Meridian Explorer uses the full-text indexing feature of the database where the repository is stored to index the text of documents, memo properties, and file properties. The indexes are generated from the files when documents are imported into the repository. As new revisions of the documents are imported into the repository, the indexes should be updated to contain the latest content.
Note The performance of full-text searches in repositories hosted in Oracle can be improved by creating a setting named MRE3203 and setting it to true in the WebConfigDto.dat file on the Meridian Enterprise Server computer as described in Troubleshooting full-text indexing.
For information about managing the property indexes for general, non-content searches, see Configuring item names and indexing.
To configure full-text indexing:
Option | Description |
---|---|
Regenerate items that failed before |
By default, if full-text indexing fails for a document, that document will be skipped during the next index update. Enable this option to retry indexing the document during each update. |
Excluded file extensions |
Type a semicolon-delimited list of file extensions to exclude from full-text indexing. These should be file extensions for which IFilters are not installed on the server. By default, all file extensions that are registered for an installed IFilter are included. To view the list of installed IFilters, see Viewing the full-text filters. |
Timeout in seconds |
The amount of time in seconds to wait for index updates to complete before an error is entered in the event logs and processing is stopped. Note The indexing process can stop responding if it encounters a problem in the source file. |
To generate the full-text indexes immediately:
To schedule full-text indexing: