Re: Removing control, finance and quaternion toolboxes

[Jaroslav Hajek]

On Wed, Jul 30, 2008 at 12:19 AM, Thomas Weber
<address@hidden> wrote:
> On 29/07/08 16:54 -0400, John W. Eaton wrote:
>> On 29-Jul-2008, Jaroslav Hajek wrote:
>>
>> | Thomas Weber has already agreed to host the release repos, but he
>> | suggested to use a single repo with branches instead. (Did you get a
>> | Cc?) The reasoning seems valid to me. Unless you have
>> | counter-arguments, I'll go ahead and dive into it.
>>
>> I think a separate archive for each 3.0.x release is not necessary,
>> but since the branch point for each new stable series (3.0.x, 3.2.x,
>> etc.) will be different, I don't see the point of having a single
>> archive for each of those.  I don't plan to have these branches in my
>> archive (as the release-3-0-x branch is now) because it doesn't seem
>> to be of much help.  But since you will be the one dealing with the
>> archive(s), it is up to you how you organize it (them).
>
> Okay, to ensure that we are all talking about the same thing:
>
> I understood Jaroslav as follows: he wants a new repository for each 3.0.x
> series, ie we have repositories:
>        octave-3.0.2
>        octave-3.0.3
>        ...
>

OK, sorry for the confusion. Perhaps you could create a separate
directory releases/ with write access for me, and I will create the
release repos myself. Is there anything else needed for web access
than the [web] section in .hg/hgrc (I guess not)?
Alternatively, you can just create release-3-2-x and release-3-0-x.

thanks a lot,

> I think that would be overkill because as soon as a version is released,
> nothing is going to happen in that specific repository anymore. And as
> octave-3.0.3 would be a clone from octave-3.0.2 anyways, it seems simpler to
> me to use just one repository octave-3.0.x and tag release in there.
>
> I don't think that repository should contain 3.2 stuff, in fact if we come to
> releases for the 3.2 series, a separate repository might be in order.
>
>        Thomas
>



-- 
RNDr. Jaroslav Hajek
computing expert
Aeronautical Research and Test Institute (VZLU)
Prague, Czech Republic
url: www.highegg.matfyz.cz