erp5_base.Base_getURLTopmostDocumentValue.xml: add a method to get the topmost traversed url in an ERP5 instance. In base it returns the portal.

Class FolderMixIn: add isURLAncestorOf method. Returns True if the given_url is child of the url of the given_url is child of the url of the document that the method is called on, False otherwise

This is too inconvenient when resizing browser size, it requires in
some cases to do taller or shorter many times only because we resized
the browser.

Also, former way was not working nicely when editor like code mirror
was used

No workflow state provides Modify Portal Content to a Role (except Manager).
Change the guard to be based on role instead of permission, so that Assignor can also trigger the transition.

This reverts commit 85c87bc6.

This publication section is not present in erp5_web_renderjs_ui or any
of its dependencies, so we cannot reference it from here.

ERP5Site_reindexAll is now called only once when --recreate_catalog=1 is explicitly specified.
Also manage_catalogClear is not called with --load unless --recreate_catalog=1 is explicitly specified.

erp5_test_result: fix AttributeError: string_index in TestResult_view/listbox_string_index tales expression

because it is a generic tool and access restriction should be done by API.

test.erp5.testUpgrader.py: update the bt5list in afterSetUp(). Otherwise the user has to invoke genbt5list manually before runnning the test. It was also confusing, because the same test would pass in the server but not locally

Solve divergence dialog was raising unauthorized error when new solver process was created
by this script (when called with someone who was not manager). Therefore use proxy role and
protect call with "View" permission. View is used instead of modify portal content since we could
imagine cases where user might need to solve divergence even though is has no right to modify
delivery.

Merge work of Ayush Tiwari (@tiwariayush) to support ERP5-side code
execution for Jupyter (= IPython) notebook:

---- 8< ----

Jupyter == IPython Notebook

Features:

1. Module - Data Notebook (To save notebook according to reference from jupyter frontend)
2. Execute python code which might/might not be remote.
3. Save local_variables for specific notebook, i.e, you can use the python
   objects which were created during code execution. After making a successful
   connection(from frontend) to specific notebook, the variables which were saved
   earlier are available while for execution.

Architecture:

- Code from Jupyter cell is being received as string by Base_executeJupyter.
- Code is executed in Base_runJupyter external method.
- Local variables saved via CMFActivity.
- Result as JSON is returned to frontend (contains mime_type of result string also).

Example Notebook:

    http://nbviewer.ipython.org/urls/lab.nexedi.com/nexedi/erp5/uploads/6778782d5753873583bd4843772cf502/Proof_of_concept_Notebook.ipynb
    https://lab.nexedi.com/nexedi/erp5/uploads/6778782d5753873583bd4843772cf502/Proof_of_concept_Notebook.ipynb

---- 8< ----

Current state is: erp5-data-notebook somehow works on server side, but is known
(see slapos!43) to have at least the following limitations:

- errors are not reported properly to users;
- state is not fully saved to ZODB.

the latter point means notebook works only if it is connected to Zope family
with only 1 zope process.

@Tyagov reviewed this work. I'm the person who does the merge.

Slapos part to integrate erp5-data-notebook bt5 and Jupyter frontend
will come in via slapos!43 .

/reviewed-by @Tyagov  (on !29 and on a lot of other places)