[ic] Bypassing the whole [more] system

Grant interchange-users@icdevgroup.org
Mon Sep 30 12:32:01 2002


>I thought I would respond since no one else has....
>
>I have been thinking about configurable missing-URL routines. Right now,
>the default is that if a page is not found, a flypage is tried. If that
>is not found, then in 4.9 the DirectoryIndex is tried (allowing index.html
>to be the index page for a directory). Then it is passed to the special
>page "missing", whatever that is defined to be.
>
>I thought of adding a "PageResolution" directive, i.e.:
>
>    PageResolution   category_page Vend::Interpolate::fly_page
>
>That would allow the Sub/GlobalSub "category_page" to get first crack
>at a missing URL, followed by a try at the flypage.
>
>Of course, the most efficient way to do this would be with a
>"category" ActionMap, i.e.:
>
>	http://foo.com/cgi-bin/bar/category/Category Name/1.html
>	http://foo.com/cgi-bin/bar/category/Category Name/2.html
>	http://foo.com/cgi-bin/bar/category/Category Name/3.html
>
>The special "category" causes the ActionMap category to be run,
>and the category could be followed. The trailing 2, etc. could
>be used for paging.

Mike -
Thank you very much for your response.  Both of your ideas sound great, but
I think I like the sound of the ActionMap better.  That way the different
category pages (Page 1 | Page 2 | Page 3) have distinct pages names.  That
way they will all be indexed by search engines.  If I understand the
PageResolution directive correctly (and I may not) all the category pages
would be called "category_page", which I think would end up with a lot of
category pages and flypages not being indexed.

I looked through the ActionMap docs online and I'm a bit confused.  If
"category" is in the URL like above, a certain action is executed which is
usually defined by a Perl subroutine.  In your example above what would
happen?  If a page called
"http://foo.com/cgi-bin/bar/category/CategoryName/1.html" is requested, the
URL gets rewritten as "http://foo.com/cgi-bin/bar/CategoryName/1.html" and
the specified subroutine is run?  Each of those HTML pages would be
"imaginary" like the process page right?  The actual page that is displayed
would be defined in the subroutine?  Is that all correct?  I think I'm
missing something.

What do you think about bypassing the [more] system in general?  Do you
forsee any problems with doing something like that?

Thanks again!

- Grant