Because relational databases are oriented toward sets, no concept of next row exists, meaning that you cannot operate on an individual row in a set. Cursor functionality solves this problem by letting a result set be processed one row at a time, similar to the way you read and update a file on a disk. A DB-Library cursor indicates the current position in a result set, just as the cursor on your screen indicates the current position in a block of text.
DB-Library cursors are client-side cursors. This means that they do not correspond to an Adaptive Server cursor, but emulate a cursor that appears to the user to be in the server. The DB-Library cursor transparently does keyset management, row positioning, and concurrency control entirely on the client side.
