Home

Awesome

Eclector: A portable and extensible Common Lisp Reader

Introduction

The eclector system provides a portable implementation of a reader following the Common Lisp specification.

eclector is under active development. Its ASDF system structure, package structure, exported symbols and protocols may change at any time but are becoming less and less likely to do so in incompatible ways. Consult the NEWS file or the "Changelog" section of the manual for lists of changes in specific versions.

This document only gives a very brief overview and highlights some features. Proper documentation can be found in the documentation directory.

Usage Overview and Highlights

Basics

In the simplest case, the eclector reader can be used like any Common Lisp reader:

Error Recovery

In contrast to many other reader implementations, eclector can recover from most errors in the input supplied to it and continue reading. This capability is realized as a restart named eclector.reader:recover which is established whenever an error is signaled for which a recovery strategy is available.

For example, the following code

(handler-bind ((error (lambda (condition)
                        (let ((restart (find-restart 'eclector.reader:recover)))
                          (format t "Recovering from error:~%~2@T~A~%using~%~2@T~A~%"
                                  condition restart))
                        (eclector.reader:recover))))
  (eclector.reader:read-from-string "`(::foo ,"))

produces this:

Recovering from error:
  A symbol token must not start with two package markers as in ::name.
using
  Treat the character as if it had been escaped.
Recovering from error:
  While reading unquote, expected an object when input ended.
using
  Use NIL in place of the missing object.
Recovering from error:
  While reading list, expected the character ) when input ended.
using
  Return a list of the already read elements.
; => (ECLECTOR.READER:QUASIQUOTE (:FOO (ECLECTOR.READER:UNQUOTE NIL))) 9

indicating that eclector recovered from multiple errors and consumed all input. Of course, the returned expression is likely unsuitable for evaluation, but recovery is useful for detecting multiple errors in one go and performing further processing such as static analysis.

Custom Parse Results

Using features provided in the eclector.parse-result package, the reader can produce parse results controlled by the client, optionally including source tracking and representation of skipped input (due to e.g. comments and reader conditionals):

(defclass my-client (eclector.parse-result:parse-result-client)
  ())

(defmethod eclector.parse-result:make-expression-result
    ((client my-client) (result t) (children t) (source t))
  (list :result result :source source :children children))

(defmethod eclector.parse-result:make-skipped-input-result
    ((client my-client) (stream t) (reason t) (children t) (source t))
  (list :reason reason :source source :children children))

(with-input-from-string (stream "(1 #|comment|# \"string\")")
  (eclector.parse-result:read (make-instance 'my-client) stream))

Concrete Syntax Trees

The eclector.concrete-syntax-tree system provides a variant of the eclector reader that produces instances of the concrete syntax tree classes provided by the concrete syntax tree library:

(eclector.concrete-syntax-tree:read-from-string "(1 2 3)")
; => #<CONCRETE-SYNTAX-TREE:CONS-CST raw: (1 2 3) {100BF94EF3}> 7 NIL