[Om3] svn.openmath.org is now opened

[James Davenport]

On Sat, 7 Jul 2007, Paul Libbrecht wrote:
> Le 7 juil. 07 à 17:28, James Davenport a écrit :
> > I would like to see a system of subdirectories of cd (and a
corresponding
> > structure for sts):
> > official - what it says
> > officialdraft - I would morve CDs into this as I sent them out for
> > comment, and move them from here to official,
>
> That sounds fine to me.
>
> > The above two should only be writable by me and the support team
>
> That should be doable. Note that for now, we can only apply this on
> honour between the (trustful) committers.
Indeed so - mine was a statement of principle as much a software
wish-list.
> > experimental - General place for experimental ones
> > named subdirs - for people who request sub-directories for various
> > collections they are building.
>
> I would really encourage projects' named directories.
Indeed so. I was thinking that we need 'experimental'
(a) for the current experimental CDs (that don't fit neatly into projects)
(b) for any new 'one-off' CDs.
Incidentally, I propose that it should be possible to put a whoe 'project'
out to review without it haing to move through officialdraft.
> > The search hierarchy should be in descending order, i.e. official
> first.
> That is a bit tricky.
In svn itself, yes. I was thinking of the search order implied by
'getting a CD from the standard place'. This could be done by that
software, or by a nightly build process.
> There is no search hierarchy involved in svn (or any other versionning
system
> I think). So I think the way would be to rely on alphabetical order
being
> correct... that'd need fancy ASCII characters somewhere, for example the
> following:
>
> _official
> _officialdraft
> experimental
>
> it's not very elegant!
True indeed.
> Haven't done anything yet.
Fine.
James