Add support for using the Scald Atom Reference field as image source

Hi.

I'm loosely familiar with Scald, though I have not used it on any official projects so the technical details are not all super-clear to me. There are a couple methods within the views style plugin that could probably be updated to support additional sources (assuming the direct source image URLs can be extracted from the rendered view object), but first I wonder if there are any existing alternatives. For example, is it possible to use some form of file relationship on your view to expose the file ids through an Atom Reference field? If so you might be able to setup a view based on Scald data/filters/context that still exposes a field that the Juicebox formatter already knows how to work with. I suppose this depends a bit on the level of views integration offered by Scald.

Any updates regarding the points in #4? I'm just curious if we should keep this open.

Hi All,

The patch works fine ,thanks

root@asad-Vostro-3550:/var/www/html/git/juicebox# git apply -v juicebox-use_scald_atom_reference_field_as_image_source-2725279-5.patch 
Checking patch includes/juicebox.field.inc...
Applied patch includes/juicebox.field.inc cleanly.
root@asad-Vostro-3550:/var/www/html/git/juicebox# 

Thanks for sharing this.

A couple questions:

• +    if (isset($item['sid'])) {
  +      // It is a scald atom. Convert to a conventional file item.
  

  I'm wondering if there is a more precise conditional check that could be used here? I'm not sure if any other modules could add an sid to any non-scald field items but it might be best not to take the risk. At a minimum I suppose we would need a module_exists('atom_reference') check.

• +      $atom = scald_atom_load($item['sid']);
  +      $fobject = file_load($atom->base_id);
  +      $item =  (array) $fobject;
  

  Are there any common cases where we might have a broken atom link (as sid without a file object) that needs to be checked for? I don't know scald internals well enough, so it's very possible that this is all fine as-is, I just wanted to check. If we are always guaranteed a file object if there is an sid then could this be reduced to one line?:

  $item = (array) file_load(scald_atom_load($item['sid'])->base_id);
  

• Does this all play nice with the existing title and caption options? How does scald store metadata like that?
• Would anyone be willing to write a test case for this?