Re: [PROPOSAL] AuthzSVNAccessFile for normal web-files

On Thu, 11 May 2006, Branko ?ibej wrote:

> Here it is then. Note that it's not exactly the fastest possible
> implementation, as I didn't attempt to cache the authz info between
> requests.

Hello Brane,
thanks for your post!

I don't know Python, but it seems to me, that the module treats only the
[groups] section, but not the paths.

Here is an example setup for a clearer description of my requirements:

* One svn repository with:
  /project1/...
  /project2/...
  /project3/...

  Here, [...] stands for source-code, C-files or similar.

* And one normal directory with:
  /project1/...
  /project2/...
  /project3/...

  Here, [...] stands for automatically generated docs from the sources in
  HEAD. These directories and their files are also served by apache with a
  normal HTTP GET (no DAV needed).

* One access file:
  [groups]
  g1 = u1, u2
  g2 = u3, u4
  all = @g1, @g2

  [/]
  @all = r

  [/project1]
  u1 = rw
  u2 = r
  @g2 =

  [/project2]
  u3 = rw
  @g1 = r
  u4 =

  [/project3]
  u4 = rw

* In the end, if user X is able to read SVN project Y, then he should be
  able to read files in the directory with the docs of project Y too.

It should be noted, that the access file can change quite often, since
users can add projects and modify access restrictions via a web interface.

Greetings, Peter

-- 
http://pmrb.free.fr/contact-DD/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org