This is based on ERP5JS implementation

No need 2 bt5 with similar goal.

This code is old code for JIO which were never used in the end, clean up for the next implementation.

  This change were commited by mistake

  Get the person created by searching the ERP5 Login, because person has not reference defined.

- Use getUserId instead getReference for person login on self.login method
- Get the person created by searching the ERP5 Login, because person has not reference defined.

slapos_erp5: testSlapOSERP5LocalPermissionSlapOSInteractionWorkflow use person.getUserId instead person.getReference

Follow up changes on erp5 security.

Follow up implementation of ERP5LoginUserManager.

Follow up changes from ERP5Security.

This API was changed, so follow up changes which come from ERP5.

  Remove useless imports
  Use ERP5Security.SUPER_SUPER instead SUPER_USER to follow up original code.
  Also use UnrestrictedMethod on SlapOSShadowAuthenticationPlugin

  Reference commit nexedi/erp5@455b24e5

 This change follow changes on ERP5 code base:
  nexedi/erp5@4b16e1dd

  By invoking validate, the interaction workflow will be kicked and allocation will be triggered. In order to control the invocation order on the test, use portal_workflow API.

  This test is failing as it is incompatible with the latest version of renderjs. The Monitor UI is going to be merged into RenderJS, so keep this test as an expected failure until this test be moved along with monitor implementation to ERP5/OfficeJS repository.

Person.getTitle do not contains patches for a while so we fall back on default the implementation of ERP5. This don't require mantains a test on project scope.

This will prevent mistaken information when display the used amount of instances.

Move manager.instance to the part of code which doesn't get preempted 

/reviewed-on nexedi/slapos.core!21

The logic implemented on client is (for started instances):

Notify instances are available to master
run promises if fail report error
Notify instances are started to master

The change on this commit prevent us to consider the notification of availability as "no error", as promise can still fail, so we just ignore and let the decision to suspend for the next run.

This makes the ease to get the message, on certain special cases.

See ERP5 commit 8995b64508f6d7c0fee1ef475de3dc3145ddc84d.

  This version include initial plugin system and cgroups integration.

  We also included a significant change for provide the partitions network informations, which now
  don't require to come from slapos master anymore, so it is possible to make local extensions.