Re: [wwwdocs] Complete revamp of our web site

[Tom Tromey <tromey at redhat dot com>]

>>>>> "Gerald" == Gerald Pfeifer <gerald@pfeifer.com> writes:

Tom> I suppose we could do something similar on gcc.gnu.org... check in one
Tom> file per news item, then have blosxom (or something) run to make files
Tom> which are then passed through mhtml.  (For all I know mhtml already
Tom> has a module for this kind of thing.  Do you know?)

Gerald> I don't, but once we have something that generates HTML or
Gerald> HTML snippets, I'll be happy to help and see what I can do
Gerald> push things onto the web pages.

Ok, here's one approach.  It is pretty hacky, but not substantially
more so than the rest of our web pages :)

I took all the news entries currently on the gcj front page and split
them into separate .txt files.  (I did a kind of lame job -- we
should use real titles instead of dates...)

Then you run the mildly hacked copy of blosxom that is included in the
attachment to generate HTML and RSS.

This is still unfinished.  You'd need to hack blosxom a little bit
more so that it doesn't try to generate a real HTML file -- this is
easy (look for the __DATA__ section to see how to eliminate headers
and footers), and then hack the meta-html stuff to actually include
the resulting file into the front page.

The idea is that we can generate news for the front page and also for
the various sub-projects, in both HTML and RSS.  To add a new item a
maintainer would simply write a new .txt file in the appropriate
subdirectory and check it in; the commit scripts would do the rest.

Tom

Attachment: news.tar.gz
Description: news tar