Feature: add support for workbench_menu: hook_query_TAG_alter() for workbench_menu_autocomplete_nid()

Manually specify the SQL join types for the views.
This helps ensure consistent and proper SQL joins.

This module was originally written for cf-1.x.
It has been modified to be compatible with cf-2.x's pseudo-backwards compatibility.
It turns out that during install, if the cf-2.x backwards compatibility is not enabled, it is then possible to create a WSOD.
To prevent this, a lot of the old cf-1.x style functions are being wrapped in conditions to see if cf-2.x is in use.

This is done to ensure forwards compatibility where $function_history will likely be set to null.

Add missing view properties that are required to ensure aggregate node views function properly.

The website was seemingly randomly printing:
  Notice: Undefined index: entity type in views_handler_field_entity->init() (line 38 of /var/www/html/sites/all/modules/views/handlers/views_handler_field_entity

It turns out that the killfile module was not defining 'entity type'.
While it is not clear that the killfile module should be defining this, for now just define 'entity type'  as 'node'.

see: https://drupal.org/node/1364064

node/[nid] for a given node should always exists, so it should be safer to use.

This will allow for testing if killfile is enabled for some node's node type before executing any rules actions.

Writing code like:
  return array(
    'key' = 'value';
  );

Is simpler to do, but the following code better promotes patching and changing:
  $items = array();
  $items['key'] = 'value';
  return $items;

This restructures the code for that reason.

Update: add watchdog messages when there is a problem with the killfile and unkillfile rules actions

At some point the function killfile_get_node_type_settings() was changed to accept an array of node types instead of a node type.
The undelete page, the delete rules action, and the undelete rules action were overlooked.

Display the node title in the message about the deleted content.

Follow the drupal coding standard here: http://drupal.org/node/161085

There are cases where $node->uri does not exist, so check that the
object property exists before use.

Use the smallest possible footprint for boolean database columns.

Now that drupal is on git, the usage of a static changelog seems like a
waste of time. It is more likely to be incorrect and it will require
druplication of effort to maintain.

Issue #1072130 by thekevinday: Some functions need to accept and array of nids or an array nodes, not just an array of nids.

For performance reasons rewrite killfile_get_node_type_settings() to accept an array so that mass node types can be loaded in a single database transaction.

Move the token hooks inside the main module and remove the token include file because the hooks were not being discovered when placed inside a separate include file.