[jifty-devel] Some naming questions about fragments

Alex Vandiver alexmv at bestpractical.com
Tue Dec 27 23:55:52 EST 2005

On Tue, 2005-12-27 at 23:02 -0500, David Glasser wrote:
> For consistency with _elements, should we rename fragments to  
> _fragments?
Fragments are accessible from any browser, so I *think* they shouldn't
get an underscore, which to me implies "private."

> Is there a good reason that we're floating around all three terms  
> "region", "page region", and "fragment"?  (Is it "region is a block  
> on the page, fragment is mason on disk" or something? Or are they  
> basically the same?)
The difference is niggling, and possibly not important.  I think I
defined them in the glossary.  Let me look.  [comes back]  Yup.  Here
you are:

      A section of HTML contained in a "region". Fragments are a kind of
      standalone Mason component which the browser can request
      individually. Because of this, they can only take strings and scalars
      as arguments, not references or objects!
      An area of the page which JavaScript can replace. The content in the
      region is a "fragment". Think of the region as the box and the
      fragment as the content in the box. See Jifty::PageRegion.

> Also, for convenience, should the app script make empty _elements and  
> fragments directories?

> (And if every region must come from a fragments path, can't we just  
> make that implicit in the path arg to Jifty->web->region? Or would  
> that break because sometimes we're loading from the share/__jifty  
> fragments dir?)
I know we considered it, but left it in mostly for explicit-ness.  The
share/__jifty/framgents could be moved to share/fragements/__jifty if we
decided that that was the way to go.
 - Alex

More information about the jifty-devel mailing list