tabledriven

Table-driven, sometimes written table-driven or table driven, refers to a programming approach in which a program's behavior, control flow, or decision-making is governed by data stored in tables rather than by hard-coded conditional logic. A lookup table associates inputs or states with corresponding actions, next states, or outputs, allowing a single implementation to handle many cases by consulting the table.

Common applications include table-driven parsers and state machines. In table-driven parsing, a parsing table indicates actions such as shift, reduce, accept, or error for combinations of states and input tokens. In state machines, a transition table maps each current state and input event to a next state and an optional action. Beyond parsing and state control, table-driven techniques are used for dispatching functions (mapping an input key to a handler function), configuring behavior, and organizing test cases or parameter variations in data sets.

Advantages of the approach include a clear separation between code and data, easier updates by editing tables rather than logic, and compact representations of complex decision logic. It can also simplify localization or reconfiguration by changing data rather than code. Drawbacks include potential readability challenges for large tables, higher risk of data-entry errors, maintenance overhead to ensure table correctness, and possible performance concerns if lookups are not optimized.

The term is common in computer science and software engineering and is often seen in the context of parsing, state machines, and data-driven testing. Related concepts include table-driven parsing and data-driven testing.