Single assignment

Single assignment is used to describe a programming language or representation in which one cannot bind a value to a name if a value has already been bound to that name. In other words, a variable is "initialized" with (i.e. "bound" to) a value at the time that it is created, and it cannot be assigned to. It prevents some types of side effects, which is claimed to reduce software bugs and simplify debugging.

Note: in some cases partial values could be used in more than one assignment as far as the partial values used in these assignments are compatible. Sometimes such variables are called declarative variables or dataflow variables. For further reading see CTM.

Implementations

Single assignment is used in several compilers and languages.

Compilers

Static single assignment form, SSA form, SSA, compiler representation

Languages

*Functional (these use single assignment to reduce side effects)
**Erlang
**Haskell
**SASL
**SISAL
**Oz
**OCaml - in OCaml, "variables" can only be single-assigned; however, members of arrays and mutable members of records and objects can be assigned to regularly
*Object-oriented
**Lava
**Oz

Some languages also support single assignment as an option.

*C# uses the "readonly" keyword next to a variable declaration.
*Java uses the "final" keyword next to a variable declaration.
*C++ uses the "const" keyword next to a variable declaration (only when the const modifies the type; not for "pointer-to-const" things).
*Scala when declaring variables with "val" instead of "var".

The use of these features, is however generally restricted to OO design decisions, such as making an object immutable.

ee also

* Immutable object
* Const-correctness