dp-import

module
v1.4.0 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: May 2, 2023 License: MIT

README

dp-import

Getting started
Required dependencies
Kafka
Neo4j
  • Run brew install neo4j
  • Disable authentication by editing /usr/local/Cellar/neo4j/3.2.0/libexec/conf/neo4j.conf
    • Set dbms.security.auth_enabled=false
  • Run brew services restart neo4j
MongoDB
  • Run brew install mongodb
  • Run brew services start mongodb
Import process

See the import sequence documentation.

Running the services

There are a number of services that provide the import process functionality. Consider using websysd that will run all of the services for you. If you do want to run services independently:

Import frontend
Import backend
Import a dataset

Ensure a dataset exists on the dataset API for the dataset specified in the recipe being used. The current stubbed recipe api specifies dataset ID cpi01 for the CPI dataset recipe.

curl --header 'internal-token:FD0108EA-825D-411C-9B1D-41EF7727F465' http://localhost:22000/datasets/cpih01

If not then create one (already set to published):

curl -X POST -d '{"release_frequency":"yearly", "state": "published", "theme": "population", "title": "CPI" }' --header 'internal-token:FD0108EA-825D-411C-9B1D-41EF7727F465' http://localhost:22000/datasets/cpih01

Ensure you have imported a full hierarchy for any hierarchies used in the dataset: https://github.com/ONSdigital/dp-hierarchy-builder

Create instance

Navigate to http://localhost:8081/florence/datasets

  • upload a file
  • select an edition
  • click submit to publishing

Get instance data from the import API - the instance state should be 'completed' if the import succeeded (copy the instance ID - it will be the last instance in the array):

Example curl command to GET instances:

curl --header 'internal-token:FD0108EA-825D-411C-9B1D-41EF7727F465' http://localhost:22000/instances | jq
Set the release date value on the instance (replace the instance ID with the one you created).

API call details:

PUT localhost:22000/instances/750102f4-2839-441f-b2e4-6cf99d26858a
{
	"release_date": "todayisfine"
}

Example curl command to PUT instance data:

curl -v -X PUT -d '{"release_date":"today"}' --header 'internal-token:FD0108EA-825D-411C-9B1D-41EF7727F465' localhost:22000/instances/750102f4-2839-441f-b2e4-6cf99d26858a
Set the instance to 'edition-confirmed' (replace the instance ID with the one you created)

This converts from an instance to a dataset version.

API call details:

PUT localhost:22000/instances/750102f4-2839-441f-b2e4-6cf99d26858a
{
    "edition":"Time-series",
    "state": "edition-confirmed"
}

Example curl command to PUT instance state:

curl -v -X PUT -d '{"state":"edition-confirmed", "edition":"Time-series"}' --header 'internal-token:FD0108EA-825D-411C-9B1D-41EF7727F465' localhost:22000/instances/750102f4-2839-441f-b2e4-6cf99d26858a
Notes

Once the instance state has been successfully updated to edition-confirmed then the edition resource is created if it does not already exist.

It is possible to combine the previous two requests into one or seperate request to update edition and state into two requests. The above request will fail if the edition is not set as the API does not know what edition this instance is related to.

Associate the dataset with a collection

You will first need to get the dataset version URL from the instances endpoint

Example curl command to GET instances:

curl --header 'internal-token:FD0108EA-825D-411C-9B1D-41EF7727F465' http://localhost:22000/instances | jq

Copy the link for the version URL to use in the following calls.

API call details:

PUT http://localhost:22000/datasets/cpih01/editions/Time-series/versions/1
{
	"collection_id": "1234",
	"state": "associated"
}

Example curl command to PUT dataset version state / collection:

curl -v -X PUT -d '{"state":"associated", "collection_id":"123"}' --header 'internal-token:FD0108EA-825D-411C-9B1D-41EF7727F465' http://localhost:22000/datasets/cpih01/editions/Time-series/versions/1
Notes

Once the version has been updated then the dataset resource is also updated, this includes the following next.collection_id, next.state and next.links are all updated.

It is possible to add a collection_id before changing the state to associated with two separate requests. If the collection_id is not set and a request is made to change the state to associated then the resource will fail to update.

Set version and hence dataset to published

API call details:

PUT http://localhost:22000/datasets/cpih01/editions/Time-series/versions/1
{
	"state": "published"
}

Example curl command to PUT dataset version state:

curl -v -X PUT -d '{"state":"published"}' --header 'internal-token:FD0108EA-825D-411C-9B1D-41EF7727F465' http://localhost:22000/datasets/cpih01/editions/Time-series/versions/1
Notes

Once the version has been updated with a state of published, then the edition is updated; this includes a change of state from created to published if the resource was not already published from a previous version and the links.latest_version is updated. Once the edition resource is successfully updated then the dataset resource is updated with a new state published on the next sub document and the next sub document is copied over to the current sub document.

Check dataset is available via the frontend

You should now see the details of the dataset via the frontend by navigating to http://localhost:20000/datasets/cpih01

Admin
Clear MongoDB

The following will drop the datasets and imports databases. Add further databases as required.

 mongo mongodb://localhost:27017 <<EOF
 use datasets
 db.dropDatabase();
 use imports
 db.dropDatabase();
EOF
Clear neo4j
brew services stop neo4j
rm -rf /usr/local/Cellar/neo4j/3.2.0/libexec/data
brew services start neo4j
clear kafka topic
  • enable kafka topic deletions: edit /usr/local/etc/kafka/server.properties and set delete.topic.enable=true

kafka-topics --delete --zookeeper localhost:2181 --topic observation-extracted

Create topic with partition / replication properties

kafka-topics --create --zookeeper localhost:2181 --topic dimensions-inserted --partitions 1 --replication-factor 1

Contributing

See CONTRIBUTING for details.

License

Copyright © 2016-2017, Office for National Statistics (https://www.ons.gov.uk)

Released under MIT license, see LICENSE for details.

Directories

Path Synopsis

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.