Question about features

[Daniel Carrera]

On Tue, Nov 3, 2009 at 12:16 PM, Maarten Bosmans <mkbosmans at gmail.com> wrote:
> For developing a website this is indeed not really a good approach.
> You can view my remarks in the light of a software development
> process, where one would have multiple active branches corresponding
> to different maintenance and development releases. In this scenario it
> really helps to know all the branches that need the bugfix.

Ok. In that context your suggestion makes more sense.

> You would probably have just one release: the live website.

Basically, yes. I have a test site where I do my work. When the
feature is complete I push to the live site. In any case, there's just
one release.

> In this
> case you can just commit to the mainline (trunk) of your project. Note
> that developing new features in separate branches from the mainline is
> really useful in this scenario too, because you can push (upload) the
> mainline directly after the bugfix commit. You now have a live website
> with the bugfix, but without your halfway done feature.

Yes, I see that. I currently use Darc's cherry picking to solve that
problem, but I could well use branches instead.


> When changes are shelved, they're only visible to the shelve plugin,
> so an upload would upload the working tree without the shelved
> changes.

:-(

Maybe I should just use rsync to send my work to the server for
testing. Then when everything works I commit... Ah, but that would
conflict with branches. If I make a separate branch to add another
feature then rsync wouldn't do the right thing anymore.

I'll have to give this more thought.
-- 
No trees were killed in the generation of this message. A large number
of electrons were, however, severely inconvenienced.