mailman/listinfo/vtk-developers
>>>
>>>