[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [sc-dev] 3.7 branch




What I proposed:

The only things that should go into 3.7.0 right now are the updates to the readme, install instructions, changelog.

Then its a beta; final check through spit and polish. Then its released.

Development work on 3.7.0 is finished.

After it is released it is merged back into master, bringing any of these minor commits with it.


This is better then cherry picking master into 3.7.  That is error prone. 

Personally I only cherry pick to fix accidents, never as an intentional work flow.


This is not proposed:

+ make important bug-fixing changes in 3.7
+ abandon master
+ eventually release 3.7 based on a tag in branch "3.7"
+ rename 3.7 to master





On Mon, Dec 14, 2015 at 7:02 PM Till Bovermann <lfsaw@xxxxxxxx> wrote:
sorry to chime in...

isn;t the usual process to

+ make all changes in master
+ cherry-pick to 3.7
+ eventually release 3.7 based on a tag in branch "3.7"

and not

+ make important bug-fixing changes in 3.7
+ abandon master
+ eventually release 3.7 based on a tag in branch "3.7"
+ rename 3.7 to master

?

a bit puzzled here since I am very well a git user as described here:
        https://xkcd.com/1597/


cheers
        Till


> On 08. Dec 2015, at 03:47, Michael Zacherl <sdiy-mz01@xxxxxxxxxxxxx> wrote:
>
>
> On 7.Dec 2015, at 13:34 , Chris Sattinger <crucialfelix@xxxxxxxxx> wrote:
>
>> you could also switch to the 3.7 branch.
>
> 3.7 is my main system now.
> I've been running 3.6.6 & current master side by side virtually all the time and just added 3.7 to it.
> Nevertheless I'm trying to keep the master branch useable.
>
>> because that also fixes another glaring bug in Quarks gui that nobody reported.
>> (if you uninstall a quark the gui still displayed it as installed)
>
> I stumbled upon this recently but didn't have the moment to clarify if that's on my side or not.
> Apparently it's not. ;)
>
>> its better if more people are using the 3.7 branch
>
> That's what I'm doing.
> I got three separate environments … while I use 3.6.6 less and less the gig tomorrow I'll play on 3.7
> Development of my stuff has been happening on current master for like a year or more now and continues on 3.7 for the time being.
>
>> On Mon, Dec 7, 2015 at 12:49 PM Michael Zacherl <sdiy-mz01@xxxxxxxxxxxxx> wrote:
>>
>> On 7.Dec 2015, at 9:03 , Chris Sattinger <crucialfelix@xxxxxxxxx> wrote:
>>
>>>
>>> Good morning. We have a 3.7 branch, and we already have a problem.
>>>
>>> There is an important fix to Quarks gui merged there (because of the string-replace issue), but not to master.
>>>
>>> We could:
>>>
>>> 1. Just update the documentation and release a 3.7 beta
>>> 2. Tell people who are compiling from source to switch to the 3.7 branch which is good for quality control
>>> 3. Start a tedious process of cherry picking every little thing and endlessly discussing it
>>
>> I just patched the master branch locally to get Quarks up and running again.
>> No real problem, no?
>> OTOH, avoiding "whinging over broken code" isn't a bad thing, if it's that easy.
>> But … where to draw the line?
>> ;)
>>
>>> I vote for 1 + 2
>
>
>
> _______________________________________________
> sc-dev mailing list
>
> info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
> archive: https://listarc.bham.ac.uk/marchives/sc-dev/
> search: https://listarc.bham.ac.uk/lists/sc-dev/search/


_______________________________________________
sc-dev mailing list

info (subscription, etc.): http://www.birmingham.ac.uk/facilities/ea-studios/research/supercollider/mailinglist.aspx
archive: https://listarc.bham.ac.uk/marchives/sc-dev/
search: https://listarc.bham.ac.uk/lists/sc-dev/search/