# Last Feedback (20-Sep-20) from the latest test in NHS.

1. When you are going to create a new dataset, you will be stuck with the draft without the option to finish dataset create, cause you cannot exit resource page explicitly.


> @mbeilin what is the current flow (in standard ckan) and what do we think the flow should be?

2. `url` property is not generated in resource, so you cannot download resource file

> @mbeilin this is a complex one and shouldn't really be fixed in editor - this is something that should be done i think in ckanext-blob-storage or elsewhere. Need to sync with shahar on this.


3. It's imposiible actually to edit schema on `resource create` step, cause once you change the schema and push th `Save Metadata` button `package_update` runs that updates `metadata` of all the resources in current dataset.

> @mbeilin i don't get how this makes it impossible ...? It just means the entire package is saved atm. (it's not changing the metadata of the other resources is it??)


But for `aircan` to be triggered again (after editing schema) for specific resource I need the `resource_update` to be run, and I can do this by changing(or inly trying to change) the resource file. In that case `resource_create` <del>with the same `resource_id` running (that's fine)</del> with the new `resource_id` and the new schema property generated ignoring my previous schema change...

> ~~@mbeilin you've got to be triggering aircan on package_update package_create as well as resource_update and resource_create ~rufus. That's a bug if we are not doing that ...~~


Let's say I've updated the field `u'FID` type from `integer` to `string`


After that, I want `aircan` to be triggered again on the same rsource but with my latest schema changes, so I try to update file (even without changing it actually) - the `resource_create` is running, but new `schema` property is generated from the file and my schema changes are ignored.

> OK, so i think this *is* a bug if i understand correctly. We should infer a schema for a file *if and only if* there is no schema already there *or* we are uploading a new file (?)


4. According to the comment here - https://github.com/datopian/datapub/issues/31#issuecomment-694126659
I need for now one custom field to be included into resource, so it's definitely project specific requirement, but this one is must have for us.
The field is `bq_table_name` and have to be generated from `id` (resource id) uuid without `-` (according BigQuery table names conventions) and have to be presented in resource, like this example:

> @mbeilin what's the underlying user story and why do we need this?
> @rufuspollock - When CSV file added to the resource we want it to be available via datastore(BigQuery) API with `bq_table_name` as datastore table id.

{
    "help": "https://ckan.nhs.staging.datopian.com/api/3/action/help_show?name=resource_show",
    "success": true,
    "result": {
    "cache_last_updated": null,
    "package_id": "cc998fb0-7752-41bd-b555-8162fb0ea574",
    "datastore_active": false,
    "id": "0411c27d-92ff-45cf-8c0d-5777c6beaa29",
    "bq_table_name": "0411c27d92ff45cf8c0d5777c6beaa29",
    "size": null,
    "restricted": "",
    "encoding": "utf-8",
    "state": "active",
    "schema": "{u'fields': [{u'type': u'string', u'name': u'FID', u'format': u'default'}, {u'type': u'number', u'name': u'MktRF', u'format': u'default'}, {u'type': u'number', u'name': u'SMB', u'format': u'default'}, {u'type': u'number', u'name': u'HML', u'format': u'default'}, {u'type': u'number', u'name': u'RF', u'format': u'default'}, {u'type': u'number', u'name': u'RFFF', u'format': u'default'}], u'missingValues': [u'']}",
    "dialect": "{u'quoteChar': u'\"', u'delimiter': u','}",
    "hash": "4c38966349b6b2f0785cfe9eb5b7063ce5ed279a6f3ba0bdcbfb73ef7368c3e8",
    "description": "",
    "format": "CSV",
    "last_modified": null,
    "url_type": null,
    "mimetype": null,
    "cache_url": null,
    "name": "ckan_resources_ef1e364a-f070-4e08-b226-5fc2dcfe4d75_hey2.csv",
    "created": "2020-09-20T12:07:48.727529",
    "url": "",
    "mimetype_inner": null,
    "position": 3,
    "revision_id": "17f810d6-aff5-4f0a-adb1-d4499c2262c3",
    "resource_type": null
}
}