[m-dev.] SourceForge web site

[Tyson Dowd]

On 09-Apr-2001, Fergus Henderson <fjh at cs.mu.OZ.AU> wrote:
> Hi,
> 
> Since users had reported problems accessing the Melbourne web site,
> I had a look at the SourceForge one.
> 
> First, a heads up on how the SourceForge site works.
> The site is named mercury.sourceforget.net.  It's running Linux.
> If you have a sourceforge account, and you are listed as
> a member of the Mercury project, then you can log in to it using ssh.
> The web pages are stored in /home/groups/mercury/.
> Files can be copied to and from the site using scp, or you can use cvs.
> It's pretty simple, really -- that's about it.
> 
> There were a lot of problems with the current SourceForge site.
> In particular:
> 
> 1. The web pages were set up with remote CVS using Tyson's account
>    (:pserver:tyson at hydra.cs.mu.oz.au) which meant that only Tyson
>    could update them, since only he had the password.
> 2. The files were not up-to-date w.r.t. our CVS repository,
>    but also contained a lot of uncommitted changes.
> 3. There were some unresolved merge conflicts in bin/install_webpages
> 4. There were many dangling links to *.html,
>    because the HTML files hadn't been created.  SourceForge runs php3
>    directly from the web server, so they're not needed; but many of the
>    links on our pages point to them.
> 
> To solve these problems, I did the following:
> 
> 1.  I modified all the CVS/Repository files so that they use anonymous CVS
>     (:pserver:guest at hydra.cs.mu.oz.au).
> 2.  I did a cvs update.
> 3.  I fixed bin/install_webpages manually.
> 4.  I copied the php3 executable to mercury.sourceforge.net, and edited the
>     Makefile.common to use that copy rather than assuming that php3 lives
>     in /usr/lib/cgi-bin, and then did `make'.
> 
> There's three issues that I haven't fully addressed:
> 	- I haven't done anything about the uncommitted changes.
> 	- I haven't done anything to ensure that the site remains
> 	  up-to-date in the future (e.g. setting up a cron job).
> 
> For the uncommitted changes, I think most of them (the ones for using
> links to *.php3 rather than *.html) should be discarded.  It's simpler
> to use the same approach with the HTML files on SourceForge as we do for
> the Melbourne site (building them statically rather than dynamically),
> and it may even be a bit more efficient that way.  Any objections?

That's fine.  The changes to use .php3 instead of .html turned out to be
more work than I expected, and so it's probably a better idea to do it
that way.

-- 
       Tyson Dowd           # 
                            #  Surreal humour isn't everyone's cup of fur.
     trd at cs.mu.oz.au        # 
http://www.cs.mu.oz.au/~trd #
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at cs.mu.oz.au
Administrative Queries: owner-mercury-developers at cs.mu.oz.au
Subscriptions:          mercury-developers-request at cs.mu.oz.au
--------------------------------------------------------------------------