Project

General

Profile

Feature #168

Contentlets should not assume template content or safe content.

Added by Stephen Burrows almost 13 years ago. Updated over 12 years ago.

Status:
Resolved
Priority:
Normal
Assignee:
-
Category:
Core
Target version:
-
Start date:
06/24/2011
Due date:
% Done:

0%

Estimated time:

Description

Currently, contentlets mark their content as safe and even compile it as a template. There is no way for template authors to prevent this. It's legacy code from before the include_string tag; now that we have that, we can and should remove the automatic compilation and the mark_safe call.

Associated revisions

Revision 34a9cd2a (diff)
Added by Stephen Burrows almost 13 years ago

Removed automatic interpretation of contentlets as templates as per issue #168.

Revision 93eac03c
Added by Joseph Spiros over 12 years ago

Merge branch 'develop' of git://github.com/melinath/philo into develop

  • 'develop' of git://github.com/melinath/philo:
    Added ordering to NavigationItemInline and removed the extra form a) because it's unnecessary, and b) to avoid grappelli issue #337.
    Added memoization to TargetURLModel.get_target_url.
    Removed unnecessary foreign key evaluations from MultiView subclass urlpatterns and get_reverse_params().
    Cleaned up shipherd docs.
    Added philo-root reversal caching to Node.construct_url.
    Pre-fetching the contentreference content may have been premature optimization. Switched to a simple fetch, as it seems to run slightly faster.
    Minor correction to ContainerContext.get_references().
    Moved caching of contentlets/contentreferences so that (for example) a page with only contentlets will not fetch content references. Set up content references to pre-fetch all related pieces of content to save queries.
    Removed template container discovery in favor of (potentially) larger queries.
    Moved container finding code into philo/utils/templates, along with template utils from philo/utils. Moved initial container fetching code for container template tags from ContainerContext.__init__ to ContainerNode.get_container_content and cleaned things up a little.
    Removed automatic interpretation of contentlets as templates as per issue #168.
    Tweaked the FeedView API slightly to pass the object from get_object into the get_items callback. This allows for FeedViews that do not have FK relationships to the object.
    A fair amount of effort is wasted getting container content. Initial commit of a branch to fix that. There's a large speedup on container fetching, but the container detection algorithm is prohibitively slow. Should be sped up or bypassed.
    Added link generation example, and fixed a typo.
    Note about context processor in Shipherd tutorial.

History

#1

Updated by Stephen Burrows almost 13 years ago

  • Status changed from New to In Progress

Included in commit 34a9cd2a670e84771d1d54e6fd6af87bcb1c279c.

#2

Updated by Stephen Burrows over 12 years ago

  • Status changed from In Progress to Resolved

Merged into develop with commit 93eac03cfd379525b6cf41825b14a5d37cd03505.

Also available in: Atom PDF