Unnecessary methods on Page

[Daniel Sheppard]

Just did a bit of a slash and burn on Page as there seemed to be a little cruft:

http://dev.radiantcms.org/radiant/changeset/568

All tests still pass, so nothing I removed seemed that valuable, but thought I'd double-check that I wasn't screwing up any untested code. Can you have a look-see over my changes?

The only actual change there should be is that render_text no longer exists, but that's because it was only ever used from the test helper (which now directly calls the private parse method)

[John W. Long]

Daniel Sheppard wrote:
> Just did a bit of a slash and burn on Page as there seemed to be a little cruft:
>
> http://dev.radiantcms.org/radiant/changeset/568

I believe this line is important:

tag.locals.page = tag.globals.actual_page || tag.globals.page

Sean would know. Perhaps it's untested.

--
John Long
http://wiseheartdesign.com

[Daniel Sheppard]

Just did a bit of a slash and burn on Page as there seemed to be a
little cruft:

http://dev.radiantcms.org/radiant/changeset/568

All tests still pass, so nothing I removed seemed that valuable, but

[Sean Cribbs]

Right.  The point of that, and the lines in the 'snippet' tag is to retain a pointer to the page that is currently rendering, while allowing snippets to render in the context of the local page, perhaps inside an iteration or 'find' tag.  I guess the tests for that weren't there.  I'll look into them as soon as I can.

Sean

[Daniel Sheppard]

But isn't tag.globals.page already a reference to the currently rendering page?  As far as I could see, the code was storing tag.globals.page as tag.globals.actual_page and then rendering a snippet through the render method of the current page... and then use the actual_page variable to

pretend that it was rendering in the context of another page instead. I changed it so that the rendering was just done by the render method of the

original page anyway - since you're not switching contexts, all the locals are there and the globals are also still correct.

 

 

---

From: radiant...@googlegroups.com [mailto:radiant...@googlegroups.com] On Behalf Of Sean Cribbs
Sent: Friday, 26 October 2007 11:44 PM
To: radiant...@googlegroups.com
Subject: [Radiant-Dev] Re: Unnecessary methods on Page

[Sean Cribbs]

Hmm, ok.  I guess we just need some new tests to verify that the desired behavior is still there.  There is a subtle difference between the approaches, I'm just not sure what the difference will be in the end.

Sean