Any reason this isn't a patch against strongarm? There are a few bits to figure out, but they are nothing compared to the issues strongarm's current code causes.

Comments

das-peter’s picture

Status: Active » Needs review
StatusFileSize
new6.51 KB

Hi there,

In lack of an existing patch I created one. It's build against the latest 7.x-2.x-dev version of Strongarm.
I also made some small additional changes:

  • Replaced the usage of theme("table_select_header_cell") - it's deprecated for D7.
  • Fixed issue in variable_features_revert() - replaced $vars['name']->value by $vars[$name]->value.
Pisco’s picture

Fixed a typo in das-peter's patch and posted it to #1094598: Performance issues from strongarm_init().

das-peter’s picture

Thank you both for keeping this alive :)
@Pisco: As promised I digged into the installation issue. I think I got it - at least it works out on all our stages.

The patch is again for 7.x-2.x-dev and includes:

To make it easier to update the patch for 6.x I've attached the diff between the both patches.
Changes:

  • Changed variable value comparison from != to !== needed to ensure arrays are really equal.
  • Changed condition in variable_features_rebuild to this
    if (isset($vars[$name]->in_code_only) || (drupal_installation_attempted() && $vars[$name]->export_type & EXPORT_IN_CODE)) {
    This allows that the variables are set during the installation.
    (Maybe we don't event need isset($default->in_code_only) now - but I'm not sure about that.)
catch’s picture

Status: Needs review » Fixed

I'm going to mark this issue fixed since #1094598: Performance issues from strongarm_init() has a patch.

Status: Fixed » Closed (fixed)

Automatically closed -- issue fixed for 2 weeks with no activity.