So what is going on for me I noticed when I tried out cck taxonomy super select. I am trying to let users tag themselves. It's no problem to tag and pick more tags later. What I CAN'T do is remove the pick, I can do it, but it's not saved. This only happens on the nodeprofile content type... Has anyone else experienced this?

Comments

aschiwi’s picture

aschiwi CreditAttribution: aschiwi commented

After doing some testing I can see that this only happens if the profile content type is edited from within the user profile, ie either by using the edit tab. It does not happen when I edit the node by itself from the contents page.

aether’s picture

aether CreditAttribution: aether commented
Version: 5.x-1.3 » 5.x-1.4

I can confirm that I am experiencing this issue as well with 5.x.1.4. Not sure if this issue belongs here or with Taxonomy Super Select as it only happens when using that module with Node Profiles.

brenda003’s picture

brenda003
she/her
CreditAttribution: brenda003 commented

I also have this bug, and we are NOT using Taxonomy Super Select.

This only happens when editing it through the user profile categories.

Doing some investigating, it appears that the node is saving twice. First, the new terms DO get inserted into term_node, all the other fields are inserted into the database, etc., but then for some reason it does it over again, deletes the nid from term_node and inserts AGAIN with the old terms selected (or none, if none were selected).

brenda003’s picture

brenda003
she/her
CreditAttribution: brenda003 commented

Doing some testing, this around line 363 seems to be the culprit:

case 'update':
  if (nodeprofile_get_settings('user_edit', $category) && node_access('create', $category)) {
    subform_element_submit();
    unset($edit[$category. '_node_form']);
  }
  break;

Changing the case to 'after_update' rather than 'update' seems to solve this without noticing any affects. It's still trying to save everything twice, it just saves the proper terms last now. But I think this is actually be a bug with subform_element, specifically the subform_element_call... can debug more later. Thoughts on this, fago?