This is just annoying to have failures from tic because portal_activities is enabled on a development instance.

This tests the fix for "slapos_jio: fix ERP5Site_activeLogin"

Those files might be customize on project related business templates, so we don't want to overwrite those files.

  Hosting Subscription requires SubscriptionItem which is now on erp5_open_trace

This fix up the default selected when no software type is pre-defined.

When you run tests, MailHost is normally changed to DummyMailHost, this script helps you to restore, in case you want to configure your development instance as a demo site.

This could be solved by adding extra asserts for wait for page be entirely load, but it is not required.

This could be solved by adding extra asserts for wait for page be entirely load, but it is not required.

The password is re-set anyway later, however it is required to comply to Authentication Policy, generating an acceptable value.

The user language will be stored on Subscription Request and on Person, defining the language used to notify the customer.

This is required when a computer reboots and there is an excessive number of break casts

The commit dee343f3 said:
slapos_jio: after activate the login come back to WebSite and not to the portal.
    
But it actually went to the WebSection and not the WebSite.
    
This fixes ERP5Site_activeLogin on slapos master because right now we
have this URL in emails sent to customer to activate their login:
    
https://slapos.vifib.com/hateoas/connection/ERP5Site_activeLogin?key=XXX
    
So the script is setupping the came from as:
    
https://slapos.vifib.com/hateoas/connection/#!login?p.page=slapos{&n.me}
    
which is not working (returns AttributeError: form_view error)
    
The URL https://slapos.vifib.com/#!login?p.page=slapos{&n.me} is working
though.


See merge request !223

The idea is to make it possible while embedding SlapOS in an existing partition to configure that some requests are forwarded to the master of the partition and to make these requests as the partition - so that when this partition is destroyed all other partitions from the embedded partitions are destroyed.

A bit more integration in slapos is still needed, but is intended to be used in theia (and why not webrunner):

At the beginning theia's instance  looks like this:

![screenshot of theia hosting subscription on slapos master, with the default 3 partitions](/uploads/9a297dbbae49996fb7ee69e89ea75da1/image.png)

after an ERP5 instance is requested in theia's embedded slapos, the partition requests are forwarded as this partition and appear here, as requested by theia:

![screenshot of theia hosting subscription on slapos master, frontend partitions requested by ERP5 appears as requested by theia](/uploads/37b8b30ca117a1251fc2bd099a5dac82/image.png)

This can also allow transparent integration, because the partition credentials are known by the partition, unlike the user credentials.

See merge request !218

The commit dee343f3 said:
slapos_jio: after activate the login come back to WebSite and not to the portal.

But it actually went to the WebSection and not the WebSite.

This fixes ERP5Site_activeLogin on slapos master because right now we
have this URL in emails sent to customer to activate their login:

https://slapos.vifib.com/hateoas/connection/ERP5Site_activeLogin?key=XXX

So the script is setupping the came from as:

https://slapos.vifib.com/hateoas/connection/#!login?p.page=slapos{&n.me}

which is not working (returns AttributeError: form_view error)

The URL https://slapos.vifib.com/#!login?p.page=slapos{&n.me} is working
though.

This is required to comply with erp5_authentication_policy

Set "value" cause some overprocessing in this case (which breaks some tests when invoked under superuser)

You can run it and increate by a certain amount the transaction ids used for payment, this may prevent ids conflics on payzen side when multiple developers are trying to test it during the same day.

If the payment is recently created (in the latest 30 min) process it every minute, if it is older them 30 min, process every 30 min.

We want to acknowledge the payment as fast as possible and usually when it occurs, it is processed on the first minutes.

... not from the Payment Transaction

slapos_subscription_request: (test) Link on the email comes from the invoice

This is an acceptable speed up as the amount of processing is small, and pontual, as states are updated quickly.

This is helpfull to configure configure computers to simulate connection parameters and allocation.

This is a change from Thomas Gambier.

This is a change from Thomas Gambier, I'm just commiting

Don't clear and allow us to pass parameters for the ansible commant which allow us pass extra variables.

  slapos_erp5: Add erp5_authentication_policy as project dependency
  slapos_cloud: Define standard configuration for authentication policy
  Update tests on various bt5.

  Modifies the Workflows meaning for Support Request Workflow

    Draft: Reservation isn't Payed, Person is created
    Planned: Reservation is Payed, User is going to be Notified
    Ordered: First month isn't payed, Hosting Subscription is created
             (started), user is notified about the payment to be done.

    Confirmed: Hosting Subscription is stopped until the payment
               is done. Once it done, instance is started and
               user is Notified.

    Started: User is available for this user.

    Stopped/Delivered aren't implemented.

  Nothing is implemented inside the workflow itself, alarms calls the
  scripts which changes the transitions once everything is ready to move
  foward.

  Allocation aoccur when if subscription request is in ordered state. We
  allocate earlier to reserve the space until the payment is done.