Read-eval-print loops

The user usually interacts with the T system via a read-eval-print loop. As illustrated above, this is a command loop which repeatedly reads an expression, evaluates it, and prints the value.

The read-eval-print loop is a simple command processor; it prints a prompt, reads a command from the terminal, executes the command, then prints another prompt, ad infinitum. A command is any executable T form, and executing the command consists of evaluating the form and printing the result.

$\begin{inset}{} The variable {\tt\char93 \char93 } always has as its value the l... ...t loop, that is, the value of the last expression typed by the user. \end{inset}$

$\begin{inset}{} {\bf Removed in {\bf T3}.} The variable {\tt ++} always has as i... ...n evaluated, so that a new expression may refer to the previous one. \end{inset}$

Expressions are read from the terminal by applying the READ operation (actually, the value of (REPL-READ); see page ) to the terminal input port. Initially, that port's read table is the standard read table, but this may be changed using SET, for example:
$\begin{codexenv} (SET (PORT-READ-TABLE (TERMINAL-INPUT)) *MY-READ-TABLE*) \end{codexenv}$

Evaluation is performed with respect to a particular variable environment and its syntax table. The read-eval-print loop may move from place to place within the environment hierarchy; one may control the evaluation environment and syntax table by setting (REPL-ENV).

Settable

$\begin{inset}{} Accesses the environment passed to {\tt (REPL-EVAL)} by the read... ...ally, this is the scratch environment (see page \pageref{USER-ENV}). \end{inset}$