>
> tis 2008-02-26 klockan 01:08 +0100 skrev Henrik Nordström:
>> tis 2008-02-26 klockan 13:01 +1300 skrev Amos Jeffries:
>>
>> > Henrik:
>> > is there something special to do with individual patches not to be
>> > merged from a group of patches which as a whole are?
>> > The include directive group which I have just done had 3 in the
>> middle
>> > which you may recall fixing-removing strtok_r usage.
>> > For now I have made them a seperate group not-merged and gone
>> straight
>> > past to your strwordtok fix.
>>
>> I usually merge groups of related HEAD patches in one single commit on
>> the STABLE branch. This question gets a lot easier then as it's the
>> whole patch group that was merged, not the individual incremental
>> steps.. and this also simplifies release maintenance of the STABLE
>> branch as there is less patches to read up on when updating changelogs,
>> release notes etc.
>
> Forgot.. splitting the groups to move out changes which has been backed
> out is not a good idea as you then easily loose context on why these was
> added and then backed out. Causes more work later in the initial
> maintenance of the next cycle.
I considered it, but wasn't too keen on loosing the author attributions
where a feature is added then fixed by others. if you have found that
okay, then I'm all for less work.
>
> Also in some cases it's not so easy to separate "bad & redone" things
> from the good in a specific subproject. Therefore merging them all
> together is highly preferable unless there is good reasons to have them
> split on the STABLE branch as well.
Ok, I've marked the removed ones as merged into the real fix patch.
Amos
Received on Mon Feb 25 2008 - 17:45:15 MST
This archive was generated by hypermail pre-2.1.9 : Sat Mar 01 2008 - 12:00:09 MST