Document Retrieval

The Document Retrieval API allows sending a query to LucidWorks Big Data and getting the matching documents in a JSON-formatted response. More information about searching the LucidWorks Big Data document indexes is available at Search.

This section describes general document retrieval, and also how to gain access to the various transformations that are possible with the ETL Workflow.

API Entry Points
Send a Query
- Input
- Output
- Examples
Request Clusters
- Input
- Output
- Examples
Request Statistically Interesting Phrases
- Input
- Output
- Examples
Request Document Similarities
- Input
- Output
- Examples
Request Specific Entities
- Input
- Output
- Examples

API Entry Points

/sda/v1/client/collections/collection/documents/retrieval: Send a query and see matching documents

Send a Query

POST /sda/v1/client/collections/collection/documents/retrieval

Input

Path Parameters

Key	Description
collection	The collection name.

Query Parameters

Key	Type	Required	Default	Description
query	JSON map	Yes	null	A valid query, expressed in key-value pairs.

Output

Output Content

The output of a query is a representation of the Solr response. Under the response, the following attributes will likely be returned:

Key	Type	Description
docs	JSON map	This includes the documents returned by the query, which, in the case of retrieval by ID, will likely only contain a single document. The fields in the `docs` response include the `body` of the document, the associated `data_source_name`, `data_source_type` and `data_source` ID, the `title` of the document, and any other fields from the document stored in the index.
maxScore	string	This is the maximum score of any document in the response. Since this is returning a single document, the `maxScore` will match the `score` found under the `docs` part of the response
numFound	integer	The number of documents found by the query.
start
responseHeader	JSON map	The responseHeader will include information on the query, output format, output settings (such as highlighting, etc.) and the time to complete the query.

Examples

Input
Retrieve all documents from the "documentation" collection.

Output