macrostep
is an Emacs minor mode for interactively stepping
through the expansion of macros in Emacs Lisp source code. It lets
you see exactly what happens at each step of the expansion process
by pretty-printing the expanded forms inline in the source buffer,
which is temporarily read-only while macro expansions are visible.
You can expand and collapse macro forms one step at a time, and
evaluate or instrument the expansions for debugging with Edebug as
normal (but see “Bugs and known limitations”, below).
Single-stepping through the expansion is particularly useful for
debugging macros that expand into another macro form. These can be
difficult to debug with Emacs’ built-in macroexpand
, which
continues expansion until the top-level form is no longer a macro
call.
Both globally-visible macros as defined by defmacro
and local
macros bound by (cl-)macrolet
or another macro-defining form can
be expanded. Within macro expansions, calls to macros and compiler
macros are fontified specially: macro forms using
macrostep-macro-face
, and functions with compiler macros using
macrostep-compiler-macro-face
. Uninterned symbols (gensyms) are
fontified based on which step in the expansion created them, to
distinguish them both from normal symbols and from other gensyms
with the same print name.
As of version 0.9, it is also possible to extend macrostep
to
work with other languages with macro systems in addition to Emacs
Lisp. An extension for Common Lisp (via SLIME) is in the works;
contributions for other languages are welcome. See “Extending
macrostep” below for details.
The standard keybindings in macrostep-mode
are the following:
- e, =, RET
- expand the macro form following point one step
- c, u, DEL
- collapse the form following point
- q, C-c C-c
- collapse all expanded forms and exit macrostep-mode
- n, TAB
- jump to the next macro form in the expansion
- p, M-TAB
- jump to the previous macro form in the expansion
It’s not very useful to enable and disable macrostep-mode
directly. Instead, bind macrostep-expand
to a key in
emacs-lisp-mode-map
, for example C-c e:
(define-key emacs-lisp-mode-map (kbd "C-c e") 'macrostep-expand)
You can then enter macrostep-mode and expand a macro form
completely by typing C-c e e e ...
as many times as necessary.
Exit macrostep-mode by typing q
or C-c C-c
, or by successively
typing c
to collapse all surrounding expansions.
Type M-x customize-group RET macrostep RET
to customize options
and faces.
To display macro expansions in a separate window, instead of inline
in the source buffer, customize
macrostep-expand-in-separate-buffer
to t
. The default is
nil
. Whichever default behavior is selected, the alternative
behavior can be obtained temporarily by giving a prefix argument to
macrostep-expand
.
To have macrostep
ignore compiler macros, customize
macrostep-expand-compiler-macros
to nil
. The default is t
.
Customize the faces macrostep-macro-face
,
macrostep-compiler-macro-face
, and macrostep-gensym-1
through
macrostep-gensym-5
to alter the appearance of macro expansions.
As of version 0.9, macrostep
can expand calls to a locally-bound
macro, whether defined by a surrounding (cl-)macrolet
form, or by
another macro-defining macro. In other words, it is possible to
expand the inner local-macro
forms in both the following
examples, whether local-macro
is defined by an enclosing
cl-macrolet
–
(cl-macrolet ((local-macro (&rest args)
`(expansion of ,args)))
(local-macro (do-something)))
– or by a macro which expands into cl-macrolet
, provided that
its definition of macro is evaluated prior to calling
macrostep-expand
:
(defmacro with-local-macro (&rest body)
`(cl-macrolet ((local-macro (&rest args)
`(expansion of ,args)))
,@body))
(with-local-macro
(local-macro (do something (else)))
See the with-js
macro in Emacs’s js.el
for a real example of
the latter kind of macro.
Expansion of locally-bound macros is implemented by instrumenting Emacs Lisp’s macro-expander to capture the environment at point. A similar trick is used to detect macro- and compiler-macro calls within expanded text so that they can be fontified accurately.
By moving point around in the macro expansion using
macrostep-next-macro
and macrostep-prev-macro
(bound to the n
and p
keys), it is possible to expand other macro calls within
the expansion before expanding the outermost form. This can
sometimes be useful, although it does not correspond to the real
order of macro expansion in Emacs Lisp, which proceeds by fully
expanding the outer form to a non-macro form before expanding
sub-forms.
The main reason to expand sub-forms out of order is to help with
debugging macros which programmatically expand their arguments in
order to rewrite them. Expanding the arguments of such a macro
lets you visualise what the macro definition would compute via
macroexpand-all
.
Since version 0.9, it is possible to extend macrostep to work with other languages besides Emacs Lisp. In typical Emacs fashion, this is implemented by setting buffer-local variables to different function values. Six buffer-local variables define the language-specific part of the implementation:
macrostep-sexp-bounds-function
macrostep-sexp-at-point-function
macrostep-environment-at-point-function
macrostep-expand-1-function
macrostep-print-function
macrostep-macro-form-p-function
Typically, an implementation for another language would set these
variables in a major-mode hook. See the docstrings of each
variable for details on how each one is called and what it should
return. At a minimum, another language implementation needs to
provide macrostep-sexp-at-point-function
,
macrostep-expand-1-function
, and macrostep-print-function
.
Lisp-like languages may be able to reuse the default
macrostep-sexp-bounds-function
if they provide another
implementation of macrostep-macro-form-p-function
. Languages
which do not implement locally-defined macros can set
macrostep-environment-at-point-function
to ignore
.
Note that the core macrostep
machinery only interprets the return
value of macrostep-sexp-bounds-function
, so implementations for
other languages can use any internal representations of code and
environments which is convenient. Although the terminology is
Lisp-specific, there is no reason that implementations could not be
provided for non-Lisp languages with macro systems, provided there
is some way of identifying macro calls and calling the compiler /
preprocessor to obtain their expansions.
You can evaluate and edebug macro-expanded forms and step through
the macro-expanded version, but the form that eval-defun
and
friends read from the buffer won’t have the uninterned symbols of
the real macro expansion. This will probably work OK with CL-style
gensyms, but may cause problems with make-symbol
symbols if they
have the same print name as another symbol in the expansion. It’s
possible that using print-circle
and print-gensym
could get
around this.
Please send other bug reports and feature requests to the author.
Thanks to:
- John Wiegley for fixing a bug with the face definitions under Emacs 24 & for plugging macrostep in his EmacsConf presentation!
- George Kettleborough for bug reports, and patches to highlight the expanded region and properly handle backquotes.
- Nic Ferrier for suggesting support for local definitions within macrolet forms
- Luís Oliveira for suggesting and implementing SLIME support
macrostep
was originally inspired by J. V. Toups’s ‘Deep Emacs
Lisp’ articles (part 1, part 2, screencast).
- v0.9, 2015-10-01:
- separate into Elisp-specific and generic components
- highlight and expand compiler macros
- improve local macro expansion and macro form identification by
instrumenting
macroexpand(-all)
- v0.8, 2014-05-29: fix a bug with printing the first element of lists
- v0.7, 2014-05-11: expand locally-defined macros within
(cl-)macrolet
forms - v0.6, 2013-05-04: better handling of quote and backquote
- v0.5, 2013-04-16: highlight region, maintain cleaner buffer state
- v0.4, 2013-04-07: only enter macrostep-mode on successful macro-expansion
- v0.3, 2012-10-30: print dotted lists correctly. autoload definitions.
Please note that as this package will be part of Emacs soon, non-trivial code contributions require FSF copyright assignment prior to acceptance.
The copyright assignment policy is at https://www.fsf.org/licensing/contributor-faq
Any legally significant contributions can only be merged after the author has completed their paperwork. https://www.gnu.org/prep/maintain/html_node/Legally-Significant.html#Legally-Significant