Overview | File Types | Environments | Queries | Updates | Generating Data Files | Integration Guides
Omnidex allows raw data files to be updated with some limitations; however, the large majority of Omnidex applications simply replace some or all of the data files and reload the Omnidex indexes. Raw data files are usually used for Omnidex Snapshots, data warehouses, data mines, transaction logs, or regularly-replaced data files. In these situations, it is easiest and quickest to simply replace the data and update the indexes.
Some applications may wish to use the SQL INSERT, UPDATE and DELETE statements to update raw data files. Different types of raw data files have different support for updates. These capabilities are shown in the table below:
| File type | INSERT | UPDATE | DELETE |
|---|---|---|---|
| Fixed-length data files | Yes | Yes | No |
| Delimited files | Yes | No | No |
| Omnidex Standalone Tables | Yes | Yes | Yes |
Like most NoSQL databases, Omnidex does not support transaction management. Each update operation is treated as a transaction of its own, and it is not possible to group multiple update operations into a single transaction.
Referential constraints are often declared in Omnidex Environment Files, and these are used by Omnidex when optimizing queries. Omnidex does not enforce these constraints during update operations, and so the application or the administrator must insure referential integrity exists between tables.
See also: