Import-Export
gdnsl import
Import collection data.
gdnsl import COLLECTION-NAME [flags]
Examples:
# Import documents in the "addresses" collection
gdnsl import addresses --json "[{\"name\": \"John\"}, {\"lname\": \"Doe\"}]"
# Import documents in the "addresses" collection with primaryKey as "name"
gdnsl import addresses --json "[{\"name\": \"John\"}, {\"lname\": \"Doe\"}]" --primary-key "name"
# Import documents with existing document having same _key in the colletion, shall be replaced
gdnsl import addresses --json "[{\"name\": \"John\"}, {\"lname\": \"Doe\"}]" --replace
# Import documents with extra information for errors and unprocessed documents returned in the result
gdnsl import addresses --json "[{\"name\": \"John\"}, {\"lname\": \"Doe\"}]" --details
# Import documents in the "addresses" collection
gdnsl import addresses --csv "John,Doe\nJane,Doe" --columns "name,lname"
# Import documents in the "addresses" collection with primaryKey as "name"
gdnsl import addresses --csv "John,Doe\nJane,Doe" --columns "name,lname" --primary-key "name"
# Import documents with existing document having same _key in the colletion, shall be replaced
gdnsl import addresses --csv "John,Doe\nJane,Doe" --columns "name,lname" --replace
# Import documents with extra information for errors and unprocessed documents returned in the result
gdnsl import addresses --csv "John,Doe\nJane,Doe" --columns "name,lname" --details
# Import documents from a JSON file
gdnsl import --file "path_to_file/import.json"
# Import documents from a CSV file
gdnsl import --file "path_to_file/import.csv"
# Import documents in the "addresses" collection with a batch size of 1000
gdnsl import addresses --file "import.json" --batch-size=1000
Options:
-h, --help Help for query import command.
--json string Should be an array of json documents. Each document is interpreted separately. For large data use --file instead.
--csv string Should be an array of csv rows. Each row is separated by a newline character. For large data use --file instead.
--columns String should be a comma separated list of column names.
--primary-key string If specified, this attribure will be used as _key of the new document. It must follow the [naming conventions](../naming-conventions.md#document-keys). If document already contains _key then it will be renamed as old_key.
--replace If true existing document having same _key in the colletion, shall be replaced.
--details If true extra information for errors and unprocessed documents will be returned in the result.
--batch-size Number of docs to be imported in a batch. Cannot be greater than the set tenant limit.
Default will the tenant limit. Works with the --file flag.
--file string Path to JSON or CSV file. File format is inferred from the file extension.
--fabric Name of the fabric to use.
Options inherited:
--config string gdnsl config file (default is ./gdnsl.yaml)
gdnsl export
Export data from a collection or via a query.
gdnsl export [flags]
Examples:
# Export data returned by the query
gdnsl export --query "FOR doc IN addresses RETURN doc"
# Export data returned by the query with based on the given filter
gdnsl export --query "FOR doc IN addresses FILTER doc.country == @country RETURN doc" --param "country=USA"
# Export data returned by the query in CSV format
gdnsl export --query "FOR doc IN addresses RETURN doc" --csv
# Export data returned by the query in JSON format to a file
gdnsl export --query "FOR doc IN addresses RETURN doc" --file test.json
# Export data from "addresses" collection
gdnsl export --collection addresses
# Export data from "addresses" collection with offset as 10
gdnsl export --collection addresses --offset 10
# Export data from "addresses" collection with limit as 10
gdnsl export --collection addresses --limit 10
# Export data from "addresses" collection in descending order
gdnsl export --collection addresses --order desc
# Export data from "addresses" collection in CSV format
gdnsl export --collection addresses --csv
# Export data from "addresses" collection in CSV format
gdnsl export --collection addresses --file test.csv
Options:
-h, --help Help for service.
--query string Query string to execute and export data.
--param string Params to be given to the query. Can be given multiple times.
--csv string To export in csv format. Default is "json".
--collection string Collection name to export data from.
--offset number This option can be used to simulate paging. Default: 0.
--limit number This option can be used to simulate paging. Limit the result. Default: 20, Max: 1000.
--order string Order the results asc or desc. Default: asc.
--file string Path to JSON or CSV file. File format is inferred from the file extension.
--fabric Name of the fabric to use.
Options inherited:
--config string gdnsl config file (default is ./gdnsl.yaml)