SearchBlox versions 8.2 and later are pre-installed with Advanced Integration [ManifoldCF]. It can be accessed from http://localhost:8080/mcf-combined-service/login.jsp.
The default credentials are
username : admin
password : admin
These can be changed in the ManifoldCF properties file.
Start SearchBlox and create a custom collection. This collection should be linked with ManifoldCF and the files will be indexed and searched from this collection.
Configuring MCF for file system indexing involves the following steps:
1. Creating an Output Connector
An output connection is a connection to SearchBlox where documents fetched from various repositories can be written. Create an output connection by clicking List Output Connections in the left-hand navigation menu and click Add a new output connection to create a new output connector.
Enter a name for the output connector and move to the Type tab. Select the Connection type as “Searchblox”. Click the Continue button.
After clicking the Continue button, two more tabs will appear. Click SearchBlox Settings.
Specify the API key from the SearchBlox admin page. When finished, click the Save button.
2. Creating a Transformation Connection
A transformation connection is an engine connection where documents fetched from various repositories can be manipulated. This typically involves metadata extraction or mapping. Click on List Transformation Connections in the menu, and click the Add a new transformation connection button to create a new transformation connector.
Enter any name that is relevant for the transformation connection. Then move to the next tab, Type.
From the Type tab, select the Connection type as “Tika content extractor ” and click the Continue button, and then click Save. This returns to List Transformation Connections, where the transformation connector created will be available.
3. Creating an Authority Group
An authority group is a collection of authorities that all cooperate to furnish security for each document from selected repositories.
Click on List authority groups from the left-hand side navigation menu, and select Add a new authority group.
Provide a name and description and click on the Save button.
4. Creating an Authority Connection
A CMIS authority connection is required to enforce security for documents retrieved from CMIS repositories. Click on the List of Authority Connections from the menu and click the Add a new connection button to create a new output connector.
Enter a name and move to the Type tab, and select the Connection type as “CMIS”. Then select the Authority group that has been previously created.
The repository configuration will only be used to track an ID for a specific CMIS repository. No calls will be performed against the CMIS repository.
Click the Save button. A summary and status for the authority connection will be shown:
5. Create a Repository Connection:
A repository connection is a connection to the repository system that contains the documents to be indexed.
Click on List Repository Connections from the menu, and click the Add a new connection button to create a new repository connector.
Enter a name and move to the Type tab. Select the Connection type as “CMIS” and select the Authority group previously created.
After clicking the Continue button, two more tabs will navigate to the Server tab. Enter the server details and click Save. For server details, contact the system administrator.
Click on Save, and the following connection status screen will appear:
6. Creating a Job
Click on List all Jobs from the menu, and click the Add a new job button.
Enter the name and click the Connection tab.
Select the Connection Name and Output from the select box and click Add output. Then select the transformation and click Insert transformation before.
The CMIS Query tab allows a query to be specified based on the CMIS Query Language to retrieve all of the result documents that need to be ingested.
For each result during the ingestion process, if the CMIS connecter finds a folder node (that must have cmis:folder as the base type), it will ingest all the children of the folder node; otherwise it will directly ingest the document (that must have cmis:document as the base type).
Enter the SearchBlox collection name in the SearchBlox settings tab.
Click the Save button to see a summary page that looks similar to this:
Navigate to the Status and Job Management menu from the left side. The jobs created will be displayed here. Start the job by clicking the Start button.
Once the job execution has been finished, documents will be updated in the SearchBlox server.