Skip to main content
Version: Next

CSV

Incubating

This plugin is used to bulk upload metadata to Datahub. It will apply glossary terms, tags, decription, owners and domain at the entity level. It can also be used to apply tags, glossary terms, and documentation at the column level. These values are read from a CSV file. You have the option to either overwrite or append existing values.

The format of the CSV is demonstrated below. The header is required and URNs should be surrounded by quotes when they contains commas (most URNs contains commas).

resource,subresource,glossary_terms,tags,owners,ownership_type,description,domain
"urn:li:dataset:(urn:li:dataPlatform:snowflake,datahub.growth.users,PROD)",,[urn:li:glossaryTerm:Users],[urn:li:tag:HighQuality],[urn:li:corpuser:lfoe|urn:li:corpuser:jdoe],TECHNICAL_OWNER,"description for users table",urn:li:domain:Engineering
"urn:li:dataset:(urn:li:dataPlatform:hive,datahub.growth.users,PROD)",first_name,[urn:li:glossaryTerm:FirstName],,,,"first_name description"
"urn:li:dataset:(urn:li:dataPlatform:hive,datahub.growth.users,PROD)",last_name,[urn:li:glossaryTerm:LastName],,,,"last_name description"

Note that the first row does not have a subresource populated. That means any glossary terms, tags, and owners will be applied at the entity field. If a subresource is populated (as it is for the second and third rows), glossary terms and tags will be applied on the column. Every row MUST have a resource. Also note that owners can only be applied at the resource level.

note

This source will not work on very large csv files that do not fit in memory.

CLI based Ingestion

Install the Plugin

The csv-enricher source works out of the box with acryl-datahub.

Starter Recipe

Check out the following recipe to get started with ingestion! See below for full configuration options.

For general pointers on writing and running a recipe, see our main recipe guide.

source:
type: csv-enricher
config:
# relative path to your csv file to ingest
filename: ./path/to/your/file.csv

# Default sink is datahub-rest and doesn't need to be configured
# See https://datahubproject.io/docs/metadata-ingestion/sink_docs/datahub for customization options

Config Details

Note that a . is used to denote nested fields in the YAML recipe.

FieldDescription
filename 
string
File path or URL of CSV file to ingest.
array_delimiter
string
Delimiter to use when parsing array fields (tags, terms and owners)
Default: |
delimiter
string
Delimiter to use when parsing CSV
Default: ,
write_semantics
string
Whether the new tags, terms and owners to be added will override the existing ones added only by this source or not. Value for this config can be "PATCH" or "OVERRIDE"
Default: PATCH

Code Coordinates

  • Class Name: datahub.ingestion.source.csv_enricher.CSVEnricherSource
  • Browse on GitHub

Questions

If you've got any questions on configuring ingestion for CSV, feel free to ping us on our Slack.