I know that each repository is independent but anyway, the server knows each
one because you can access with the right path.
It is something I don't know and I think this can be explained by
developpers only. How does the SVN server work ?
Let's say we have 3 directories under the SVN Parent PathParent path is
"D:\SVNrep" and subdir are called "Repos1", "Repos2" and "Repos3".Each one
contains a repository.
Using SVN server, you have to specify address like this
svn://<hostname>/Reposx
where x is value 1, 2 or 3, depending on which repository you want to
access.
Does the server know it is 3 different repositories in his ParentPath or
does he only goes into D:\SVNRep\Reposx to see if a repostory exists ?
I'm asking this because I don't take a look to Server code but I think that
can be added either
1- easily if server check all repositories at startup and keep them in
memory
2- or "quite easily" by browsing subdirs to find repositories when the
server receive the command for repositories listing.
For point 2, it is necessary to know how to identify a repository in order
to be sure subdir is a real repository.
But it is only my point of view.
Franck
----- Original Message -----
From: "John Peacock" <jpeacock@rowman.com>
To: "ISIT - Service Technique" <isitech@isit.fr>
Cc: <dev@subversion.tigris.org>
Sent: Tuesday, August 07, 2007 3:25 PM
Subject: Re: New basic feature ?
> ISIT - Service Technique wrote:
> > The problem is that we don't want to use nor apache server either web
> > browsing tools for this. We just plan to use SVN internal server.
> > That's why I would prefer to have a specific SVN command that would list
all
> > repositories.
> >
> > Do you think this is something that can be added ?
>
> Not as such, because each repository is completely independent. There
> is no "master config file" where the repositories could be registered.
> Apache can get away with it because the webserver itself can enumerate
> the directories underneath SVNParentPath.
>
> Why are you so opposed to running the Apache server? You can limit it
> to HTTPS only (for security) or run it on a non-standard port (e.g. if
> you are already running Apache 1.x), and of course you needn't run it on
> a public IP address (you can even use SSH tunneling to further control
> access).
>
> John
>
> --
> John Peacock
> Director of Information Research and Technology
> Rowman & Littlefield Publishing Group
> 4501 Forbes Boulevard
> Suite H
> Lanham, MD 20706
> 301-459-3366 x.5010
> fax 301-429-5748
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Aug 7 16:58:26 2007