mirror of
https://git.savannah.gnu.org/git/emacs.git
synced 2024-12-22 10:26:20 +00:00
3952 lines
166 KiB
EmacsLisp
3952 lines
166 KiB
EmacsLisp
;;; advice.el --- an overloading mechanism for Emacs Lisp functions
|
|
|
|
;; Copyright (C) 1993, 1994, 2000, 2001, 2002, 2003, 2004,
|
|
;; 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
|
|
|
|
;; Author: Hans Chalupsky <hans@cs.buffalo.edu>
|
|
;; Maintainer: FSF
|
|
;; Created: 12 Dec 1992
|
|
;; Keywords: extensions, lisp, tools
|
|
|
|
;; This file is part of GNU Emacs.
|
|
|
|
;; GNU Emacs is free software: you can redistribute it and/or modify
|
|
;; it under the terms of the GNU General Public License as published by
|
|
;; the Free Software Foundation, either version 3 of the License, or
|
|
;; (at your option) any later version.
|
|
|
|
;; GNU Emacs is distributed in the hope that it will be useful,
|
|
;; but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
;; GNU General Public License for more details.
|
|
|
|
;; You should have received a copy of the GNU General Public License
|
|
;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
;; LCD Archive Entry:
|
|
;; advice|Hans Chalupsky|hans@cs.buffalo.edu|
|
|
;; Overloading mechanism for Emacs Lisp functions|
|
|
;; 1994/08/05 03:42:04|2.14|~/packages/advice.el.Z|
|
|
|
|
|
|
;;; Commentary:
|
|
|
|
;; NOTE: This documentation is slightly out of date. In particular, all the
|
|
;; references to Emacs-18 are obsolete now, because it is not any longer
|
|
;; supported by this version of Advice.
|
|
|
|
;; Advice is documented in the Emacs Lisp Manual.
|
|
|
|
;; @ Introduction:
|
|
;; ===============
|
|
;; This package implements a full-fledged Lisp-style advice mechanism
|
|
;; for Emacs Lisp. Advice is a clean and efficient way to modify the
|
|
;; behavior of Emacs Lisp functions without having to keep personal
|
|
;; modified copies of such functions around. A great number of such
|
|
;; modifications can be achieved by treating the original function as a
|
|
;; black box and specifying a different execution environment for it
|
|
;; with a piece of advice. Think of a piece of advice as a kind of fancy
|
|
;; hook that you can attach to any function/macro/subr.
|
|
|
|
;; @ Highlights:
|
|
;; =============
|
|
;; - Clean definition of multiple, named before/around/after advices
|
|
;; for functions, macros, subrs and special forms
|
|
;; - Full control over the arguments an advised function will receive,
|
|
;; the binding environment in which it will be executed, as well as the
|
|
;; value it will return.
|
|
;; - Allows re/definition of interactive behavior for functions and subrs
|
|
;; - Every piece of advice can have its documentation string which will be
|
|
;; combined with the original documentation of the advised function at
|
|
;; call-time of `documentation' for proper command-key substitution.
|
|
;; - The execution of every piece of advice can be protected against error
|
|
;; and non-local exits in preceding code or advices.
|
|
;; - Simple argument access either by name, or, more portable but as
|
|
;; efficient, via access macros
|
|
;; - Allows the specification of a different argument list for the advised
|
|
;; version of a function.
|
|
;; - Advised functions can be byte-compiled either at file-compile time
|
|
;; (see preactivation) or activation time.
|
|
;; - Separation of advice definition and activation
|
|
;; - Forward advice is possible, that is
|
|
;; as yet undefined or autoload functions can be advised without having to
|
|
;; preload the file in which they are defined.
|
|
;; - Forward redefinition is possible because around advice can be used to
|
|
;; completely redefine a function.
|
|
;; - A caching mechanism for advised definition provides for cheap deactivation
|
|
;; and reactivation of advised functions.
|
|
;; - Preactivation allows efficient construction and compilation of advised
|
|
;; definitions at file compile time without giving up the flexibility of
|
|
;; the advice mechanism.
|
|
;; - En/disablement mechanism allows the use of different "views" of advised
|
|
;; functions depending on what pieces of advice are currently en/disabled
|
|
;; - Provides manipulation mechanisms for sets of advised functions via
|
|
;; regular expressions that match advice names
|
|
|
|
;; @ How to get Advice for Emacs-18:
|
|
;; =================================
|
|
;; `advice18.el', a version of Advice that also works in Emacs-18 is available
|
|
;; either via anonymous ftp from `ftp.cs.buffalo.edu (128.205.32.9)' with
|
|
;; pathname `/pub/Emacs/advice18.el', or from one of the Emacs Lisp archive
|
|
;; sites, or send email to <hans@cs.buffalo.edu> and I'll mail it to you.
|
|
|
|
;; @ Overview, or how to read this file:
|
|
;; =====================================
|
|
;; NOTE: This documentation is slightly out of date. In particular, all the
|
|
;; references to Emacs-18 are obsolete now, because it is not any longer
|
|
;; supported by this version of Advice. An up-to-date version will soon be
|
|
;; available as an info file (thanks to the kind help of Jack Vinson and
|
|
;; David M. Smith). Until then you can use `outline-mode' to help you read
|
|
;; this documentation (set `outline-regexp' to `";; @+"').
|
|
;;
|
|
;; The four major sections of this file are:
|
|
;;
|
|
;; @ This initial information ...installation, customization etc.
|
|
;; @ Advice documentation: ...general documentation
|
|
;; @ Foo games: An advice tutorial ...teaches about Advice by example
|
|
;; @ Advice implementation: ...actual code, yeah!!
|
|
;;
|
|
;; The latter three are actual headings which you can search for
|
|
;; directly in case `outline-mode' doesn't work for you.
|
|
|
|
;; @ Restrictions:
|
|
;; ===============
|
|
;; - This version of Advice only works for Emacs 19.26 and later. It uses
|
|
;; new versions of the built-in functions `fset/defalias' which are not
|
|
;; yet available in Lucid Emacs, hence, it won't work there.
|
|
;; - Advised functions/macros/subrs will only exhibit their advised behavior
|
|
;; when they are invoked via their function cell. This means that advice will
|
|
;; not work for the following:
|
|
;; + advised subrs that are called directly from other subrs or C-code
|
|
;; + advised subrs that got replaced with their byte-code during
|
|
;; byte-compilation (e.g., car)
|
|
;; + advised macros which were expanded during byte-compilation before
|
|
;; their advice was activated.
|
|
|
|
;; @ Credits:
|
|
;; ==========
|
|
;; This package is an extension and generalization of packages such as
|
|
;; insert-hooks.el written by Noah S. Friedman, and advise.el written by
|
|
;; Raul J. Acevedo. Some ideas used in here come from these packages,
|
|
;; others come from the various Lisp advice mechanisms I've come across
|
|
;; so far, and a few are simply mine.
|
|
|
|
;; @ Comments, suggestions, bug reports:
|
|
;; =====================================
|
|
;; If you find any bugs, have suggestions for new advice features, find the
|
|
;; documentation wrong, confusing, incomplete, or otherwise unsatisfactory,
|
|
;; have any questions about Advice, or have otherwise enlightening
|
|
;; comments feel free to send me email at <hans@cs.buffalo.edu>.
|
|
|
|
;; @ Safety Rules and Emergency Exits:
|
|
;; ===================================
|
|
;; Before we begin: CAUTION!!
|
|
;; Advice provides you with a lot of rope to hang yourself on very
|
|
;; easily accessible trees, so, here are a few important things you
|
|
;; should know: Once Advice has been started with `ad-start-advice'
|
|
;; (which happens automatically when you load this file), it
|
|
;; generates an advised definition of the `documentation' function, and
|
|
;; it will enable automatic advice activation when functions get defined.
|
|
;; All of this can be undone at any time with `M-x ad-stop-advice'.
|
|
;;
|
|
;; If you experience any strange behavior/errors etc. that you attribute to
|
|
;; Advice or to some ill-advised function do one of the following:
|
|
|
|
;; - M-x ad-deactivate FUNCTION (if you have a definite suspicion what
|
|
;; function gives you problems)
|
|
;; - M-x ad-deactivate-all (if you don't have a clue what's going wrong)
|
|
;; - M-x ad-stop-advice (if you think the problem is related to the
|
|
;; advised functions used by Advice itself)
|
|
;; - M-x ad-recover-normality (for real emergencies)
|
|
;; - If none of the above solves your Advice-related problem go to another
|
|
;; terminal, kill your Emacs process and send me some hate mail.
|
|
|
|
;; The first three measures have restarts, i.e., once you've figured out
|
|
;; the problem you can reactivate advised functions with either `ad-activate',
|
|
;; `ad-activate-all', or `ad-start-advice'. `ad-recover-normality' unadvises
|
|
;; everything so you won't be able to reactivate any advised functions, you'll
|
|
;; have to stick with their standard incarnations for the rest of the session.
|
|
|
|
;; IMPORTANT: With Advice loaded always do `M-x ad-deactivate-all' before
|
|
;; you byte-compile a file, because advised special forms and macros can lead
|
|
;; to unwanted compilation results. When you are done compiling use
|
|
;; `M-x ad-activate-all' to go back to the advised state of all your
|
|
;; advised functions.
|
|
|
|
;; RELAX: Advice is pretty safe even if you are oblivious to the above.
|
|
;; I use it extensively and haven't run into any serious trouble in a long
|
|
;; time. Just wanted you to be warned.
|
|
|
|
;; @ Customization:
|
|
;; ================
|
|
|
|
;; Look at the documentation of `ad-redefinition-action' for possible values
|
|
;; of this variable. Its default value is `warn' which will print a warning
|
|
;; message when an already defined advised function gets redefined with a
|
|
;; new original definition and de/activated.
|
|
|
|
;; Look at the documentation of `ad-default-compilation-action' for possible
|
|
;; values of this variable. Its default value is `maybe' which will compile
|
|
;; advised definitions during activation in case the byte-compiler is already
|
|
;; loaded. Otherwise, it will leave them uncompiled.
|
|
|
|
;; @ Motivation:
|
|
;; =============
|
|
;; Before I go on explaining how advice works, here are four simple examples
|
|
;; how this package can be used. The first three are very useful, the last one
|
|
;; is just a joke:
|
|
|
|
;;(defadvice switch-to-buffer (before existing-buffers-only activate)
|
|
;; "When called interactively switch to existing buffers only, unless
|
|
;;when called with a prefix argument."
|
|
;; (interactive
|
|
;; (list (read-buffer "Switch to buffer: " (other-buffer)
|
|
;; (null current-prefix-arg)))))
|
|
;;
|
|
;;(defadvice switch-to-buffer (around confirm-non-existing-buffers activate)
|
|
;; "Switch to non-existing buffers only upon confirmation."
|
|
;; (interactive "BSwitch to buffer: ")
|
|
;; (if (or (get-buffer (ad-get-arg 0))
|
|
;; (y-or-n-p (format "`%s' does not exist, create? " (ad-get-arg 0))))
|
|
;; ad-do-it))
|
|
;;
|
|
;;(defadvice find-file (before existing-files-only activate)
|
|
;; "Find existing files only"
|
|
;; (interactive "fFind file: "))
|
|
;;
|
|
;;(defadvice car (around interactive activate)
|
|
;; "Make `car' an interactive function."
|
|
;; (interactive "xCar of list: ")
|
|
;; ad-do-it
|
|
;; (if (interactive-p)
|
|
;; (message "%s" ad-return-value)))
|
|
|
|
|
|
;; @ Advice documentation:
|
|
;; =======================
|
|
;; Below is general documentation of the various features of advice. For more
|
|
;; concrete examples check the corresponding sections in the tutorial part.
|
|
|
|
;; @@ Terminology:
|
|
;; ===============
|
|
;; - Emacs, Emacs-19: Emacs as released by the GNU Project
|
|
;; - Lemacs: Lucid's version of Emacs with major version 19
|
|
;; - v18: Any Emacs with major version 18 or built as an extension to that
|
|
;; (such as Epoch)
|
|
;; - v19: Any Emacs with major version 19
|
|
;; - jwz: Jamie Zawinski - former keeper of Lemacs and creator of the optimizing
|
|
;; byte-compiler used in v19s.
|
|
;; - Advice: The name of this package.
|
|
;; - advices: Short for "pieces of advice".
|
|
|
|
;; @@ Defining a piece of advice with `defadvice':
|
|
;; ===============================================
|
|
;; The main means of defining a piece of advice is the macro `defadvice',
|
|
;; there is no interactive way of specifying a piece of advice. A call to
|
|
;; `defadvice' has the following syntax which is similar to the syntax of
|
|
;; `defun/defmacro':
|
|
;;
|
|
;; (defadvice <function> (<class> <name> [<position>] [<arglist>] {<flags>}*)
|
|
;; [ [<documentation-string>] [<interactive-form>] ]
|
|
;; {<body-form>}* )
|
|
|
|
;; <function> is the name of the function/macro/subr to be advised.
|
|
|
|
;; <class> is the class of the advice which has to be one of `before',
|
|
;; `around', `after', `activation' or `deactivation' (the last two allow
|
|
;; definition of special act/deactivation hooks).
|
|
|
|
;; <name> is the name of the advice which has to be a non-nil symbol.
|
|
;; Names uniquely identify a piece of advice in a certain advice class,
|
|
;; hence, advices can be redefined by defining an advice with the same class
|
|
;; and name. Advice names are global symbols, hence, the same name space
|
|
;; conventions used for function names should be applied.
|
|
|
|
;; An optional <position> specifies where in the current list of advices of
|
|
;; the specified <class> this new advice will be placed. <position> has to
|
|
;; be either `first', `last' or a number that specifies a zero-based
|
|
;; position (`first' is equivalent to 0). If no position is specified
|
|
;; `first' will be used as a default. If this call to `defadvice' redefines
|
|
;; an already existing advice (see above) then the position argument will
|
|
;; be ignored and the position of the already existing advice will be used.
|
|
|
|
;; An optional <arglist> which has to be a list can be used to define the
|
|
;; argument list of the advised function. This argument list should of
|
|
;; course be compatible with the argument list of the original function,
|
|
;; otherwise functions that call the advised function with the original
|
|
;; argument list in mind will break. If more than one advice specify an
|
|
;; argument list then the first one (the one with the smallest position)
|
|
;; found in the list of before/around/after advices will be used.
|
|
|
|
;; <flags> is a list of symbols that specify further information about the
|
|
;; advice. All flags can be specified with unambiguous initial substrings.
|
|
;; `activate': Specifies that the advice information of the advised
|
|
;; function should be activated right after this advice has been
|
|
;; defined. In forward advices `activate' will be ignored.
|
|
;; `protect': Specifies that this advice should be protected against
|
|
;; non-local exits and errors in preceding code/advices.
|
|
;; `compile': Specifies that the advised function should be byte-compiled.
|
|
;; This flag will be ignored unless `activate' is also specified.
|
|
;; `disable': Specifies that the defined advice should be disabled, hence,
|
|
;; it will not be used in an activation until somebody enables it.
|
|
;; `preactivate': Specifies that the advised function should get preactivated
|
|
;; at macro-expansion/compile time of this `defadvice'. This
|
|
;; generates a compiled advised definition according to the
|
|
;; current advice state which will be used during activation
|
|
;; if appropriate. Only use this if the `defadvice' gets
|
|
;; actually compiled (with a v18 byte-compiler put the `defadvice'
|
|
;; into the body of a `defun' to accomplish proper compilation).
|
|
|
|
;; An optional <documentation-string> can be supplied to document the advice.
|
|
;; On call of the `documentation' function it will be combined with the
|
|
;; documentation strings of the original function and other advices.
|
|
|
|
;; An optional <interactive-form> form can be supplied to change/add
|
|
;; interactive behavior of the original function. If more than one advice
|
|
;; has an `(interactive ...)' specification then the first one (the one
|
|
;; with the smallest position) found in the list of before/around/after
|
|
;; advices will be used.
|
|
|
|
;; A possibly empty list of <body-forms> specifies the body of the advice in
|
|
;; an implicit progn. The body of an advice can access/change arguments,
|
|
;; the return value, the binding environment, and can have all sorts of
|
|
;; other side effects.
|
|
|
|
;; @@ Assembling advised definitions:
|
|
;; ==================================
|
|
;; Suppose a function/macro/subr/special-form has N pieces of before advice,
|
|
;; M pieces of around advice and K pieces of after advice. Assuming none of
|
|
;; the advices is protected, its advised definition will look like this
|
|
;; (body-form indices correspond to the position of the respective advice in
|
|
;; that advice class):
|
|
|
|
;; ([macro] lambda <arglist>
|
|
;; [ [<advised-docstring>] [(interactive ...)] ]
|
|
;; (let (ad-return-value)
|
|
;; {<before-0-body-form>}*
|
|
;; ....
|
|
;; {<before-N-1-body-form>}*
|
|
;; {<around-0-body-form>}*
|
|
;; {<around-1-body-form>}*
|
|
;; ....
|
|
;; {<around-M-1-body-form>}*
|
|
;; (setq ad-return-value
|
|
;; <apply original definition to <arglist>>)
|
|
;; {<other-around-M-1-body-form>}*
|
|
;; ....
|
|
;; {<other-around-1-body-form>}*
|
|
;; {<other-around-0-body-form>}*
|
|
;; {<after-0-body-form>}*
|
|
;; ....
|
|
;; {<after-K-1-body-form>}*
|
|
;; ad-return-value))
|
|
|
|
;; Macros and special forms will be redefined as macros, hence the optional
|
|
;; [macro] in the beginning of the definition.
|
|
|
|
;; <arglist> is either the argument list of the original function or the
|
|
;; first argument list defined in the list of before/around/after advices.
|
|
;; The values of <arglist> variables can be accessed/changed in the body of
|
|
;; an advice by simply referring to them by their original name, however,
|
|
;; more portable argument access macros are also provided (see below). For
|
|
;; subrs/special-forms for which neither explicit argument list definitions
|
|
;; are available, nor their documentation strings contain such definitions
|
|
;; (as they do v19s), `(&rest ad-subr-args)' will be used.
|
|
|
|
;; <advised-docstring> is an optional, special documentation string which will
|
|
;; be expanded into a proper documentation string upon call of `documentation'.
|
|
|
|
;; (interactive ...) is an optional interactive form either taken from the
|
|
;; original function or from a before/around/after advice. For advised
|
|
;; interactive subrs that do not have an interactive form specified in any
|
|
;; advice we have to use (interactive) and then call the subr interactively
|
|
;; if the advised function was called interactively, because the
|
|
;; interactive specification of subrs is not accessible. This is the only
|
|
;; case where changing the values of arguments will not have an affect
|
|
;; because they will be reset by the interactive specification of the subr.
|
|
;; If this is a problem one can always specify an interactive form in a
|
|
;; before/around/after advice to gain control over argument values that
|
|
;; were supplied interactively.
|
|
;;
|
|
;; Then the body forms of the various advices in the various classes of advice
|
|
;; are assembled in order. The forms of around advice L are normally part of
|
|
;; one of the forms of around advice L-1. An around advice can specify where
|
|
;; the forms of the wrapped or surrounded forms should go with the special
|
|
;; keyword `ad-do-it', which will be substituted with a `progn' containing the
|
|
;; forms of the surrounded code.
|
|
|
|
;; The innermost part of the around advice onion is
|
|
;; <apply original definition to <arglist>>
|
|
;; whose form depends on the type of the original function. The variable
|
|
;; `ad-return-value' will be set to its result. This variable is visible to
|
|
;; all pieces of advice which can access and modify it before it gets returned.
|
|
;;
|
|
;; The semantic structure of advised functions that contain protected pieces
|
|
;; of advice is the same. The only difference is that `unwind-protect' forms
|
|
;; make sure that the protected advice gets executed even if some previous
|
|
;; piece of advice had an error or a non-local exit. If any around advice is
|
|
;; protected then the whole around advice onion will be protected.
|
|
|
|
;; @@ Argument access in advised functions:
|
|
;; ========================================
|
|
;; As already mentioned, the simplest way to access the arguments of an
|
|
;; advised function in the body of an advice is to refer to them by name. To
|
|
;; do that, the advice programmer needs to know either the names of the
|
|
;; argument variables of the original function, or the names used in the
|
|
;; argument list redefinition given in a piece of advice. While this simple
|
|
;; method might be sufficient in many cases, it has the disadvantage that it
|
|
;; is not very portable because it hardcodes the argument names into the
|
|
;; advice. If the definition of the original function changes the advice
|
|
;; might break even though the code might still be correct. Situations like
|
|
;; that arise, for example, if one advises a subr like `eval-region' which
|
|
;; gets redefined in a non-advice style into a function by the edebug
|
|
;; package. If the advice assumes `eval-region' to be a subr it might break
|
|
;; once edebug is loaded. Similar situations arise when one wants to use the
|
|
;; same piece of advice across different versions of Emacs. Some subrs in a
|
|
;; v18 Emacs are functions in v19 and vice versa, but for the most part the
|
|
;; semantics remain the same, hence, the same piece of advice might be usable
|
|
;; in both Emacs versions.
|
|
|
|
;; As a solution to that advice provides argument list access macros that get
|
|
;; translated into the proper access forms at activation time, i.e., when the
|
|
;; advised definition gets constructed. Access macros access actual arguments
|
|
;; by position regardless of how these actual argument get distributed onto
|
|
;; the argument variables of a function. The rational behind this is that in
|
|
;; Emacs Lisp the semantics of an argument is strictly determined by its
|
|
;; position (there are no keyword arguments).
|
|
|
|
;; Suppose the function `foo' is defined as
|
|
;;
|
|
;; (defun foo (x y &optional z &rest r) ....)
|
|
;;
|
|
;; and is then called with
|
|
;;
|
|
;; (foo 0 1 2 3 4 5 6)
|
|
|
|
;; which means that X=0, Y=1, Z=2 and R=(3 4 5 6). The assumption is that
|
|
;; the semantics of an actual argument is determined by its position. It is
|
|
;; this semantics that has to be known by the advice programmer. Then s/he
|
|
;; can access these arguments in a piece of advice with some of the
|
|
;; following macros (the arrows indicate what value they will return):
|
|
|
|
;; (ad-get-arg 0) -> 0
|
|
;; (ad-get-arg 1) -> 1
|
|
;; (ad-get-arg 2) -> 2
|
|
;; (ad-get-arg 3) -> 3
|
|
;; (ad-get-args 2) -> (2 3 4 5 6)
|
|
;; (ad-get-args 4) -> (4 5 6)
|
|
|
|
;; `(ad-get-arg <position>)' will return the actual argument that was supplied
|
|
;; at <position>, `(ad-get-args <position>)' will return the list of actual
|
|
;; arguments supplied starting at <position>. Note that these macros can be
|
|
;; used without any knowledge about the form of the actual argument list of
|
|
;; the original function.
|
|
|
|
;; Similarly, `(ad-set-arg <position> <value-form>)' can be used to set the
|
|
;; value of the actual argument at <position> to <value-form>. For example,
|
|
;;
|
|
;; (ad-set-arg 5 "five")
|
|
;;
|
|
;; will have the effect that R=(3 4 "five" 6) once the original function is
|
|
;; called. `(ad-set-args <position> <value-list-form>)' can be used to set
|
|
;; the list of actual arguments starting at <position> to <value-list-form>.
|
|
;; For example,
|
|
;;
|
|
;; (ad-set-args 0 '(5 4 3 2 1 0))
|
|
;;
|
|
;; will have the effect that X=5, Y=4, Z=3 and R=(2 1 0) once the original
|
|
;; function is called.
|
|
|
|
;; All these access macros are text macros rather than real Lisp macros. When
|
|
;; the advised definition gets constructed they get replaced with actual access
|
|
;; forms depending on the argument list of the advised function, i.e., after
|
|
;; that argument access is in most cases as efficient as using the argument
|
|
;; variable names directly.
|
|
|
|
;; @@@ Accessing argument bindings of arbitrary functions:
|
|
;; =======================================================
|
|
;; Some functions (such as `trace-function' defined in trace.el) need a
|
|
;; method of accessing the names and bindings of the arguments of an
|
|
;; arbitrary advised function. To do that within an advice one can use the
|
|
;; special keyword `ad-arg-bindings' which is a text macro that will be
|
|
;; substituted with a form that will evaluate to a list of binding
|
|
;; specifications, one for every argument variable. These binding
|
|
;; specifications can then be examined in the body of the advice. For
|
|
;; example, somewhere in an advice we could do this:
|
|
;;
|
|
;; (let* ((bindings ad-arg-bindings)
|
|
;; (firstarg (car bindings))
|
|
;; (secondarg (car (cdr bindings))))
|
|
;; ;; Print info about first argument
|
|
;; (print (format "%s=%s (%s)"
|
|
;; (ad-arg-binding-field firstarg 'name)
|
|
;; (ad-arg-binding-field firstarg 'value)
|
|
;; (ad-arg-binding-field firstarg 'type)))
|
|
;; ....)
|
|
;;
|
|
;; The `type' of an argument is either `required', `optional' or `rest'.
|
|
;; Wherever `ad-arg-bindings' appears a form will be inserted that evaluates
|
|
;; to the list of bindings, hence, in order to avoid multiple unnecessary
|
|
;; evaluations one should always bind it to some variable.
|
|
|
|
;; @@@ Argument list mapping:
|
|
;; ==========================
|
|
;; Because `defadvice' allows the specification of the argument list of the
|
|
;; advised function we need a mapping mechanism that maps this argument list
|
|
;; onto that of the original function. For example, somebody might specify
|
|
;; `(sym newdef)' as the argument list of `fset', while advice might use
|
|
;; `(&rest ad-subr-args)' as the argument list of the original function
|
|
;; (depending on what Emacs version is used). Hence SYM and NEWDEF have to
|
|
;; be properly mapped onto the &rest variable when the original definition is
|
|
;; called. Advice automatically takes care of that mapping, hence, the advice
|
|
;; programmer can specify an argument list without having to know about the
|
|
;; exact structure of the original argument list as long as the new argument
|
|
;; list takes a compatible number/magnitude of actual arguments.
|
|
|
|
;; @@@ Definition of subr argument lists:
|
|
;; ======================================
|
|
;; When advice constructs the advised definition of a function it has to
|
|
;; know the argument list of the original function. For functions and macros
|
|
;; the argument list can be determined from the actual definition, however,
|
|
;; for subrs there is no such direct access available. In Lemacs and for some
|
|
;; subrs in Emacs-19 the argument list of a subr can be determined from
|
|
;; its documentation string, in a v18 Emacs even that is not possible. If
|
|
;; advice cannot at all determine the argument list of a subr it uses
|
|
;; `(&rest ad-subr-args)' which will always work but is inefficient because
|
|
;; it conses up arguments. The macro `ad-define-subr-args' can be used by
|
|
;; the advice programmer to explicitly tell advice about the argument list
|
|
;; of a certain subr, for example,
|
|
;;
|
|
;; (ad-define-subr-args 'fset '(sym newdef))
|
|
;;
|
|
;; is used by advice itself to tell a v18 Emacs about the arguments of `fset'.
|
|
;; The following can be used to undo such a definition:
|
|
;;
|
|
;; (ad-undefine-subr-args 'fset)
|
|
;;
|
|
;; The argument list definition is stored on the property list of the subr
|
|
;; name symbol. When an argument list could be determined from the
|
|
;; documentation string it will be cached under that property. The general
|
|
;; mechanism for looking up the argument list of a subr is the following:
|
|
;; 1) look for a definition stored on the property list
|
|
;; 2) if that failed try to infer it from the documentation string and
|
|
;; if successful cache it on the property list
|
|
;; 3) otherwise use `(&rest ad-subr-args)'
|
|
|
|
;; @@ Activation and deactivation:
|
|
;; ===============================
|
|
;; The definition of an advised function does not change until all its advice
|
|
;; gets actually activated. Activation can either happen with the `activate'
|
|
;; flag specified in the `defadvice', with an explicit call or interactive
|
|
;; invocation of `ad-activate', or if forward advice is enabled (i.e., the
|
|
;; value of `ad-activate-on-definition' is t) at the time an already advised
|
|
;; function gets defined.
|
|
|
|
;; When a function gets first activated its original definition gets saved,
|
|
;; all defined and enabled pieces of advice will get combined with the
|
|
;; original definition, the resulting definition might get compiled depending
|
|
;; on some conditions described below, and then the function will get
|
|
;; redefined with the advised definition. This also means that undefined
|
|
;; functions cannot get activated even though they might be already advised.
|
|
|
|
;; The advised definition will get compiled either if `ad-activate' was called
|
|
;; interactively with a prefix argument, or called explicitly with its second
|
|
;; argument as t, or, if `ad-default-compilation-action' justifies it according
|
|
;; to the current system state. If the advised definition was
|
|
;; constructed during "preactivation" (see below) then that definition will
|
|
;; be already compiled because it was constructed during byte-compilation of
|
|
;; the file that contained the `defadvice' with the `preactivate' flag.
|
|
|
|
;; `ad-deactivate' can be used to back-define an advised function to its
|
|
;; original definition. It can be called interactively or directly. Because
|
|
;; `ad-activate' caches the advised definition the function can be
|
|
;; reactivated via `ad-activate' with only minor overhead (it is checked
|
|
;; whether the current advice state is consistent with the cached
|
|
;; definition, see the section on caching below).
|
|
|
|
;; `ad-activate-regexp' and `ad-deactivate-regexp' can be used to de/activate
|
|
;; all currently advised function that have a piece of advice with a name that
|
|
;; contains a match for a regular expression. These functions can be used to
|
|
;; de/activate sets of functions depending on certain advice naming
|
|
;; conventions.
|
|
|
|
;; Finally, `ad-activate-all' and `ad-deactivate-all' can be used to
|
|
;; de/activate all currently advised functions. These are useful to
|
|
;; (temporarily) return to an un/advised state.
|
|
|
|
;; @@@ Reasons for the separation of advice definition and activation:
|
|
;; ===================================================================
|
|
;; As already mentioned, advising happens in two stages:
|
|
|
|
;; 1) definition of various pieces of advice
|
|
;; 2) activation of all advice currently defined and enabled
|
|
|
|
;; The advantage of this is that various pieces of advice can be defined
|
|
;; before they get combined into an advised definition which avoids
|
|
;; unnecessary constructions of intermediate advised definitions. The more
|
|
;; important advantage is that it allows the implementation of forward advice.
|
|
;; Advice information for a certain function accumulates as the value of the
|
|
;; `advice-info' property of the function symbol. This accumulation is
|
|
;; completely independent of the fact that that function might not yet be
|
|
;; defined. The special forms `defun' and `defmacro' have been advised to
|
|
;; check whether the function/macro they defined had advice information
|
|
;; associated with it. If so and forward advice is enabled, the original
|
|
;; definition will be saved, and then the advice will be activated. When a
|
|
;; file is loaded in a v18 Emacs the functions/macros it defines are also
|
|
;; defined with calls to `defun/defmacro'. Hence, we can forward advise
|
|
;; functions/macros which will be defined later during a load/autoload of some
|
|
;; file (for compiled files generated by jwz's byte-compiler in a v19 Emacs
|
|
;; this is slightly more complicated but the basic idea is the same).
|
|
|
|
;; @@ Enabling/disabling pieces or sets of advice:
|
|
;; ===============================================
|
|
;; A major motivation for the development of this advice package was to bring
|
|
;; a little bit more structure into the function overloading chaos in Emacs
|
|
;; Lisp. Many packages achieve some of their functionality by adding a little
|
|
;; bit (or a lot) to the standard functionality of some Emacs Lisp function.
|
|
;; ange-ftp is a very popular package that achieves its magic by overloading
|
|
;; most Emacs Lisp functions that deal with files. A popular function that's
|
|
;; overloaded by many packages is `expand-file-name'. The situation that one
|
|
;; function is multiply overloaded can arise easily.
|
|
|
|
;; Once in a while it would be desirable to be able to disable some/all
|
|
;; overloads of a particular package while keeping all the rest. Ideally -
|
|
;; at least in my opinion - these overloads would all be done with advice,
|
|
;; I know I am dreaming right now... In that ideal case the enable/disable
|
|
;; mechanism of advice could be used to achieve just that.
|
|
|
|
;; Every piece of advice is associated with an enablement flag. When the
|
|
;; advised definition of a particular function gets constructed (e.g., during
|
|
;; activation) only the currently enabled pieces of advice will be considered.
|
|
;; This mechanism allows one to have different "views" of an advised function
|
|
;; dependent on what pieces of advice are currently enabled.
|
|
|
|
;; Another motivation for this mechanism is that it allows one to define a
|
|
;; piece of advice for some function yet keep it dormant until a certain
|
|
;; condition is met. Until then activation of the function will not make use
|
|
;; of that piece of advice. Once the condition is met the advice can be
|
|
;; enabled and a reactivation of the function will add its functionality as
|
|
;; part of the new advised definition. For example, the advices of `defun'
|
|
;; etc. used by advice itself will stay disabled until `ad-start-advice' is
|
|
;; called and some variables have the proper values. Hence, if somebody
|
|
;; else advised these functions too and activates them the advices defined
|
|
;; by advice will get used only if they are intended to be used.
|
|
|
|
;; The main interface to this mechanism are the interactive functions
|
|
;; `ad-enable-advice' and `ad-disable-advice'. For example, the following
|
|
;; would disable a particular advice of the function `foo':
|
|
;;
|
|
;; (ad-disable-advice 'foo 'before 'my-advice)
|
|
;;
|
|
;; This call by itself only changes the flag, to get the proper effect in
|
|
;; the advised definition too one has to activate `foo' with
|
|
;;
|
|
;; (ad-activate 'foo)
|
|
;;
|
|
;; or interactively. To disable whole sets of advices one can use a regular
|
|
;; expression mechanism. For example, let us assume that ange-ftp actually
|
|
;; used advice to overload all its functions, and that it used the
|
|
;; "ange-ftp-" prefix for all its advice names, then we could temporarily
|
|
;; disable all its advices with
|
|
;;
|
|
;; (ad-disable-regexp "^ange-ftp-")
|
|
;;
|
|
;; and the following call would put that actually into effect:
|
|
;;
|
|
;; (ad-activate-regexp "^ange-ftp-")
|
|
;;
|
|
;; A saver way would have been to use
|
|
;;
|
|
;; (ad-update-regexp "^ange-ftp-")
|
|
;;
|
|
;; instead which would have only reactivated currently actively advised
|
|
;; functions, but not functions that were currently deactivated. All these
|
|
;; functions can also be called interactively.
|
|
|
|
;; A certain piece of advice is considered a match if its name contains a
|
|
;; match for the regular expression. To enable ange-ftp again we would use
|
|
;; `ad-enable-regexp' and then activate or update again.
|
|
|
|
;; @@ Forward advice, automatic advice activation:
|
|
;; ===============================================
|
|
;; Because most Emacs Lisp packages are loaded on demand via an autoload
|
|
;; mechanism it is essential to be able to "forward advise" functions.
|
|
;; Otherwise, proper advice definition and activation would make it necessary
|
|
;; to preload every file that defines a certain function before it can be
|
|
;; advised, which would partly defeat the purpose of the advice mechanism.
|
|
|
|
;; In the following, "forward advice" always implies its automatic activation
|
|
;; once a function gets defined, and not just the accumulation of advice
|
|
;; information for a possibly undefined function.
|
|
|
|
;; Advice implements forward advice mainly via the following: 1) Separation
|
|
;; of advice definition and activation that makes it possible to accumulate
|
|
;; advice information without having the original function already defined,
|
|
;; 2) special versions of the built-in functions `fset/defalias' which check
|
|
;; for advice information whenever they define a function. If advice
|
|
;; information was found then the advice will immediately get activated when
|
|
;; the function gets defined.
|
|
|
|
;; Automatic advice activation means, that whenever a function gets defined
|
|
;; with either `defun', `defmacro', `fset' or by loading a byte-compiled
|
|
;; file, and the function has some advice-info stored with it then that
|
|
;; advice will get activated right away.
|
|
|
|
;; @@@ Enabling automatic advice activation:
|
|
;; =========================================
|
|
;; Automatic advice activation is enabled by default. It can be disabled by
|
|
;; doint `M-x ad-stop-advice' and enabled again with `M-x ad-start-advice'.
|
|
|
|
;; @@ Caching of advised definitions:
|
|
;; ==================================
|
|
;; After an advised definition got constructed it gets cached as part of the
|
|
;; advised function's advice-info so it can be reused, for example, after an
|
|
;; intermediate deactivation. Because the advice-info of a function might
|
|
;; change between the time of caching and reuse a cached definition gets
|
|
;; a cache-id associated with it so it can be verified whether the cached
|
|
;; definition is still valid (the main application of this is preactivation
|
|
;; - see below).
|
|
|
|
;; When an advised function gets activated and a verifiable cached definition
|
|
;; is available, then that definition will be used instead of creating a new
|
|
;; advised definition from scratch. If you want to make sure that a new
|
|
;; definition gets constructed then you should use `ad-clear-cache' before you
|
|
;; activate the advised function.
|
|
|
|
;; @@ Preactivation:
|
|
;; =================
|
|
;; Constructing an advised definition is moderately expensive. In a situation
|
|
;; where one package defines a lot of advised functions it might be
|
|
;; prohibitively expensive to do all the advised definition construction at
|
|
;; runtime. Preactivation is a mechanism that allows compile-time construction
|
|
;; of compiled advised definitions that can be activated cheaply during
|
|
;; runtime. Preactivation uses the caching mechanism to do that. Here's how it
|
|
;; works:
|
|
|
|
;; When the byte-compiler compiles a `defadvice' that has the `preactivate'
|
|
;; flag specified, it uses the current original definition of the advised
|
|
;; function plus the advice specified in this `defadvice' (even if it is
|
|
;; specified as disabled) and all other currently enabled pieces of advice to
|
|
;; construct an advised definition and an identifying cache-id and makes them
|
|
;; part of the `defadvice' expansion which will then be compiled by the
|
|
;; byte-compiler (to ensure that in a v18 emacs you have to put the
|
|
;; `defadvice' inside a `defun' to get it compiled and then you have to call
|
|
;; that compiled `defun' in order to actually execute the `defadvice'). When
|
|
;; the file with the compiled, preactivating `defadvice' gets loaded the
|
|
;; precompiled advised definition will be cached on the advised function's
|
|
;; advice-info. When it gets activated (can be immediately on execution of the
|
|
;; `defadvice' or any time later) the cache-id gets checked against the
|
|
;; current state of advice and if it is verified the precompiled definition
|
|
;; will be used directly (the verification is pretty cheap). If it couldn't get
|
|
;; verified a new advised definition for that function will be built from
|
|
;; scratch, hence, the efficiency added by the preactivation mechanism does
|
|
;; not at all impair the flexibility of the advice mechanism.
|
|
|
|
;; MORAL: In order get all the efficiency out of preactivation the advice
|
|
;; state of an advised function at the time the file with the
|
|
;; preactivating `defadvice' gets byte-compiled should be exactly
|
|
;; the same as it will be when the advice of that function gets
|
|
;; actually activated. If it is not there is a high chance that the
|
|
;; cache-id will not match and hence a new advised definition will
|
|
;; have to be constructed at runtime.
|
|
|
|
;; Preactivation and forward advice do not contradict each other. It is
|
|
;; perfectly ok to load a file with a preactivating `defadvice' before the
|
|
;; original definition of the advised function is available. The constructed
|
|
;; advised definition will be used once the original function gets defined and
|
|
;; its advice gets activated. The only constraint is that at the time the
|
|
;; file with the preactivating `defadvice' got compiled the original function
|
|
;; definition was available.
|
|
|
|
;; TIPS: Here are some indications that a preactivation did not work the way
|
|
;; you intended it to work:
|
|
;; - Activation of the advised function takes longer than usual/expected
|
|
;; - The byte-compiler gets loaded while an advised function gets
|
|
;; activated
|
|
;; - `byte-compile' is part of the `features' variable even though you
|
|
;; did not use the byte-compiler
|
|
;; Right now advice does not provide an elegant way to find out whether
|
|
;; and why a preactivation failed. What you can do is to trace the
|
|
;; function `ad-cache-id-verification-code' (with the function
|
|
;; `trace-function-background' defined in my trace.el package) before
|
|
;; any of your advised functions get activated. After they got
|
|
;; activated check whether all calls to `ad-cache-id-verification-code'
|
|
;; returned `verified' as a result. Other values indicate why the
|
|
;; verification failed which should give you enough information to
|
|
;; fix your preactivation/compile/load/activation sequence.
|
|
|
|
;; IMPORTANT: There is one case (that I am aware of) that can make
|
|
;; preactivation fail, i.e., a preconstructed advised definition that does
|
|
;; NOT match the current state of advice gets used nevertheless. That case
|
|
;; arises if one package defines a certain piece of advice which gets used
|
|
;; during preactivation, and another package incompatibly redefines that
|
|
;; very advice (i.e., same function/class/name), and it is the second advice
|
|
;; that is available when the preconstructed definition gets activated, and
|
|
;; that was the only definition of that advice so far (`ad-add-advice'
|
|
;; catches advice redefinitions and clears the cache in such a case).
|
|
;; Catching that would make the cache verification too expensive.
|
|
|
|
;; MORAL-II: Redefining somebody else's advice is BAAAAD (to speak with
|
|
;; George Walker Bush), and why would you redefine your own advice anyway?
|
|
;; Advice is a mechanism to facilitate function redefinition, not advice
|
|
;; redefinition (wait until I write Meta-Advice :-). If you really have
|
|
;; to undo somebody else's advice try to write a "neutralizing" advice.
|
|
|
|
;; @@ Advising macros and special forms and other dangerous things:
|
|
;; ================================================================
|
|
;; Look at the corresponding tutorial sections for more information on
|
|
;; these topics. Here it suffices to point out that the special treatment
|
|
;; of macros and special forms by the byte-compiler can lead to problems
|
|
;; when they get advised. Macros can create problems because they get
|
|
;; expanded at compile time, hence, they might not have all the necessary
|
|
;; runtime support and such advice cannot be de/activated or changed as
|
|
;; it is possible for functions. Special forms create problems because they
|
|
;; have to be advised "into" macros, i.e., an advised special form is a
|
|
;; implemented as a macro, hence, in most cases the byte-compiler will
|
|
;; not recognize it as a special form anymore which can lead to very strange
|
|
;; results.
|
|
;;
|
|
;; MORAL: - Only advise macros or special forms when you are absolutely sure
|
|
;; what you are doing.
|
|
;; - As a safety measure, always do `ad-deactivate-all' before you
|
|
;; byte-compile a file to make sure that even if some inconsiderate
|
|
;; person advised some special forms you'll get proper compilation
|
|
;; results. After compilation do `ad-activate-all' to get back to
|
|
;; the previous state.
|
|
|
|
;; @@ Adding a piece of advice with `ad-add-advice':
|
|
;; =================================================
|
|
;; The non-interactive function `ad-add-advice' can be used to add a piece of
|
|
;; advice to some function without using `defadvice'. This is useful if advice
|
|
;; has to be added somewhere by a function (also look at `ad-make-advice').
|
|
|
|
;; @@ Activation/deactivation advices, file load hooks:
|
|
;; ====================================================
|
|
;; There are two special classes of advice called `activation' and
|
|
;; `deactivation'. The body forms of these advices are not included into the
|
|
;; advised definition of a function, rather they are assembled into a hook
|
|
;; form which will be evaluated whenever the advice-info of the advised
|
|
;; function gets activated or deactivated. One application of this mechanism
|
|
;; is to define file load hooks for files that do not provide such hooks
|
|
;; (v19s already come with a general file-load-hook mechanism, v18s don't).
|
|
;; For example, suppose you want to print a message whenever `file-x' gets
|
|
;; loaded, and suppose the last function defined in `file-x' is
|
|
;; `file-x-last-fn'. Then we can define the following advice:
|
|
;;
|
|
;; (defadvice file-x-last-fn (activation file-x-load-hook)
|
|
;; "Executed whenever file-x is loaded"
|
|
;; (if load-in-progress (message "Loaded file-x")))
|
|
;;
|
|
;; This will constitute a forward advice for function `file-x-last-fn' which
|
|
;; will get activated when `file-x' is loaded (only if forward advice is
|
|
;; enabled of course). Because there are no "real" pieces of advice
|
|
;; available for it, its definition will not be changed, but the activation
|
|
;; advice will be run during its activation which is equivalent to having a
|
|
;; file load hook for `file-x'.
|
|
|
|
;; @@ Summary of main advice concepts:
|
|
;; ===================================
|
|
;; - Definition:
|
|
;; A piece of advice gets defined with `defadvice' and added to the
|
|
;; `advice-info' property of a function.
|
|
;; - Enablement:
|
|
;; Every piece of advice has an enablement flag associated with it. Only
|
|
;; enabled advices are considered during construction of an advised
|
|
;; definition.
|
|
;; - Activation:
|
|
;; Redefine an advised function with its advised definition. Constructs
|
|
;; an advised definition from scratch if no verifiable cached advised
|
|
;; definition is available and caches it.
|
|
;; - Deactivation:
|
|
;; Back-define an advised function to its original definition.
|
|
;; - Update:
|
|
;; Reactivate an advised function but only if its advice is currently
|
|
;; active. This can be used to bring all currently advised function up
|
|
;; to date with the current state of advice without also activating
|
|
;; currently deactivated functions.
|
|
;; - Caching:
|
|
;; Is the saving of an advised definition and an identifying cache-id so
|
|
;; it can be reused, for example, for activation after deactivation.
|
|
;; - Preactivation:
|
|
;; Is the construction of an advised definition according to the current
|
|
;; state of advice during byte-compilation of a file with a preactivating
|
|
;; `defadvice'. That advised definition can then rather cheaply be used
|
|
;; during activation without having to construct an advised definition
|
|
;; from scratch at runtime.
|
|
|
|
;; @@ Summary of interactive advice manipulation functions:
|
|
;; ========================================================
|
|
;; The following interactive functions can be used to manipulate the state
|
|
;; of advised functions (all of them support completion on function names,
|
|
;; advice classes and advice names):
|
|
|
|
;; - ad-activate to activate the advice of a FUNCTION
|
|
;; - ad-deactivate to deactivate the advice of a FUNCTION
|
|
;; - ad-update to activate the advice of a FUNCTION unless it was not
|
|
;; yet activated or is currently deactivated.
|
|
;; - ad-unadvise deactivates a FUNCTION and removes all of its advice
|
|
;; information, hence, it cannot be activated again
|
|
;; - ad-recover tries to redefine a FUNCTION to its original definition and
|
|
;; discards all advice information (a low-level `ad-unadvise').
|
|
;; Use only in emergencies.
|
|
|
|
;; - ad-remove-advice removes a particular piece of advice of a FUNCTION.
|
|
;; You still have to do call `ad-activate' or `ad-update' to
|
|
;; activate the new state of advice.
|
|
;; - ad-enable-advice enables a particular piece of advice of a FUNCTION.
|
|
;; - ad-disable-advice disables a particular piece of advice of a FUNCTION.
|
|
;; - ad-enable-regexp maps over all currently advised functions and enables
|
|
;; every advice whose name contains a match for a regular
|
|
;; expression.
|
|
;; - ad-disable-regexp disables matching advices.
|
|
|
|
;; - ad-activate-regexp activates all advised function with a matching advice
|
|
;; - ad-deactivate-regexp deactivates all advised function with matching advice
|
|
;; - ad-update-regexp updates all advised function with a matching advice
|
|
;; - ad-activate-all activates all advised functions
|
|
;; - ad-deactivate-all deactivates all advised functions
|
|
;; - ad-update-all updates all advised functions
|
|
;; - ad-unadvise-all unadvises all advised functions
|
|
;; - ad-recover-all recovers all advised functions
|
|
|
|
;; - ad-compile byte-compiles a function/macro if it is compilable.
|
|
|
|
;; @@ Summary of forms with special meanings when used within an advice:
|
|
;; =====================================================================
|
|
;; ad-return-value name of the return value variable (get/settable)
|
|
;; ad-subr-args name of &rest argument variable used for advised
|
|
;; subrs whose actual argument list cannot be
|
|
;; determined (get/settable)
|
|
;; (ad-get-arg <pos>), (ad-get-args <pos>),
|
|
;; (ad-set-arg <pos> <value>), (ad-set-args <pos> <value-list>)
|
|
;; argument access text macros to get/set the values of
|
|
;; actual arguments at a certain position
|
|
;; ad-arg-bindings text macro that returns the actual names, values
|
|
;; and types of the arguments as a list of bindings. The
|
|
;; order of the bindings corresponds to the order of the
|
|
;; arguments. The individual fields of every binding (name,
|
|
;; value and type) can be accessed with the function
|
|
;; `ad-arg-binding-field' (see example above).
|
|
;; ad-do-it text macro that identifies the place where the original
|
|
;; or wrapped definition should go in an around advice
|
|
|
|
|
|
;; @ Foo games: An advice tutorial
|
|
;; ===============================
|
|
;; The following tutorial was created in Emacs 18.59. Left-justified
|
|
;; s-expressions are input forms followed by one or more result forms.
|
|
;; First we have to start the advice magic:
|
|
;;
|
|
;; (ad-start-advice)
|
|
;; nil
|
|
;;
|
|
;; We start by defining an innocent looking function `foo' that simply
|
|
;; adds 1 to its argument X:
|
|
;;
|
|
;; (defun foo (x)
|
|
;; "Add 1 to X."
|
|
;; (1+ x))
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; 4
|
|
;;
|
|
;; @@ Defining a simple piece of advice:
|
|
;; =====================================
|
|
;; Now let's define the first piece of advice for `foo'. To do that we
|
|
;; use the macro `defadvice' which takes a function name, a list of advice
|
|
;; specifiers and a list of body forms as arguments. The first element of
|
|
;; the advice specifiers is the class of the advice, the second is its name,
|
|
;; the third its position and the rest are some flags. The class of our
|
|
;; first advice is `before', its name is `fg-add2', its position among the
|
|
;; currently defined before advices (none so far) is `first', and the advice
|
|
;; will be `activate'ed immediately. Advice names are global symbols, hence,
|
|
;; the name space conventions used for function names should be applied. All
|
|
;; advice names in this tutorial will be prefixed with `fg' for `Foo Games'
|
|
;; (because everybody has the right to be inconsistent all the function names
|
|
;; used in this tutorial do NOT follow this convention).
|
|
;;
|
|
;; In the body of an advice we can refer to the argument variables of the
|
|
;; original function by name. Here we add 1 to X so the effect of calling
|
|
;; `foo' will be to actually add 2. All of the advice definitions below only
|
|
;; have one body form for simplicity, but there is no restriction to that
|
|
;; extent. Every piece of advice can have a documentation string which will
|
|
;; be combined with the documentation of the original function.
|
|
;;
|
|
;; (defadvice foo (before fg-add2 first activate)
|
|
;; "Add 2 to X."
|
|
;; (setq x (1+ x)))
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; 5
|
|
;;
|
|
;; @@ Specifying the position of an advice:
|
|
;; ========================================
|
|
;; Now we define the second before advice which will cancel the effect of
|
|
;; the previous advice. This time we specify the position as 0 which is
|
|
;; equivalent to `first'. A number can be used to specify the zero-based
|
|
;; position of an advice among the list of advices in the same class. This
|
|
;; time we already have one before advice hence the position specification
|
|
;; actually has an effect. So, after the following definition the position
|
|
;; of the previous advice will be 1 even though we specified it with `first'
|
|
;; above, the reason for this is that the position argument is relative to
|
|
;; the currently defined pieces of advice which by now has changed.
|
|
;;
|
|
;; (defadvice foo (before fg-cancel-add2 0 activate)
|
|
;; "Again only add 1 to X."
|
|
;; (setq x (1- x)))
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; 4
|
|
;;
|
|
;; @@ Redefining a piece of advice:
|
|
;; ================================
|
|
;; Now we define an advice with the same class and same name but with a
|
|
;; different position. Defining an advice in a class in which an advice with
|
|
;; that name already exists is interpreted as a redefinition of that
|
|
;; particular advice, in which case the position argument will be ignored
|
|
;; and the previous position of the redefined piece of advice is used.
|
|
;; Advice flags can be specified with non-ambiguous initial substrings, hence,
|
|
;; from now on we'll use `act' instead of the verbose `activate'.
|
|
;;
|
|
;; (defadvice foo (before fg-cancel-add2 last act)
|
|
;; "Again only add 1 to X."
|
|
;; (setq x (1- x)))
|
|
;; foo
|
|
;;
|
|
;; @@ Assembly of advised documentation:
|
|
;; =====================================
|
|
;; The documentation strings of the various pieces of advice are assembled
|
|
;; in order which shows that advice `fg-cancel-add2' is still the first
|
|
;; `before' advice even though we specified position `last' above:
|
|
;;
|
|
;; (documentation 'foo)
|
|
;; "Add 1 to X.
|
|
;;
|
|
;; This function is advised with the following advice(s):
|
|
;;
|
|
;; fg-cancel-add2 (before):
|
|
;; Again only add 1 to X.
|
|
;;
|
|
;; fg-add2 (before):
|
|
;; Add 2 to X."
|
|
;;
|
|
;; @@ Advising interactive behavior:
|
|
;; =================================
|
|
;; We can make a function interactive (or change its interactive behavior)
|
|
;; by specifying an interactive form in one of the before or around
|
|
;; advices (there could also be body forms in this advice). The particular
|
|
;; definition always assigns 5 as an argument to X which gives us 6 as a
|
|
;; result when we call foo interactively:
|
|
;;
|
|
;; (defadvice foo (before fg-inter last act)
|
|
;; "Use 5 as argument when called interactively."
|
|
;; (interactive (list 5)))
|
|
;; foo
|
|
;;
|
|
;; (call-interactively 'foo)
|
|
;; 6
|
|
;;
|
|
;; If more than one advice have an interactive declaration, then the one of
|
|
;; the advice with the smallest position will be used (before advices go
|
|
;; before around and after advices), hence, the declaration below does
|
|
;; not have any effect:
|
|
;;
|
|
;; (defadvice foo (before fg-inter2 last act)
|
|
;; (interactive (list 6)))
|
|
;; foo
|
|
;;
|
|
;; (call-interactively 'foo)
|
|
;; 6
|
|
;;
|
|
;; Let's have a look at what the definition of `foo' looks like now
|
|
;; (indentation added by hand for legibility):
|
|
;;
|
|
;; (symbol-function 'foo)
|
|
;; (lambda (x)
|
|
;; "$ad-doc: foo$"
|
|
;; (interactive (list 5))
|
|
;; (let (ad-return-value)
|
|
;; (setq x (1- x))
|
|
;; (setq x (1+ x))
|
|
;; (setq ad-return-value (ad-Orig-foo x))
|
|
;; ad-return-value))
|
|
;;
|
|
;; @@ Around advices:
|
|
;; ==================
|
|
;; Now we'll try some `around' advices. An around advice is a wrapper around
|
|
;; the original definition. It can shadow or establish bindings for the
|
|
;; original definition, and it can look at and manipulate the value returned
|
|
;; by the original function. The position of the special keyword `ad-do-it'
|
|
;; specifies where the code of the original function will be executed. The
|
|
;; keyword can appear multiple times which will result in multiple calls of
|
|
;; the original function in the resulting advised code. Note, that if we don't
|
|
;; specify a position argument (i.e., `first', `last' or a number), then
|
|
;; `first' (or 0) is the default):
|
|
;;
|
|
;; (defadvice foo (around fg-times-2 act)
|
|
;; "First double X."
|
|
;; (let ((x (* x 2)))
|
|
;; ad-do-it))
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; 7
|
|
;;
|
|
;; Around advices are assembled like onion skins where the around advice
|
|
;; with position 0 is the outermost skin and the advice at the last position
|
|
;; is the innermost skin which is directly wrapped around the call of the
|
|
;; original definition of the function. Hence, after the next `defadvice' we
|
|
;; will first multiply X by 2 then add 1 and then call the original
|
|
;; definition (i.e., add 1 again):
|
|
;;
|
|
;; (defadvice foo (around fg-add-1 last act)
|
|
;; "Add 1 to X."
|
|
;; (let ((x (1+ x)))
|
|
;; ad-do-it))
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; 8
|
|
;;
|
|
;; Again, let's see what the definition of `foo' looks like so far:
|
|
;;
|
|
;; (symbol-function 'foo)
|
|
;; (lambda (x)
|
|
;; "$ad-doc: foo$"
|
|
;; (interactive (list 5))
|
|
;; (let (ad-return-value)
|
|
;; (setq x (1- x))
|
|
;; (setq x (1+ x))
|
|
;; (let ((x (* x 2)))
|
|
;; (let ((x (1+ x)))
|
|
;; (setq ad-return-value (ad-Orig-foo x))))
|
|
;; ad-return-value))
|
|
;;
|
|
;; @@ Controlling advice activation:
|
|
;; =================================
|
|
;; In every `defadvice' so far we have used the flag `activate' to activate
|
|
;; the advice immediately after its definition, and that's what we want in
|
|
;; most cases. However, if we define multiple pieces of advice for a single
|
|
;; function then activating every advice immediately is inefficient. A
|
|
;; better way to do this is to only activate the last defined advice.
|
|
;; For example:
|
|
;;
|
|
;; (defadvice foo (after fg-times-x)
|
|
;; "Multiply the result with X."
|
|
;; (setq ad-return-value (* ad-return-value x)))
|
|
;; foo
|
|
;;
|
|
;; This still yields the same result as before:
|
|
;; (foo 3)
|
|
;; 8
|
|
;;
|
|
;; Now we define another advice and activate which will also activate the
|
|
;; previous advice `fg-times-x'. Note the use of the special variable
|
|
;; `ad-return-value' in the body of the advice which is set to the result of
|
|
;; the original function. If we change its value then the value returned by
|
|
;; the advised function will be changed accordingly:
|
|
;;
|
|
;; (defadvice foo (after fg-times-x-again act)
|
|
;; "Again multiply the result with X."
|
|
;; (setq ad-return-value (* ad-return-value x)))
|
|
;; foo
|
|
;;
|
|
;; Now the advices have an effect:
|
|
;;
|
|
;; (foo 3)
|
|
;; 72
|
|
;;
|
|
;; @@ Protecting advice execution:
|
|
;; ===============================
|
|
;; Once in a while we define an advice to perform some cleanup action,
|
|
;; for example:
|
|
;;
|
|
;; (defadvice foo (after fg-cleanup last act)
|
|
;; "Do some cleanup."
|
|
;; (print "Let's clean up now!"))
|
|
;; foo
|
|
;;
|
|
;; However, in case of an error the cleanup won't be performed:
|
|
;;
|
|
;; (condition-case error
|
|
;; (foo t)
|
|
;; (error 'error-in-foo))
|
|
;; error-in-foo
|
|
;;
|
|
;; To make sure a certain piece of advice gets executed even if some error or
|
|
;; non-local exit occurred in any preceding code, we can protect it by using
|
|
;; the `protect' keyword. (if any of the around advices is protected then the
|
|
;; whole around advice onion will be protected):
|
|
;;
|
|
;; (defadvice foo (after fg-cleanup prot act)
|
|
;; "Do some protected cleanup."
|
|
;; (print "Let's clean up now!"))
|
|
;; foo
|
|
;;
|
|
;; Now the cleanup form will be executed even in case of an error:
|
|
;;
|
|
;; (condition-case error
|
|
;; (foo t)
|
|
;; (error 'error-in-foo))
|
|
;; "Let's clean up now!"
|
|
;; error-in-foo
|
|
;;
|
|
;; Again, let's see what `foo' looks like:
|
|
;;
|
|
;; (symbol-function 'foo)
|
|
;; (lambda (x)
|
|
;; "$ad-doc: foo$"
|
|
;; (interactive (list 5))
|
|
;; (let (ad-return-value)
|
|
;; (unwind-protect
|
|
;; (progn (setq x (1- x))
|
|
;; (setq x (1+ x))
|
|
;; (let ((x (* x 2)))
|
|
;; (let ((x (1+ x)))
|
|
;; (setq ad-return-value (ad-Orig-foo x))))
|
|
;; (setq ad-return-value (* ad-return-value x))
|
|
;; (setq ad-return-value (* ad-return-value x)))
|
|
;; (print "Let's clean up now!"))
|
|
;; ad-return-value))
|
|
;;
|
|
;; @@ Compilation of advised definitions:
|
|
;; ======================================
|
|
;; Finally, we can specify the `compile' keyword in a `defadvice' to say
|
|
;; that we want the resulting advised function to be byte-compiled
|
|
;; (`compile' will be ignored unless we also specified `activate'):
|
|
;;
|
|
;; (defadvice foo (after fg-cleanup prot act comp)
|
|
;; "Do some protected cleanup."
|
|
;; (print "Let's clean up now!"))
|
|
;; foo
|
|
;;
|
|
;; Now `foo' is byte-compiled:
|
|
;;
|
|
;; (symbol-function 'foo)
|
|
;; (lambda (x)
|
|
;; "$ad-doc: foo$"
|
|
;; (interactive (byte-code "....." [5] 1))
|
|
;; (byte-code "....." [ad-return-value x nil ((byte-code "....." [print "Let's clean up now!"] 2)) * 2 ad-Orig-foo] 6))
|
|
;;
|
|
;; (foo 3)
|
|
;; "Let's clean up now!"
|
|
;; 72
|
|
;;
|
|
;; @@ Enabling and disabling pieces of advice:
|
|
;; ===========================================
|
|
;; Once in a while it is desirable to temporarily disable a piece of advice
|
|
;; so that it won't be considered during activation, for example, if two
|
|
;; different packages advise the same function and one wants to temporarily
|
|
;; neutralize the effect of the advice of one of the packages.
|
|
;;
|
|
;; The following disables the after advice `fg-times-x' in the function `foo'.
|
|
;; All that does is to change a flag for this particular advice. All the
|
|
;; other information defining it will be left unchanged (e.g., its relative
|
|
;; position in this advice class, etc.).
|
|
;;
|
|
;; (ad-disable-advice 'foo 'after 'fg-times-x)
|
|
;; nil
|
|
;;
|
|
;; For this to have an effect we have to activate `foo':
|
|
;;
|
|
;; (ad-activate 'foo)
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; "Let's clean up now!"
|
|
;; 24
|
|
;;
|
|
;; If we want to disable all multiplication advices in `foo' we can use a
|
|
;; regular expression that matches the names of such advices. Actually, any
|
|
;; advice name that contains a match for the regular expression will be
|
|
;; called a match. A special advice class `any' can be used to consider
|
|
;; all advice classes:
|
|
;;
|
|
;; (ad-disable-advice 'foo 'any "^fg-.*times")
|
|
;; nil
|
|
;;
|
|
;; (ad-activate 'foo)
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; "Let's clean up now!"
|
|
;; 5
|
|
;;
|
|
;; To enable the disabled advice we could use either `ad-enable-advice'
|
|
;; similar to `ad-disable-advice', or as an alternative `ad-enable-regexp'
|
|
;; which will enable matching advices in ALL currently advised functions.
|
|
;; Hence, this can be used to dis/enable advices made by a particular
|
|
;; package to a set of functions as long as that package obeys standard
|
|
;; advice name conventions. We prefixed all advice names with `fg-', hence
|
|
;; the following will do the trick (`ad-enable-regexp' returns the number
|
|
;; of matched advices):
|
|
;;
|
|
;; (ad-enable-regexp "^fg-")
|
|
;; 9
|
|
;;
|
|
;; The following will activate all currently active advised functions that
|
|
;; contain some advice matched by the regular expression. This is a save
|
|
;; way to update the activation of advised functions whose advice changed
|
|
;; in some way or other without accidentally also activating currently
|
|
;; deactivated functions:
|
|
;;
|
|
;; (ad-update-regexp "^fg-")
|
|
;; nil
|
|
;;
|
|
;; (foo 3)
|
|
;; "Let's clean up now!"
|
|
;; 72
|
|
;;
|
|
;; Another use for the dis/enablement mechanism is to define a piece of advice
|
|
;; and keep it "dormant" until a particular condition is satisfied, i.e., until
|
|
;; then the advice will not be used during activation. The `disable' flag lets
|
|
;; one do that with `defadvice':
|
|
;;
|
|
;; (defadvice foo (before fg-1-more dis)
|
|
;; "Add yet 1 more."
|
|
;; (setq x (1+ x)))
|
|
;; foo
|
|
;;
|
|
;; (ad-activate 'foo)
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; "Let's clean up now!"
|
|
;; 72
|
|
;;
|
|
;; (ad-enable-advice 'foo 'before 'fg-1-more)
|
|
;; nil
|
|
;;
|
|
;; (ad-activate 'foo)
|
|
;; foo
|
|
;;
|
|
;; (foo 3)
|
|
;; "Let's clean up now!"
|
|
;; 160
|
|
;;
|
|
;; @@ Caching:
|
|
;; ===========
|
|
;; Advised definitions get cached to allow efficient activation/deactivation
|
|
;; without having to reconstruct them if nothing in the advice-info of a
|
|
;; function has changed. The following idiom can be used to temporarily
|
|
;; deactivate functions that have a piece of advice defined by a certain
|
|
;; package (we save the old definition to check out caching):
|
|
;;
|
|
;; (setq old-definition (symbol-function 'foo))
|
|
;; (lambda (x) ....)
|
|
;;
|
|
;; (ad-deactivate-regexp "^fg-")
|
|
;; nil
|
|
;;
|
|
;; (foo 3)
|
|
;; 4
|
|
;;
|
|
;; (ad-activate-regexp "^fg-")
|
|
;; nil
|
|
;;
|
|
;; (eq old-definition (symbol-function 'foo))
|
|
;; t
|
|
;;
|
|
;; (foo 3)
|
|
;; "Let's clean up now!"
|
|
;; 160
|
|
;;
|
|
;; @@ Forward advice:
|
|
;; ==================
|
|
;; To enable automatic activation of forward advice we first have to set
|
|
;; `ad-activate-on-definition' to t and restart advice:
|
|
;;
|
|
;; (setq ad-activate-on-definition t)
|
|
;; t
|
|
;;
|
|
;; (ad-start-advice)
|
|
;; (ad-activate-defined-function)
|
|
;;
|
|
;; Let's define a piece of advice for an undefined function:
|
|
;;
|
|
;; (defadvice bar (before fg-sub-1-more act)
|
|
;; "Subtract one more from X."
|
|
;; (setq x (1- x)))
|
|
;; bar
|
|
;;
|
|
;; `bar' is not yet defined:
|
|
;; (fboundp 'bar)
|
|
;; nil
|
|
;;
|
|
;; Now we define it and the forward advice will get activated (only because
|
|
;; `ad-activate-on-definition' was t when we started advice above with
|
|
;; `ad-start-advice'):
|
|
;;
|
|
;; (defun bar (x)
|
|
;; "Subtract 1 from X."
|
|
;; (1- x))
|
|
;; bar
|
|
;;
|
|
;; (bar 4)
|
|
;; 2
|
|
;;
|
|
;; Redefinition will activate any available advice if the value of
|
|
;; `ad-redefinition-action' is either `warn', `accept' or `discard':
|
|
;;
|
|
;; (defun bar (x)
|
|
;; "Subtract 2 from X."
|
|
;; (- x 2))
|
|
;; bar
|
|
;;
|
|
;; (bar 4)
|
|
;; 1
|
|
;;
|
|
;; @@ Preactivation:
|
|
;; =================
|
|
;; Constructing advised definitions is moderately expensive, hence, it is
|
|
;; desirable to have a way to construct them at byte-compile time.
|
|
;; Preactivation is a mechanism that allows one to do that.
|
|
;;
|
|
;; (defun fie (x)
|
|
;; "Multiply X by 2."
|
|
;; (* x 2))
|
|
;; fie
|
|
;;
|
|
;; (defadvice fie (before fg-times-4 preact)
|
|
;; "Multiply X by 4."
|
|
;; (setq x (* x 2)))
|
|
;; fie
|
|
;;
|
|
;; This advice did not affect `fie'...
|
|
;;
|
|
;; (fie 2)
|
|
;; 4
|
|
;;
|
|
;; ...but it constructed a cached definition that will be used once `fie' gets
|
|
;; activated as long as its current advice state is the same as it was during
|
|
;; preactivation:
|
|
;;
|
|
;; (setq cached-definition (ad-get-cache-definition 'fie))
|
|
;; (lambda (x) ....)
|
|
;;
|
|
;; (ad-activate 'fie)
|
|
;; fie
|
|
;;
|
|
;; (eq cached-definition (symbol-function 'fie))
|
|
;; t
|
|
;;
|
|
;; (fie 2)
|
|
;; 8
|
|
;;
|
|
;; If you put a preactivating `defadvice' into a Lisp file that gets byte-
|
|
;; compiled then the constructed advised definition will get compiled by
|
|
;; the byte-compiler. For that to occur in a v18 emacs you have to put the
|
|
;; `defadvice' inside a `defun' because the v18 compiler does not compile
|
|
;; top-level forms other than `defun' or `defmacro', for example,
|
|
;;
|
|
;; (defun fg-defadvice-fum ()
|
|
;; (defadvice fum (before fg-times-4 preact act)
|
|
;; "Multiply X by 4."
|
|
;; (setq x (* x 2))))
|
|
;; fg-defadvice-fum
|
|
;;
|
|
;; So far, no `defadvice' for `fum' got executed, but when we compile
|
|
;; `fg-defadvice-fum' the `defadvice' will be expanded by the byte compiler.
|
|
;; In order for preactivation to be effective we have to have a proper
|
|
;; definition of `fum' around at preactivation time, hence, we define it now:
|
|
;;
|
|
;; (defun fum (x)
|
|
;; "Multiply X by 2."
|
|
;; (* x 2))
|
|
;; fum
|
|
;;
|
|
;; Now we compile the defining function which will construct an advised
|
|
;; definition during expansion of the `defadvice', compile it and store it
|
|
;; as part of the compiled `fg-defadvice-fum':
|
|
;;
|
|
;; (ad-compile-function 'fg-defadvice-fum)
|
|
;; (lambda nil (byte-code ...))
|
|
;;
|
|
;; `fum' is still completely unaffected:
|
|
;;
|
|
;; (fum 2)
|
|
;; 4
|
|
;;
|
|
;; (ad-get-advice-info 'fum)
|
|
;; nil
|
|
;;
|
|
;; (fg-defadvice-fum)
|
|
;; fum
|
|
;;
|
|
;; Now the advised version of `fum' is compiled because the compiled definition
|
|
;; constructed during preactivation was used, even though we did not specify
|
|
;; the `compile' flag:
|
|
;;
|
|
;; (symbol-function 'fum)
|
|
;; (lambda (x)
|
|
;; "$ad-doc: fum$"
|
|
;; (byte-code "....." [ad-return-value x nil * 2 ad-Orig-fum] 4))
|
|
;;
|
|
;; (fum 2)
|
|
;; 8
|
|
;;
|
|
;; A preactivated definition will only be used if it matches the current
|
|
;; function definition and advice information. If it does not match it
|
|
;; will simply be discarded and a new advised definition will be constructed
|
|
;; from scratch. For example, let's first remove all advice-info for `fum':
|
|
;;
|
|
;; (ad-unadvise 'fum)
|
|
;; (("fie") ("bar") ("foo") ...)
|
|
;;
|
|
;; And now define a new piece of advice:
|
|
;;
|
|
;; (defadvice fum (before fg-interactive act)
|
|
;; "Make fum interactive."
|
|
;; (interactive "nEnter x: "))
|
|
;; fum
|
|
;;
|
|
;; When we now try to use a preactivation it will not be used because the
|
|
;; current advice state is different from the one at preactivation time. This
|
|
;; is no tragedy, everything will work as expected just not as efficient,
|
|
;; because a new advised definition has to be constructed from scratch:
|
|
;;
|
|
;; (fg-defadvice-fum)
|
|
;; fum
|
|
;;
|
|
;; A new uncompiled advised definition got constructed:
|
|
;;
|
|
;; (ad-compiled-p (symbol-function 'fum))
|
|
;; nil
|
|
;;
|
|
;; (fum 2)
|
|
;; 8
|
|
;;
|
|
;; MORAL: To get all the efficiency out of preactivation the function
|
|
;; definition and advice state at preactivation time must be the same as the
|
|
;; state at activation time. Preactivation does work with forward advice, all
|
|
;; that's necessary is that the definition of the forward advised function is
|
|
;; available when the `defadvice' with the preactivation gets compiled.
|
|
;;
|
|
;; @@ Portable argument access:
|
|
;; ============================
|
|
;; So far, we always used the actual argument variable names to access an
|
|
;; argument in a piece of advice. For many advice applications this is
|
|
;; perfectly ok and keeps advices simple. However, it decreases portability
|
|
;; of advices because it assumes specific argument variable names. For example,
|
|
;; if one advises a subr such as `eval-region' which then gets redefined by
|
|
;; some package (e.g., edebug) into a function with different argument names,
|
|
;; then a piece of advice written for `eval-region' that was written with
|
|
;; the subr arguments in mind will break. Similar situations arise when one
|
|
;; switches between major Emacs versions, e.g., certain subrs in v18 are
|
|
;; functions in v19 and vice versa. Also, in v19s subr argument lists
|
|
;; are available and will be used, while they are not available in v18.
|
|
;;
|
|
;; Argument access text macros allow one to access arguments of an advised
|
|
;; function in a portable way without having to worry about all these
|
|
;; possibilities. These macros will be translated into the proper access forms
|
|
;; at activation time, hence, argument access will be as efficient as if
|
|
;; the arguments had been used directly in the definition of the advice.
|
|
;;
|
|
;; (defun fuu (x y z)
|
|
;; "Add 3 numbers."
|
|
;; (+ x y z))
|
|
;; fuu
|
|
;;
|
|
;; (fuu 1 1 1)
|
|
;; 3
|
|
;;
|
|
;; Argument access macros specify actual arguments at a certain position.
|
|
;; Position 0 access the first actual argument, position 1 the second etc.
|
|
;; For example, the following advice adds 1 to each of the 3 arguments:
|
|
;;
|
|
;; (defadvice fuu (before fg-add-1-to-all act)
|
|
;; "Adds 1 to all arguments."
|
|
;; (ad-set-arg 0 (1+ (ad-get-arg 0)))
|
|
;; (ad-set-arg 1 (1+ (ad-get-arg 1)))
|
|
;; (ad-set-arg 2 (1+ (ad-get-arg 2))))
|
|
;; fuu
|
|
;;
|
|
;; (fuu 1 1 1)
|
|
;; 6
|
|
;;
|
|
;; Now suppose somebody redefines `fuu' with a rest argument. Our advice
|
|
;; will still work because we used access macros (note, that automatic
|
|
;; advice activation is still in effect, hence, the redefinition of `fuu'
|
|
;; will automatically activate all its advice):
|
|
;;
|
|
;; (defun fuu (&rest numbers)
|
|
;; "Add NUMBERS."
|
|
;; (apply '+ numbers))
|
|
;; fuu
|
|
;;
|
|
;; (fuu 1 1 1)
|
|
;; 6
|
|
;;
|
|
;; (fuu 1 1 1 1 1 1)
|
|
;; 9
|
|
;;
|
|
;; What's important to notice is that argument access macros access actual
|
|
;; arguments regardless of how they got distributed onto argument variables.
|
|
;; In Emacs Lisp the semantics of an actual argument is determined purely
|
|
;; by position, hence, as long as nobody changes the semantics of what a
|
|
;; certain actual argument at a certain position means the access macros
|
|
;; will do the right thing.
|
|
;;
|
|
;; Because of &rest arguments we need a second kind of access macro that
|
|
;; can access all actual arguments starting from a certain position:
|
|
;;
|
|
;; (defadvice fuu (before fg-print-args act)
|
|
;; "Print all arguments."
|
|
;; (print (ad-get-args 0)))
|
|
;; fuu
|
|
;;
|
|
;; (fuu 1 2 3 4 5)
|
|
;; (1 2 3 4 5)
|
|
;; 18
|
|
;;
|
|
;; (defadvice fuu (before fg-set-args act)
|
|
;; "Swaps 2nd and 3rd arg and discards all the rest."
|
|
;; (ad-set-args 1 (list (ad-get-arg 2) (ad-get-arg 1))))
|
|
;; fuu
|
|
;;
|
|
;; (fuu 1 2 3 4 4 4 4 4 4)
|
|
;; (1 3 2)
|
|
;; 9
|
|
;;
|
|
;; (defun fuu (x y z)
|
|
;; "Add 3 numbers."
|
|
;; (+ x y z))
|
|
;;
|
|
;; (fuu 1 2 3)
|
|
;; (1 3 2)
|
|
;; 9
|
|
;;
|
|
;; @@ Defining the argument list of an advised function:
|
|
;; =====================================================
|
|
;; Once in a while it might be desirable to advise a function and additionally
|
|
;; give it an extra argument that controls the advised code, for example, one
|
|
;; might want to make an interactive function sensitive to a prefix argument.
|
|
;; For such cases `defadvice' allows the specification of an argument list
|
|
;; for the advised function. Similar to the redefinition of interactive
|
|
;; behavior, the first argument list specification found in the list of before/
|
|
;; around/after advices will be used. Of course, the specified argument list
|
|
;; should be downward compatible with the original argument list, otherwise
|
|
;; functions that call the advised function with the original argument list
|
|
;; in mind will break.
|
|
;;
|
|
;; (defun fii (x)
|
|
;; "Add 1 to X."
|
|
;; (1+ x))
|
|
;; fii
|
|
;;
|
|
;; Now we advise `fii' to use an optional second argument that controls the
|
|
;; amount of incrementation. A list following the (optional) position
|
|
;; argument of the advice will be interpreted as an argument list
|
|
;; specification. This means you cannot specify an empty argument list, and
|
|
;; why would you want to anyway?
|
|
;;
|
|
;; (defadvice fii (before fg-inc-x (x &optional incr) act)
|
|
;; "Increment X by INCR (default is 1)."
|
|
;; (setq x (+ x (1- (or incr 1)))))
|
|
;; fii
|
|
;;
|
|
;; (fii 3)
|
|
;; 4
|
|
;;
|
|
;; (fii 3 2)
|
|
;; 5
|
|
;;
|
|
;; @@ Specifying argument lists of subrs:
|
|
;; ======================================
|
|
;; The argument lists of subrs cannot be determined directly from Lisp.
|
|
;; This means that Advice has to use `(&rest ad-subr-args)' as the
|
|
;; argument list of the advised subr which is not very efficient. In Lemacs
|
|
;; subr argument lists can be determined from their documentation string, in
|
|
;; Emacs-19 this is the case for some but not all subrs. To accommodate
|
|
;; for the cases where the argument lists cannot be determined (e.g., in a
|
|
;; v18 Emacs) Advice comes with a specification mechanism that allows the
|
|
;; advice programmer to tell advice what the argument list of a certain subr
|
|
;; really is.
|
|
;;
|
|
;; In a v18 Emacs the following will return the &rest idiom:
|
|
;;
|
|
;; (ad-arglist (symbol-function 'car))
|
|
;; (&rest ad-subr-args)
|
|
;;
|
|
;; To tell advice what the argument list of `car' really is we
|
|
;; can do the following:
|
|
;;
|
|
;; (ad-define-subr-args 'car '(list))
|
|
;; ((list))
|
|
;;
|
|
;; Now `ad-arglist' will return the proper argument list (this method is
|
|
;; actually used by advice itself for the advised definition of `fset'):
|
|
;;
|
|
;; (ad-arglist (symbol-function 'car))
|
|
;; (list)
|
|
;;
|
|
;; The defined argument list will be stored on the property list of the
|
|
;; subr name symbol. When advice looks for a subr argument list it first
|
|
;; checks for a definition on the property list, if that fails it tries
|
|
;; to infer it from the documentation string and caches it on the property
|
|
;; list if it was successful, otherwise `(&rest ad-subr-args)' will be used.
|
|
;;
|
|
;; @@ Advising interactive subrs:
|
|
;; ==============================
|
|
;; For the most part there is no difference between advising functions and
|
|
;; advising subrs. There is one situation though where one might have to write
|
|
;; slightly different advice code for subrs than for functions. This case
|
|
;; arises when one wants to access subr arguments in a before/around advice
|
|
;; when the arguments were determined by an interactive call to the subr.
|
|
;; Advice cannot determine what `interactive' form determines the interactive
|
|
;; behavior of the subr, hence, when it calls the original definition in an
|
|
;; interactive subr invocation it has to use `call-interactively' to generate
|
|
;; the proper interactive behavior. Thus up to that call the arguments of the
|
|
;; interactive subr will be nil. For example, the following advice for
|
|
;; `kill-buffer' will not work in an interactive invocation...
|
|
;;
|
|
;; (defadvice kill-buffer (before fg-kill-buffer-hook first act preact comp)
|
|
;; (my-before-kill-buffer-hook (ad-get-arg 0)))
|
|
;; kill-buffer
|
|
;;
|
|
;; ...because the buffer argument will be nil in that case. The way out of
|
|
;; this dilemma is to provide an `interactive' specification that mirrors
|
|
;; the interactive behavior of the unadvised subr, for example, the following
|
|
;; will do the right thing even when `kill-buffer' is called interactively:
|
|
;;
|
|
;; (defadvice kill-buffer (before fg-kill-buffer-hook first act preact comp)
|
|
;; (interactive "bKill buffer: ")
|
|
;; (my-before-kill-buffer-hook (ad-get-arg 0)))
|
|
;; kill-buffer
|
|
;;
|
|
;; @@ Advising macros:
|
|
;; ===================
|
|
;; Advising macros is slightly different because there are two significant
|
|
;; time points in the invocation of a macro: Expansion and evaluation time.
|
|
;; For an advised macro instead of evaluating the original definition we
|
|
;; use `macroexpand', that is, changing argument values and binding
|
|
;; environments by pieces of advice has an affect during macro expansion
|
|
;; but not necessarily during evaluation. In particular, any side effects
|
|
;; of pieces of advice will occur during macro expansion. To also affect
|
|
;; the behavior during evaluation time one has to change the value of
|
|
;; `ad-return-value' in a piece of after advice. For example:
|
|
;;
|
|
;; (defmacro foom (x)
|
|
;; (` (list (, x))))
|
|
;; foom
|
|
;;
|
|
;; (foom '(a))
|
|
;; ((a))
|
|
;;
|
|
;; (defadvice foom (before fg-print-x act)
|
|
;; "Print the value of X."
|
|
;; (print x))
|
|
;; foom
|
|
;;
|
|
;; The following works as expected because evaluation immediately follows
|
|
;; macro expansion:
|
|
;;
|
|
;; (foom '(a))
|
|
;; (quote (a))
|
|
;; ((a))
|
|
;;
|
|
;; However, the printing happens during expansion (or byte-compile) time:
|
|
;;
|
|
;; (macroexpand '(foom '(a)))
|
|
;; (quote (a))
|
|
;; (list (quote (a)))
|
|
;;
|
|
;; If we want it to happen during evaluation time we have to do the
|
|
;; following (first remove the old advice):
|
|
;;
|
|
;; (ad-remove-advice 'foom 'before 'fg-print-x)
|
|
;; nil
|
|
;;
|
|
;; (defadvice foom (after fg-print-x act)
|
|
;; "Print the value of X."
|
|
;; (setq ad-return-value
|
|
;; (` (progn (print (, x))
|
|
;; (, ad-return-value)))))
|
|
;; foom
|
|
;;
|
|
;; (macroexpand '(foom '(a)))
|
|
;; (progn (print (quote (a))) (list (quote (a))))
|
|
;;
|
|
;; (foom '(a))
|
|
;; (a)
|
|
;; ((a))
|
|
;;
|
|
;; While this method might seem somewhat cumbersome, it is very general
|
|
;; because it allows one to influence macro expansion as well as evaluation.
|
|
;; In general, advising macros should be a rather rare activity anyway, in
|
|
;; particular, because compile-time macro expansion takes away a lot of the
|
|
;; flexibility and effectiveness of the advice mechanism. Macros that were
|
|
;; compile-time expanded before the advice was activated will of course never
|
|
;; exhibit the advised behavior.
|
|
;;
|
|
;; @@ Advising special forms:
|
|
;; ==========================
|
|
;; Now for something that should be even more rare than advising macros:
|
|
;; Advising special forms. Because special forms are irregular in their
|
|
;; argument evaluation behavior (e.g., `setq' evaluates the second but not
|
|
;; the first argument) they have to be advised into macros. A dangerous
|
|
;; consequence of this is that the byte-compiler will not recognize them
|
|
;; as special forms anymore (well, in most cases) and use their expansion
|
|
;; rather than the proper byte-code. Also, because the original definition
|
|
;; of a special form cannot be `funcall'ed, `eval' has to be used instead
|
|
;; which is less efficient.
|
|
;;
|
|
;; MORAL: Do not advise special forms unless you are completely sure about
|
|
;; what you are doing (some of the forward advice behavior is
|
|
;; implemented via advice of the special forms `defun' and `defmacro').
|
|
;; As a safety measure one should always do `ad-deactivate-all' before
|
|
;; one byte-compiles a file to avoid any interference of advised
|
|
;; special forms.
|
|
;;
|
|
;; Apart from the safety concerns advising special forms is not any different
|
|
;; from advising plain functions or subrs.
|
|
|
|
|
|
;;; Code:
|
|
|
|
;; @ Advice implementation:
|
|
;; ========================
|
|
|
|
;; @@ Compilation idiosyncrasies:
|
|
;; ==============================
|
|
|
|
;; `defadvice' expansion needs quite a few advice functions and variables,
|
|
;; hence, I need to preload the file before it can be compiled. To avoid
|
|
;; interference of bogus compiled files I always preload the source file:
|
|
(provide 'advice-preload)
|
|
;; During a normal load this is a noop:
|
|
(require 'advice-preload "advice.el")
|
|
|
|
|
|
;; @@ Variable definitions:
|
|
;; ========================
|
|
|
|
(defgroup advice nil
|
|
"An overloading mechanism for Emacs Lisp functions."
|
|
:prefix "ad-"
|
|
:link '(custom-manual "(elisp)Advising Functions")
|
|
:group 'lisp)
|
|
|
|
(defconst ad-version "2.14")
|
|
|
|
;;;###autoload
|
|
(defcustom ad-redefinition-action 'warn
|
|
"*Defines what to do with redefinitions during Advice de/activation.
|
|
Redefinition occurs if a previously activated function that already has an
|
|
original definition associated with it gets redefined and then de/activated.
|
|
In such a case we can either accept the current definition as the new
|
|
original definition, discard the current definition and replace it with the
|
|
old original, or keep it and raise an error. The values `accept', `discard',
|
|
`error' or `warn' govern what will be done. `warn' is just like `accept' but
|
|
it additionally prints a warning message. All other values will be
|
|
interpreted as `error'."
|
|
:type '(choice (const accept) (const discard) (const warn)
|
|
(other :tag "error" error))
|
|
:group 'advice)
|
|
|
|
;;;###autoload
|
|
(defcustom ad-default-compilation-action 'maybe
|
|
"*Defines whether to compile advised definitions during activation.
|
|
A value of `always' will result in unconditional compilation, `never' will
|
|
always avoid compilation, `maybe' will compile if the byte-compiler is already
|
|
loaded, and `like-original' will compile if the original definition of the
|
|
advised function is compiled or a built-in function. Every other value will
|
|
be interpreted as `maybe'. This variable will only be considered if the
|
|
COMPILE argument of `ad-activate' was supplied as nil."
|
|
:type '(choice (const always) (const never) (const like-original)
|
|
(other :tag "maybe" maybe))
|
|
:group 'advice)
|
|
|
|
|
|
|
|
;; @@ Some utilities:
|
|
;; ==================
|
|
|
|
;; We don't want the local arguments to interfere with anything
|
|
;; referenced in the supplied functions => the cryptic casing:
|
|
(defun ad-substitute-tree (sUbTrEe-TeSt fUnCtIoN tReE)
|
|
"Substitute qualifying subTREEs with result of FUNCTION(subTREE).
|
|
Only proper subtrees are considered, for example, if TREE is (1 (2 (3)) 4)
|
|
then the subtrees will be 1 (2 (3)) 2 (3) 3 4, dotted structures are
|
|
allowed too. Once a qualifying subtree has been found its subtrees will
|
|
not be considered anymore. (ad-substitute-tree 'atom 'identity tree)
|
|
generates a copy of TREE."
|
|
(cond ((consp tReE)
|
|
(cons (if (funcall sUbTrEe-TeSt (car tReE))
|
|
(funcall fUnCtIoN (car tReE))
|
|
(if (consp (car tReE))
|
|
(ad-substitute-tree sUbTrEe-TeSt fUnCtIoN (car tReE))
|
|
(car tReE)))
|
|
(ad-substitute-tree sUbTrEe-TeSt fUnCtIoN (cdr tReE))))
|
|
((funcall sUbTrEe-TeSt tReE)
|
|
(funcall fUnCtIoN tReE))
|
|
(t tReE)))
|
|
|
|
;; this is just faster than `ad-substitute-tree':
|
|
(defun ad-copy-tree (tree)
|
|
"Return a copy of the list structure of TREE."
|
|
(cond ((consp tree)
|
|
(cons (ad-copy-tree (car tree))
|
|
(ad-copy-tree (cdr tree))))
|
|
(t tree)))
|
|
|
|
(defmacro ad-dolist (varform &rest body)
|
|
"A Common-Lisp-style dolist iterator with the following syntax:
|
|
|
|
(ad-dolist (VAR INIT-FORM [RESULT-FORM])
|
|
BODY-FORM...)
|
|
|
|
which will iterate over the list yielded by INIT-FORM binding VAR to the
|
|
current head at every iteration. If RESULT-FORM is supplied its value will
|
|
be returned at the end of the iteration, nil otherwise. The iteration can be
|
|
exited prematurely with `(ad-do-return [VALUE])'."
|
|
(let ((expansion
|
|
`(let ((ad-dO-vAr ,(car (cdr varform)))
|
|
,(car varform))
|
|
(while ad-dO-vAr
|
|
(setq ,(car varform) (car ad-dO-vAr))
|
|
,@body
|
|
;;work around a backquote bug:
|
|
;;(` ((,@ '(foo)) (bar))) => (append '(foo) '(((bar)))) wrong
|
|
;;(` ((,@ '(foo)) (, '(bar)))) => (append '(foo) (list '(bar)))
|
|
,'(setq ad-dO-vAr (cdr ad-dO-vAr)))
|
|
,(car (cdr (cdr varform))))))
|
|
;;ok, this wastes some cons cells but only during compilation:
|
|
(if (catch 'contains-return
|
|
(ad-substitute-tree
|
|
(function (lambda (subtree)
|
|
(cond ((eq (car-safe subtree) 'ad-dolist))
|
|
((eq (car-safe subtree) 'ad-do-return)
|
|
(throw 'contains-return t)))))
|
|
'identity body)
|
|
nil)
|
|
`(catch 'ad-dO-eXiT ,expansion)
|
|
expansion)))
|
|
|
|
(defmacro ad-do-return (value)
|
|
`(throw 'ad-dO-eXiT ,value))
|
|
|
|
(if (not (get 'ad-dolist 'lisp-indent-hook))
|
|
(put 'ad-dolist 'lisp-indent-hook 1))
|
|
|
|
|
|
;; @@ Save real definitions of subrs used by Advice:
|
|
;; =================================================
|
|
;; Advice depends on the real, unmodified functionality of various subrs,
|
|
;; we save them here so advised versions will not interfere (eventually,
|
|
;; we will save all subrs used in code generated by Advice):
|
|
|
|
(defmacro ad-save-real-definition (function)
|
|
(let ((saved-function (intern (format "ad-real-%s" function))))
|
|
;; Make sure the compiler is loaded during macro expansion:
|
|
(require 'byte-compile "bytecomp")
|
|
`(if (not (fboundp ',saved-function))
|
|
(progn (fset ',saved-function (symbol-function ',function))
|
|
;; Copy byte-compiler properties:
|
|
,@(if (get function 'byte-compile)
|
|
`((put ',saved-function 'byte-compile
|
|
',(get function 'byte-compile))))
|
|
,@(if (get function 'byte-opcode)
|
|
`((put ',saved-function 'byte-opcode
|
|
',(get function 'byte-opcode))))))))
|
|
|
|
(defun ad-save-real-definitions ()
|
|
;; Macro expansion will hardcode the values of the various byte-compiler
|
|
;; properties into the compiled version of this function such that the
|
|
;; proper values will be available at runtime without loading the compiler:
|
|
(ad-save-real-definition fset)
|
|
(ad-save-real-definition documentation))
|
|
|
|
(ad-save-real-definitions)
|
|
|
|
|
|
;; @@ Advice info access fns:
|
|
;; ==========================
|
|
|
|
;; Advice information for a particular function is stored on the
|
|
;; advice-info property of the function symbol. It is stored as an
|
|
;; alist of the following format:
|
|
;;
|
|
;; ((active . t/nil)
|
|
;; (before adv1 adv2 ...)
|
|
;; (around adv1 adv2 ...)
|
|
;; (after adv1 adv2 ...)
|
|
;; (activation adv1 adv2 ...)
|
|
;; (deactivation adv1 adv2 ...)
|
|
;; (origname . <symbol fbound to origdef>)
|
|
;; (cache . (<advised-definition> . <id>)))
|
|
|
|
;; List of currently advised though not necessarily activated functions
|
|
;; (this list is maintained as a completion table):
|
|
(defvar ad-advised-functions nil)
|
|
|
|
(defmacro ad-pushnew-advised-function (function)
|
|
"Add FUNCTION to `ad-advised-functions' unless its already there."
|
|
`(if (not (assoc (symbol-name ,function) ad-advised-functions))
|
|
(setq ad-advised-functions
|
|
(cons (list (symbol-name ,function))
|
|
ad-advised-functions))))
|
|
|
|
(defmacro ad-pop-advised-function (function)
|
|
"Remove FUNCTION from `ad-advised-functions'."
|
|
`(setq ad-advised-functions
|
|
(delq (assoc (symbol-name ,function) ad-advised-functions)
|
|
ad-advised-functions)))
|
|
|
|
(defmacro ad-do-advised-functions (varform &rest body)
|
|
"`ad-dolist'-style iterator that maps over `ad-advised-functions'.
|
|
\(ad-do-advised-functions (VAR [RESULT-FORM])
|
|
BODY-FORM...)
|
|
On each iteration VAR will be bound to the name of an advised function
|
|
\(a symbol)."
|
|
`(ad-dolist (,(car varform)
|
|
ad-advised-functions
|
|
,(car (cdr varform)))
|
|
(setq ,(car varform) (intern (car ,(car varform))))
|
|
,@body))
|
|
|
|
(if (not (get 'ad-do-advised-functions 'lisp-indent-hook))
|
|
(put 'ad-do-advised-functions 'lisp-indent-hook 1))
|
|
|
|
(defun ad-get-advice-info (function)
|
|
(get function 'ad-advice-info))
|
|
|
|
(defmacro ad-get-advice-info-macro (function)
|
|
`(get ,function 'ad-advice-info))
|
|
|
|
(defmacro ad-set-advice-info (function advice-info)
|
|
`(put ,function 'ad-advice-info ,advice-info))
|
|
|
|
(defmacro ad-copy-advice-info (function)
|
|
`(ad-copy-tree (get ,function 'ad-advice-info)))
|
|
|
|
(defmacro ad-is-advised (function)
|
|
"Return non-nil if FUNCTION has any advice info associated with it.
|
|
This does not mean that the advice is also active."
|
|
(list 'ad-get-advice-info-macro function))
|
|
|
|
(defun ad-initialize-advice-info (function)
|
|
"Initialize the advice info for FUNCTION.
|
|
Assumes that FUNCTION has not yet been advised."
|
|
(ad-pushnew-advised-function function)
|
|
(ad-set-advice-info function (list (cons 'active nil))))
|
|
|
|
(defmacro ad-get-advice-info-field (function field)
|
|
"Retrieve the value of the advice info FIELD of FUNCTION."
|
|
`(cdr (assq ,field (ad-get-advice-info-macro ,function))))
|
|
|
|
(defun ad-set-advice-info-field (function field value)
|
|
"Destructively modify VALUE of the advice info FIELD of FUNCTION."
|
|
(and (ad-is-advised function)
|
|
(cond ((assq field (ad-get-advice-info-macro function))
|
|
;; A field with that name is already present:
|
|
(rplacd (assq field (ad-get-advice-info-macro function)) value))
|
|
(t;; otherwise, create a new field with that name:
|
|
(nconc (ad-get-advice-info-macro function)
|
|
(list (cons field value)))))))
|
|
|
|
;; Don't make this a macro so we can use it as a predicate:
|
|
(defun ad-is-active (function)
|
|
"Return non-nil if FUNCTION is advised and activated."
|
|
(ad-get-advice-info-field function 'active))
|
|
|
|
|
|
;; @@ Access fns for single pieces of advice and related predicates:
|
|
;; =================================================================
|
|
|
|
(defun ad-make-advice (name protect enable definition)
|
|
"Constructs single piece of advice to be stored in some advice-info.
|
|
NAME should be a non-nil symbol, PROTECT and ENABLE should each be
|
|
either t or nil, and DEFINITION should be a list of the form
|
|
`(advice lambda ARGLIST [DOCSTRING] [INTERACTIVE-FORM] BODY...)'."
|
|
(list name protect enable definition))
|
|
|
|
;; ad-find-advice uses the alist structure directly ->
|
|
;; change if this data structure changes!!
|
|
(defmacro ad-advice-name (advice)
|
|
(list 'car advice))
|
|
(defmacro ad-advice-protected (advice)
|
|
(list 'nth 1 advice))
|
|
(defmacro ad-advice-enabled (advice)
|
|
(list 'nth 2 advice))
|
|
(defmacro ad-advice-definition (advice)
|
|
(list 'nth 3 advice))
|
|
|
|
(defun ad-advice-set-enabled (advice flag)
|
|
(rplaca (cdr (cdr advice)) flag))
|
|
|
|
(defun ad-class-p (thing)
|
|
(memq thing ad-advice-classes))
|
|
(defun ad-name-p (thing)
|
|
(and thing (symbolp thing)))
|
|
(defun ad-position-p (thing)
|
|
(or (natnump thing)
|
|
(memq thing '(first last))))
|
|
|
|
|
|
;; @@ Advice access functions:
|
|
;; ===========================
|
|
|
|
;; List of defined advice classes:
|
|
(defvar ad-advice-classes '(before around after activation deactivation))
|
|
|
|
(defun ad-has-enabled-advice (function class)
|
|
"True if at least one of FUNCTION's advices in CLASS is enabled."
|
|
(ad-dolist (advice (ad-get-advice-info-field function class))
|
|
(if (ad-advice-enabled advice) (ad-do-return t))))
|
|
|
|
(defun ad-has-redefining-advice (function)
|
|
"True if FUNCTION's advice info defines at least 1 redefining advice.
|
|
Redefining advices affect the construction of an advised definition."
|
|
(and (ad-is-advised function)
|
|
(or (ad-has-enabled-advice function 'before)
|
|
(ad-has-enabled-advice function 'around)
|
|
(ad-has-enabled-advice function 'after))))
|
|
|
|
(defun ad-has-any-advice (function)
|
|
"True if the advice info of FUNCTION defines at least one advice."
|
|
(and (ad-is-advised function)
|
|
(ad-dolist (class ad-advice-classes nil)
|
|
(if (ad-get-advice-info-field function class)
|
|
(ad-do-return t)))))
|
|
|
|
(defun ad-get-enabled-advices (function class)
|
|
"Return the list of enabled advices of FUNCTION in CLASS."
|
|
(let (enabled-advices)
|
|
(ad-dolist (advice (ad-get-advice-info-field function class))
|
|
(if (ad-advice-enabled advice)
|
|
(push advice enabled-advices)))
|
|
(reverse enabled-advices)))
|
|
|
|
|
|
;; @@ Dealing with automatic advice activation via `fset/defalias':
|
|
;; ================================================================
|
|
|
|
;; Since Emacs 19.26 the built-in versions of `fset' and `defalias'
|
|
;; take care of automatic advice activation, hence, we don't have to
|
|
;; hack it anymore by advising `fset/defun/defmacro/byte-code/etc'.
|
|
|
|
;; The functionality of the new `fset' is as follows:
|
|
;;
|
|
;; fset(sym,newdef)
|
|
;; assign NEWDEF to SYM
|
|
;; if (get SYM 'ad-advice-info)
|
|
;; ad-activate-internal(SYM, nil)
|
|
;; return (symbol-function SYM)
|
|
;;
|
|
;; Whether advised definitions created by automatic activations will be
|
|
;; compiled depends on the value of `ad-default-compilation-action'.
|
|
|
|
;; Since calling `ad-activate-internal' in the built-in definition of `fset' can
|
|
;; create major disasters we have to be a bit careful. One precaution is
|
|
;; to provide a dummy definition for `ad-activate-internal' which can be used to
|
|
;; turn off automatic advice activation (e.g., when `ad-stop-advice' or
|
|
;; `ad-recover-normality' are called). Another is to avoid recursive calls
|
|
;; to `ad-activate' by using `ad-with-auto-activation-disabled' where
|
|
;; appropriate, especially in a safe version of `fset'.
|
|
|
|
;; For now define `ad-activate-internal' to the dummy definition:
|
|
(defun ad-activate-internal (function &optional compile)
|
|
"Automatic advice activation is disabled. `ad-start-advice' enables it."
|
|
nil)
|
|
|
|
;; This is just a copy of the above:
|
|
(defun ad-activate-internal-off (function &optional compile)
|
|
"Automatic advice activation is disabled. `ad-start-advice' enables it."
|
|
nil)
|
|
|
|
;; This will be t for top-level calls to `ad-activate-internal-on':
|
|
(defvar ad-activate-on-top-level t)
|
|
|
|
(defmacro ad-with-auto-activation-disabled (&rest body)
|
|
`(let ((ad-activate-on-top-level nil))
|
|
,@body))
|
|
|
|
(defun ad-safe-fset (symbol definition)
|
|
"A safe `fset' which will never call `ad-activate-internal' recursively."
|
|
(ad-with-auto-activation-disabled
|
|
(ad-real-fset symbol definition)))
|
|
|
|
|
|
;; @@ Access functions for original definitions:
|
|
;; ============================================
|
|
;; The advice-info of an advised function contains its `origname' which is
|
|
;; a symbol that is fbound to the original definition available at the first
|
|
;; proper activation of the function after a valid re/definition. If the
|
|
;; original was defined via fcell indirection then `origname' will be defined
|
|
;; just so. Hence, to get hold of the actual original definition of a function
|
|
;; we need to use `ad-real-orig-definition'.
|
|
|
|
(defun ad-make-origname (function)
|
|
"Make name to be used to call the original FUNCTION."
|
|
(intern (format "ad-Orig-%s" function)))
|
|
|
|
(defmacro ad-get-orig-definition (function)
|
|
`(let ((origname (ad-get-advice-info-field ,function 'origname)))
|
|
(if (fboundp origname)
|
|
(symbol-function origname))))
|
|
|
|
(defmacro ad-set-orig-definition (function definition)
|
|
`(ad-safe-fset
|
|
(ad-get-advice-info-field function 'origname) ,definition))
|
|
|
|
(defmacro ad-clear-orig-definition (function)
|
|
`(fmakunbound (ad-get-advice-info-field ,function 'origname)))
|
|
|
|
|
|
;; @@ Interactive input functions:
|
|
;; ===============================
|
|
|
|
(defun ad-read-advised-function (&optional prompt predicate default)
|
|
"Read name of advised function with completion from the minibuffer.
|
|
An optional PROMPT will be used to prompt for the function. PREDICATE
|
|
plays the same role as for `try-completion' (which see). DEFAULT will
|
|
be returned on empty input (defaults to the first advised function for
|
|
which PREDICATE returns non-nil)."
|
|
(if (null ad-advised-functions)
|
|
(error "ad-read-advised-function: There are no advised functions"))
|
|
(setq default
|
|
(or default
|
|
(ad-do-advised-functions (function)
|
|
(if (or (null predicate)
|
|
(funcall predicate function))
|
|
(ad-do-return function)))
|
|
(error "ad-read-advised-function: %s"
|
|
"There are no qualifying advised functions")))
|
|
(let* ((ad-pReDiCaTe predicate)
|
|
(function
|
|
(completing-read
|
|
(format "%s (default %s): " (or prompt "Function") default)
|
|
ad-advised-functions
|
|
(if predicate
|
|
(function
|
|
(lambda (function)
|
|
;; Oops, no closures - the joys of dynamic scoping:
|
|
;; `predicate' clashed with the `predicate' argument
|
|
;; of Lemacs' `completing-read'.....
|
|
(funcall ad-pReDiCaTe (intern (car function))))))
|
|
t)))
|
|
(if (equal function "")
|
|
(if (ad-is-advised default)
|
|
default
|
|
(error "ad-read-advised-function: `%s' is not advised" default))
|
|
(intern function))))
|
|
|
|
(defvar ad-advice-class-completion-table
|
|
(mapcar (lambda (class) (list (symbol-name class)))
|
|
ad-advice-classes))
|
|
|
|
(defun ad-read-advice-class (function &optional prompt default)
|
|
"Read a valid advice class with completion from the minibuffer.
|
|
An optional PROMPT will be used to prompt for the class. DEFAULT will
|
|
be returned on empty input (defaults to the first non-empty advice
|
|
class of FUNCTION)."
|
|
(setq default
|
|
(or default
|
|
(ad-dolist (class ad-advice-classes)
|
|
(if (ad-get-advice-info-field function class)
|
|
(ad-do-return class)))
|
|
(error "ad-read-advice-class: `%s' has no advices" function)))
|
|
(let ((class (completing-read
|
|
(format "%s (default %s): " (or prompt "Class") default)
|
|
ad-advice-class-completion-table nil t)))
|
|
(if (equal class "")
|
|
default
|
|
(intern class))))
|
|
|
|
(defun ad-read-advice-name (function class &optional prompt)
|
|
"Read name of existing advice of CLASS for FUNCTION with completion.
|
|
An optional PROMPT is used to prompt for the name."
|
|
(let* ((name-completion-table
|
|
(mapcar (function (lambda (advice)
|
|
(list (symbol-name (ad-advice-name advice)))))
|
|
(ad-get-advice-info-field function class)))
|
|
(default
|
|
(if (null name-completion-table)
|
|
(error "ad-read-advice-name: `%s' has no %s advice"
|
|
function class)
|
|
(car (car name-completion-table))))
|
|
(prompt (format "%s (default %s): " (or prompt "Name") default))
|
|
(name (completing-read prompt name-completion-table nil t)))
|
|
(if (equal name "")
|
|
(intern default)
|
|
(intern name))))
|
|
|
|
(defun ad-read-advice-specification (&optional prompt)
|
|
"Read a complete function/class/name specification from minibuffer.
|
|
The list of read symbols will be returned. The optional PROMPT will
|
|
be used to prompt for the function."
|
|
(let* ((function (ad-read-advised-function prompt))
|
|
(class (ad-read-advice-class function))
|
|
(name (ad-read-advice-name function class)))
|
|
(list function class name)))
|
|
|
|
;; Use previous regexp as a default:
|
|
(defvar ad-last-regexp "")
|
|
|
|
(defun ad-read-regexp (&optional prompt)
|
|
"Read a regular expression from the minibuffer."
|
|
(let ((regexp (read-from-minibuffer
|
|
(concat (or prompt "Regular expression")
|
|
(if (equal ad-last-regexp "") ": "
|
|
(format " (default %s): " ad-last-regexp))))))
|
|
(setq ad-last-regexp
|
|
(if (equal regexp "") ad-last-regexp regexp))))
|
|
|
|
|
|
;; @@ Finding, enabling, adding and removing pieces of advice:
|
|
;; ===========================================================
|
|
|
|
(defmacro ad-find-advice (function class name)
|
|
"Find the first advice of FUNCTION in CLASS with NAME."
|
|
`(assq ,name (ad-get-advice-info-field ,function ,class)))
|
|
|
|
(defun ad-advice-position (function class name)
|
|
"Return position of first advice of FUNCTION in CLASS with NAME."
|
|
(let* ((found-advice (ad-find-advice function class name))
|
|
(advices (ad-get-advice-info-field function class)))
|
|
(if found-advice
|
|
(- (length advices) (length (memq found-advice advices))))))
|
|
|
|
(defun ad-find-some-advice (function class name)
|
|
"Find the first of FUNCTION's advices in CLASS matching NAME.
|
|
NAME can be a symbol or a regular expression matching part of an advice name.
|
|
If CLASS is `any' all valid advice classes will be checked."
|
|
(if (ad-is-advised function)
|
|
(let (found-advice)
|
|
(ad-dolist (advice-class ad-advice-classes)
|
|
(if (or (eq class 'any) (eq advice-class class))
|
|
(setq found-advice
|
|
(ad-dolist (advice (ad-get-advice-info-field
|
|
function advice-class))
|
|
(if (or (and (stringp name)
|
|
(string-match
|
|
name (symbol-name
|
|
(ad-advice-name advice))))
|
|
(eq name (ad-advice-name advice)))
|
|
(ad-do-return advice)))))
|
|
(if found-advice (ad-do-return found-advice))))))
|
|
|
|
(defun ad-enable-advice-internal (function class name flag)
|
|
"Set enable FLAG of FUNCTION's advices in CLASS matching NAME.
|
|
If NAME is a string rather than a symbol then it's interpreted as a regular
|
|
expression and all advices whose name contain a match for it will be
|
|
affected. If CLASS is `any' advices in all valid advice classes will be
|
|
considered. The number of changed advices will be returned (or nil if
|
|
FUNCTION was not advised)."
|
|
(if (ad-is-advised function)
|
|
(let ((matched-advices 0))
|
|
(ad-dolist (advice-class ad-advice-classes)
|
|
(if (or (eq class 'any) (eq advice-class class))
|
|
(ad-dolist (advice (ad-get-advice-info-field
|
|
function advice-class))
|
|
(cond ((or (and (stringp name)
|
|
(string-match
|
|
name (symbol-name (ad-advice-name advice))))
|
|
(eq name (ad-advice-name advice)))
|
|
(setq matched-advices (1+ matched-advices))
|
|
(ad-advice-set-enabled advice flag))))))
|
|
matched-advices)))
|
|
|
|
;;;###autoload
|
|
(defun ad-enable-advice (function class name)
|
|
"Enables the advice of FUNCTION with CLASS and NAME."
|
|
(interactive (ad-read-advice-specification "Enable advice of"))
|
|
(if (ad-is-advised function)
|
|
(if (eq (ad-enable-advice-internal function class name t) 0)
|
|
(error "ad-enable-advice: `%s' has no %s advice matching `%s'"
|
|
function class name))
|
|
(error "ad-enable-advice: `%s' is not advised" function)))
|
|
|
|
;;;###autoload
|
|
(defun ad-disable-advice (function class name)
|
|
"Disable the advice of FUNCTION with CLASS and NAME."
|
|
(interactive (ad-read-advice-specification "Disable advice of"))
|
|
(if (ad-is-advised function)
|
|
(if (eq (ad-enable-advice-internal function class name nil) 0)
|
|
(error "ad-disable-advice: `%s' has no %s advice matching `%s'"
|
|
function class name))
|
|
(error "ad-disable-advice: `%s' is not advised" function)))
|
|
|
|
(defun ad-enable-regexp-internal (regexp class flag)
|
|
"Set enable FLAGs of all CLASS advices whose name contains a REGEXP match.
|
|
If CLASS is `any' all valid advice classes are considered. The number of
|
|
affected advices will be returned."
|
|
(let ((matched-advices 0))
|
|
(ad-do-advised-functions (advised-function)
|
|
(setq matched-advices
|
|
(+ matched-advices
|
|
(or (ad-enable-advice-internal
|
|
advised-function class regexp flag)
|
|
0))))
|
|
matched-advices))
|
|
|
|
(defun ad-enable-regexp (regexp)
|
|
"Enables all advices with names that contain a match for REGEXP.
|
|
All currently advised functions will be considered."
|
|
(interactive
|
|
(list (ad-read-regexp "Enable advices via regexp")))
|
|
(let ((matched-advices (ad-enable-regexp-internal regexp 'any t)))
|
|
(if (interactive-p)
|
|
(message "%d matching advices enabled" matched-advices))
|
|
matched-advices))
|
|
|
|
(defun ad-disable-regexp (regexp)
|
|
"Disable all advices with names that contain a match for REGEXP.
|
|
All currently advised functions will be considered."
|
|
(interactive
|
|
(list (ad-read-regexp "Disable advices via regexp")))
|
|
(let ((matched-advices (ad-enable-regexp-internal regexp 'any nil)))
|
|
(if (interactive-p)
|
|
(message "%d matching advices disabled" matched-advices))
|
|
matched-advices))
|
|
|
|
(defun ad-remove-advice (function class name)
|
|
"Remove FUNCTION's advice with NAME from its advices in CLASS.
|
|
If such an advice was found it will be removed from the list of advices
|
|
in that CLASS."
|
|
(interactive (ad-read-advice-specification "Remove advice of"))
|
|
(if (ad-is-advised function)
|
|
(let ((advice-to-remove (ad-find-advice function class name)))
|
|
(if advice-to-remove
|
|
(ad-set-advice-info-field
|
|
function class
|
|
(delq advice-to-remove (ad-get-advice-info-field function class)))
|
|
(error "ad-remove-advice: `%s' has no %s advice `%s'"
|
|
function class name)))
|
|
(error "ad-remove-advice: `%s' is not advised" function)))
|
|
|
|
;;;###autoload
|
|
(defun ad-add-advice (function advice class position)
|
|
"Add a piece of ADVICE to FUNCTION's list of advices in CLASS.
|
|
If FUNCTION already has one or more pieces of advice of the specified
|
|
CLASS then POSITION determines where the new piece will go. The value
|
|
of POSITION can either be `first', `last' or a number where 0 corresponds
|
|
to `first'. Numbers outside the range will be mapped to the closest
|
|
extreme position. If there was already a piece of ADVICE with the same
|
|
name, then the position argument will be ignored and the old advice
|
|
will be overwritten with the new one.
|
|
If the FUNCTION was not advised already, then its advice info will be
|
|
initialized. Redefining a piece of advice whose name is part of the cache-id
|
|
will clear the cache."
|
|
(cond ((not (ad-is-advised function))
|
|
(ad-initialize-advice-info function)
|
|
(ad-set-advice-info-field
|
|
function 'origname (ad-make-origname function))))
|
|
(let* ((previous-position
|
|
(ad-advice-position function class (ad-advice-name advice)))
|
|
(advices (ad-get-advice-info-field function class))
|
|
;; Determine a numerical position for the new advice:
|
|
(position (cond (previous-position)
|
|
((eq position 'first) 0)
|
|
((eq position 'last) (length advices))
|
|
((numberp position)
|
|
(max 0 (min position (length advices))))
|
|
(t 0))))
|
|
;; Check whether we have to clear the cache:
|
|
(if (memq (ad-advice-name advice) (ad-get-cache-class-id function class))
|
|
(ad-clear-cache function))
|
|
(if previous-position
|
|
(setcar (nthcdr position advices) advice)
|
|
(if (= position 0)
|
|
(ad-set-advice-info-field function class (cons advice advices))
|
|
(setcdr (nthcdr (1- position) advices)
|
|
(cons advice (nthcdr position advices)))))))
|
|
|
|
|
|
;; @@ Accessing and manipulating function definitions:
|
|
;; ===================================================
|
|
|
|
(defmacro ad-macrofy (definition)
|
|
"Take a lambda function DEFINITION and make a macro out of it."
|
|
`(cons 'macro ,definition))
|
|
|
|
(defmacro ad-lambdafy (definition)
|
|
"Take a macro function DEFINITION and make a lambda out of it."
|
|
`(cdr ,definition))
|
|
|
|
(defun ad-special-form-p (definition)
|
|
"Non-nil if and only if DEFINITION is a special form."
|
|
(if (and (symbolp definition) (fboundp definition))
|
|
(setq definition (indirect-function definition)))
|
|
(and (subrp definition) (eq (cdr (subr-arity definition)) 'unevalled)))
|
|
|
|
(defmacro ad-subr-p (definition)
|
|
;;"non-nil if DEFINITION is a subr."
|
|
(list 'subrp definition))
|
|
|
|
(defmacro ad-macro-p (definition)
|
|
;;"non-nil if DEFINITION is a macro."
|
|
`(eq (car-safe ,definition) 'macro))
|
|
|
|
(defmacro ad-lambda-p (definition)
|
|
;;"non-nil if DEFINITION is a lambda expression."
|
|
`(eq (car-safe ,definition) 'lambda))
|
|
|
|
;; see ad-make-advice for the format of advice definitions:
|
|
(defmacro ad-advice-p (definition)
|
|
;;"non-nil if DEFINITION is a piece of advice."
|
|
`(eq (car-safe ,definition) 'advice))
|
|
|
|
;; Emacs/Lemacs cross-compatibility
|
|
;; (compiled-function-p is an obsolete function in Emacs):
|
|
(if (and (not (fboundp 'byte-code-function-p))
|
|
(fboundp 'compiled-function-p))
|
|
(ad-safe-fset 'byte-code-function-p 'compiled-function-p))
|
|
|
|
(defmacro ad-compiled-p (definition)
|
|
"Return non-nil if DEFINITION is a compiled byte-code object."
|
|
`(or (byte-code-function-p ,definition)
|
|
(and (ad-macro-p ,definition)
|
|
(byte-code-function-p (ad-lambdafy ,definition)))))
|
|
|
|
(defmacro ad-compiled-code (compiled-definition)
|
|
"Return the byte-code object of a COMPILED-DEFINITION."
|
|
`(if (ad-macro-p ,compiled-definition)
|
|
(ad-lambdafy ,compiled-definition)
|
|
,compiled-definition))
|
|
|
|
(defun ad-lambda-expression (definition)
|
|
"Return the lambda expression of a function/macro/advice DEFINITION."
|
|
(cond ((ad-lambda-p definition)
|
|
definition)
|
|
((ad-macro-p definition)
|
|
(ad-lambdafy definition))
|
|
((ad-advice-p definition)
|
|
(cdr definition))
|
|
(t nil)))
|
|
|
|
(defun ad-arglist (definition &optional name)
|
|
"Return the argument list of DEFINITION.
|
|
If DEFINITION could be from a subr then its NAME should be
|
|
supplied to make subr arglist lookup more efficient."
|
|
(cond ((ad-compiled-p definition)
|
|
(aref (ad-compiled-code definition) 0))
|
|
((consp definition)
|
|
(car (cdr (ad-lambda-expression definition))))
|
|
((ad-subr-p definition)
|
|
(if name
|
|
(ad-subr-arglist name)
|
|
;; otherwise get it from its printed representation:
|
|
(setq name (format "%s" definition))
|
|
(string-match "^#<subr \\([^>]+\\)>$" name)
|
|
(ad-subr-arglist (intern (match-string 1 name)))))))
|
|
|
|
;; Store subr-args as `((arg1 arg2 ...))' so I can distinguish
|
|
;; a defined empty arglist `(nil)' from an undefined arglist:
|
|
(defmacro ad-define-subr-args (subr arglist)
|
|
`(put ,subr 'ad-subr-arglist (list ,arglist)))
|
|
(defmacro ad-undefine-subr-args (subr)
|
|
`(put ,subr 'ad-subr-arglist nil))
|
|
(defmacro ad-subr-args-defined-p (subr)
|
|
`(get ,subr 'ad-subr-arglist))
|
|
(defmacro ad-get-subr-args (subr)
|
|
`(car (get ,subr 'ad-subr-arglist)))
|
|
|
|
(defun ad-subr-arglist (subr-name)
|
|
"Retrieve arglist of the subr with SUBR-NAME.
|
|
Either use the one stored under the `ad-subr-arglist' property,
|
|
or try to retrieve it from the docstring and cache it under
|
|
that property, or otherwise use `(&rest ad-subr-args)'."
|
|
(if (ad-subr-args-defined-p subr-name)
|
|
(ad-get-subr-args subr-name)
|
|
;; says jwz: Should use this for Lemacs 19.8 and above:
|
|
;;((fboundp 'subr-min-args)
|
|
;; ...)
|
|
;; says hans: I guess what Jamie means is that I should use the values
|
|
;; of `subr-min-args' and `subr-max-args' to construct the subr arglist
|
|
;; without having to look it up via parsing the docstring, e.g.,
|
|
;; values 1 and 2 would suggest `(arg1 &optional arg2)' as an
|
|
;; argument list. However, that won't work because there is no
|
|
;; way to distinguish a subr with args `(a &optional b &rest c)' from
|
|
;; one with args `(a &rest c)' using that mechanism. Also, the argument
|
|
;; names from the docstring are more meaningful. Hence, I'll stick with
|
|
;; the old way of doing things.
|
|
(let ((doc (or (ad-real-documentation subr-name t) "")))
|
|
(if (not (string-match "\n\n\\((.+)\\)\\'" doc))
|
|
;; Signalling an error leads to bugs during bootstrapping because
|
|
;; the DOC file is not yet built (which is an error, BTW).
|
|
;; (error "The usage info is missing from the subr %s" subr-name)
|
|
'(&rest ad-subr-args)
|
|
(ad-define-subr-args
|
|
subr-name
|
|
(cdr (car (read-from-string
|
|
(downcase (match-string 1 doc))))))
|
|
(ad-get-subr-args subr-name)))))
|
|
|
|
(defun ad-docstring (definition)
|
|
"Return the unexpanded docstring of DEFINITION."
|
|
(let ((docstring
|
|
(if (ad-compiled-p definition)
|
|
(ad-real-documentation definition t)
|
|
(car (cdr (cdr (ad-lambda-expression definition)))))))
|
|
(if (or (stringp docstring)
|
|
(natnump docstring))
|
|
docstring)))
|
|
|
|
(defun ad-interactive-form (definition)
|
|
"Return the interactive form of DEFINITION.
|
|
Like `interactive-form', but also works on pieces of advice."
|
|
(interactive-form
|
|
(if (ad-advice-p definition)
|
|
(ad-lambda-expression definition)
|
|
definition)))
|
|
|
|
(defun ad-body-forms (definition)
|
|
"Return the list of body forms of DEFINITION."
|
|
(cond ((ad-compiled-p definition)
|
|
nil)
|
|
((consp definition)
|
|
(nthcdr (+ (if (ad-docstring definition) 1 0)
|
|
(if (ad-interactive-form definition) 1 0))
|
|
(cdr (cdr (ad-lambda-expression definition)))))))
|
|
|
|
(defun ad-make-advised-definition-docstring (function)
|
|
"Make an identifying docstring for the advised definition of FUNCTION.
|
|
Put function name into the documentation string so we can infer
|
|
the name of the advised function from the docstring. This is needed
|
|
to generate a proper advised docstring even if we are just given a
|
|
definition (see the code for `documentation')."
|
|
(propertize "Advice doc string" 'ad-advice-info function))
|
|
|
|
(defun ad-advised-definition-p (definition)
|
|
"Return non-nil if DEFINITION was generated from advice information."
|
|
(if (or (ad-lambda-p definition)
|
|
(ad-macro-p definition)
|
|
(ad-compiled-p definition))
|
|
(let ((docstring (ad-docstring definition)))
|
|
(and (stringp docstring)
|
|
(get-text-property 0 'ad-advice-info docstring)))))
|
|
|
|
(defun ad-definition-type (definition)
|
|
"Return symbol that describes the type of DEFINITION."
|
|
(if (ad-macro-p definition)
|
|
'macro
|
|
(if (ad-subr-p definition)
|
|
(if (ad-special-form-p definition)
|
|
'special-form
|
|
'subr)
|
|
(if (or (ad-lambda-p definition)
|
|
(ad-compiled-p definition))
|
|
'function
|
|
(if (ad-advice-p definition)
|
|
'advice)))))
|
|
|
|
(defun ad-has-proper-definition (function)
|
|
"True if FUNCTION is a symbol with a proper definition.
|
|
For that it has to be fbound with a non-autoload definition."
|
|
(and (symbolp function)
|
|
(fboundp function)
|
|
(not (eq (car-safe (symbol-function function)) 'autoload))))
|
|
|
|
;; The following two are necessary for the sake of packages such as
|
|
;; ange-ftp which redefine functions via fcell indirection:
|
|
(defun ad-real-definition (function)
|
|
"Find FUNCTION's definition at the end of function cell indirection."
|
|
(if (ad-has-proper-definition function)
|
|
(let ((definition (symbol-function function)))
|
|
(if (symbolp definition)
|
|
(ad-real-definition definition)
|
|
definition))))
|
|
|
|
(defun ad-real-orig-definition (function)
|
|
"Find FUNCTION's real original definition starting from its `origname'."
|
|
(if (ad-is-advised function)
|
|
(ad-real-definition (ad-get-advice-info-field function 'origname))))
|
|
|
|
(defun ad-is-compilable (function)
|
|
"True if FUNCTION has an interpreted definition that can be compiled."
|
|
(and (ad-has-proper-definition function)
|
|
(or (ad-lambda-p (symbol-function function))
|
|
(ad-macro-p (symbol-function function)))
|
|
(not (ad-compiled-p (symbol-function function)))))
|
|
|
|
(defun ad-compile-function (function)
|
|
"Byte-compiles FUNCTION (or macro) if it is not yet compiled."
|
|
(interactive "aByte-compile function: ")
|
|
(if (ad-is-compilable function)
|
|
;; Need to turn off auto-activation
|
|
;; because `byte-compile' uses `fset':
|
|
(ad-with-auto-activation-disabled
|
|
(require 'bytecomp)
|
|
(let ((symbol (make-symbol "advice-compilation"))
|
|
(byte-compile-warnings byte-compile-warnings))
|
|
(if (featurep 'cl)
|
|
(byte-compile-disable-warning 'cl-functions))
|
|
(fset symbol (symbol-function function))
|
|
(byte-compile symbol)
|
|
(fset function (symbol-function symbol))))))
|
|
|
|
|
|
;; @@ Constructing advised definitions:
|
|
;; ====================================
|
|
;;
|
|
;; Main design decisions about the form of advised definitions:
|
|
;;
|
|
;; A) How will original definitions be called?
|
|
;; B) What will argument lists of advised functions look like?
|
|
;;
|
|
;; Ad A)
|
|
;; I chose to use function indirection for all four types of original
|
|
;; definitions (functions, macros, subrs and special forms), i.e., create
|
|
;; a unique symbol `ad-Orig-<name>' which is fbound to the original
|
|
;; definition and call it according to type and arguments. Functions and
|
|
;; subrs that don't have any &rest arguments can be called directly in a
|
|
;; `(ad-Orig-<name> ....)' form. If they have a &rest argument we have to
|
|
;; use `apply'. Macros will be called with
|
|
;; `(macroexpand '(ad-Orig-<name> ....))', and special forms also need a
|
|
;; form like that with `eval' instead of `macroexpand'.
|
|
;;
|
|
;; Ad B)
|
|
;; Use original arguments where possible and `(&rest ad-subr-args)'
|
|
;; otherwise, even though this seems to be more complicated and less
|
|
;; uniform than a general `(&rest args)' approach. My reason to still
|
|
;; do it that way is that in most cases my approach leads to the more
|
|
;; efficient form for the advised function, and portability (e.g., to
|
|
;; make the same advice work regardless of whether something is a
|
|
;; function or a subr) can still be achieved with argument access macros.
|
|
|
|
|
|
(defun ad-prognify (forms)
|
|
(cond ((<= (length forms) 1)
|
|
(car forms))
|
|
(t (cons 'progn forms))))
|
|
|
|
;; @@@ Accessing argument lists:
|
|
;; =============================
|
|
|
|
(defun ad-parse-arglist (arglist)
|
|
"Parse ARGLIST into its required, optional and rest parameters.
|
|
A three-element list is returned, where the 1st element is the list of
|
|
required arguments, the 2nd is the list of optional arguments, and the 3rd
|
|
is the name of an optional rest parameter (or nil)."
|
|
(let (required optional rest)
|
|
(setq rest (car (cdr (memq '&rest arglist))))
|
|
(if rest (setq arglist (reverse (cdr (memq '&rest (reverse arglist))))))
|
|
(setq optional (cdr (memq '&optional arglist)))
|
|
(if optional
|
|
(setq required (reverse (cdr (memq '&optional (reverse arglist)))))
|
|
(setq required arglist))
|
|
(list required optional rest)))
|
|
|
|
(defun ad-retrieve-args-form (arglist)
|
|
"Generate a form which evaluates into names/values/types of ARGLIST.
|
|
When the form gets evaluated within a function with that argument list
|
|
it will result in a list with one entry for each argument, where the
|
|
first element of each entry is the name of the argument, the second
|
|
element is its actual current value, and the third element is either
|
|
`required', `optional' or `rest' depending on the type of the argument."
|
|
(let* ((parsed-arglist (ad-parse-arglist arglist))
|
|
(rest (nth 2 parsed-arglist)))
|
|
`(list
|
|
,@(mapcar (function
|
|
(lambda (req)
|
|
`(list ',req ,req 'required)))
|
|
(nth 0 parsed-arglist))
|
|
,@(mapcar (function
|
|
(lambda (opt)
|
|
`(list ',opt ,opt 'optional)))
|
|
(nth 1 parsed-arglist))
|
|
,@(if rest (list `(list ',rest ,rest 'rest))))))
|
|
|
|
(defun ad-arg-binding-field (binding field)
|
|
(cond ((eq field 'name) (car binding))
|
|
((eq field 'value) (car (cdr binding)))
|
|
((eq field 'type) (car (cdr (cdr binding))))))
|
|
|
|
(defun ad-list-access (position list)
|
|
(cond ((= position 0) list)
|
|
((= position 1) (list 'cdr list))
|
|
(t (list 'nthcdr position list))))
|
|
|
|
(defun ad-element-access (position list)
|
|
(cond ((= position 0) (list 'car list))
|
|
((= position 1) `(car (cdr ,list)))
|
|
(t (list 'nth position list))))
|
|
|
|
(defun ad-access-argument (arglist index)
|
|
"Tell how to access ARGLIST's actual argument at position INDEX.
|
|
For a required/optional arg it simply returns it, if a rest argument has
|
|
to be accessed, it returns a list with the index and name."
|
|
(let* ((parsed-arglist (ad-parse-arglist arglist))
|
|
(reqopt-args (append (nth 0 parsed-arglist)
|
|
(nth 1 parsed-arglist)))
|
|
(rest-arg (nth 2 parsed-arglist)))
|
|
(cond ((< index (length reqopt-args))
|
|
(nth index reqopt-args))
|
|
(rest-arg
|
|
(list (- index (length reqopt-args)) rest-arg)))))
|
|
|
|
(defun ad-get-argument (arglist index)
|
|
"Return form to access ARGLIST's actual argument at position INDEX."
|
|
(let ((argument-access (ad-access-argument arglist index)))
|
|
(cond ((consp argument-access)
|
|
(ad-element-access
|
|
(car argument-access) (car (cdr argument-access))))
|
|
(argument-access))))
|
|
|
|
(defun ad-set-argument (arglist index value-form)
|
|
"Return form to set ARGLIST's actual arg at INDEX to VALUE-FORM."
|
|
(let ((argument-access (ad-access-argument arglist index)))
|
|
(cond ((consp argument-access)
|
|
;; should this check whether there actually is something to set?
|
|
`(setcar ,(ad-list-access
|
|
(car argument-access) (car (cdr argument-access)))
|
|
,value-form))
|
|
(argument-access
|
|
`(setq ,argument-access ,value-form))
|
|
(t (error "ad-set-argument: No argument at position %d of `%s'"
|
|
index arglist)))))
|
|
|
|
(defun ad-get-arguments (arglist index)
|
|
"Return form to access all actual arguments starting at position INDEX."
|
|
(let* ((parsed-arglist (ad-parse-arglist arglist))
|
|
(reqopt-args (append (nth 0 parsed-arglist)
|
|
(nth 1 parsed-arglist)))
|
|
(rest-arg (nth 2 parsed-arglist))
|
|
args-form)
|
|
(if (< index (length reqopt-args))
|
|
(setq args-form `(list ,@(nthcdr index reqopt-args))))
|
|
(if rest-arg
|
|
(if args-form
|
|
(setq args-form `(nconc ,args-form ,rest-arg))
|
|
(setq args-form (ad-list-access (- index (length reqopt-args))
|
|
rest-arg))))
|
|
args-form))
|
|
|
|
(defun ad-set-arguments (arglist index values-form)
|
|
"Make form to assign elements of VALUES-FORM as actual ARGLIST args.
|
|
The assignment starts at position INDEX."
|
|
(let ((values-index 0)
|
|
argument-access set-forms)
|
|
(while (setq argument-access (ad-access-argument arglist index))
|
|
(if (symbolp argument-access)
|
|
(setq set-forms
|
|
(cons (ad-set-argument
|
|
arglist index
|
|
(ad-element-access values-index 'ad-vAlUeS))
|
|
set-forms))
|
|
(setq set-forms
|
|
(cons (if (= (car argument-access) 0)
|
|
(list 'setq
|
|
(car (cdr argument-access))
|
|
(ad-list-access values-index 'ad-vAlUeS))
|
|
(list 'setcdr
|
|
(ad-list-access (1- (car argument-access))
|
|
(car (cdr argument-access)))
|
|
(ad-list-access values-index 'ad-vAlUeS)))
|
|
set-forms))
|
|
;; terminate loop
|
|
(setq arglist nil))
|
|
(setq index (1+ index))
|
|
(setq values-index (1+ values-index)))
|
|
(if (null set-forms)
|
|
(error "ad-set-arguments: No argument at position %d of `%s'"
|
|
index arglist)
|
|
(if (= (length set-forms) 1)
|
|
;; For exactly one set-form we can use values-form directly,...
|
|
(ad-substitute-tree
|
|
(function (lambda (form) (eq form 'ad-vAlUeS)))
|
|
(function (lambda (form) values-form))
|
|
(car set-forms))
|
|
;; ...if we have more we have to bind it to a variable:
|
|
`(let ((ad-vAlUeS ,values-form))
|
|
,@(reverse set-forms)
|
|
;; work around the old backquote bug:
|
|
,'ad-vAlUeS)))))
|
|
|
|
(defun ad-insert-argument-access-forms (definition arglist)
|
|
"Expands arg-access text macros in DEFINITION according to ARGLIST."
|
|
(ad-substitute-tree
|
|
(function
|
|
(lambda (form)
|
|
(or (eq form 'ad-arg-bindings)
|
|
(and (memq (car-safe form)
|
|
'(ad-get-arg ad-get-args ad-set-arg ad-set-args))
|
|
(integerp (car-safe (cdr form)))))))
|
|
(function
|
|
(lambda (form)
|
|
(if (eq form 'ad-arg-bindings)
|
|
(ad-retrieve-args-form arglist)
|
|
(let ((accessor (car form))
|
|
(index (car (cdr form)))
|
|
(val (car (cdr (ad-insert-argument-access-forms
|
|
(cdr form) arglist)))))
|
|
(cond ((eq accessor 'ad-get-arg)
|
|
(ad-get-argument arglist index))
|
|
((eq accessor 'ad-set-arg)
|
|
(ad-set-argument arglist index val))
|
|
((eq accessor 'ad-get-args)
|
|
(ad-get-arguments arglist index))
|
|
((eq accessor 'ad-set-args)
|
|
(ad-set-arguments arglist index val)))))))
|
|
definition))
|
|
|
|
;; @@@ Mapping argument lists:
|
|
;; ===========================
|
|
;; Here is the problem:
|
|
;; Suppose function foo was called with (foo 1 2 3 4 5), and foo has the
|
|
;; argument list (x y &rest z), and we want to call the function bar which
|
|
;; has argument list (a &rest b) with a combination of x, y and z so that
|
|
;; the effect is just as if we had called (bar 1 2 3 4 5) directly.
|
|
;; The mapping should work for any two argument lists.
|
|
|
|
(defun ad-map-arglists (source-arglist target-arglist)
|
|
"Make `funcall/apply' form to map SOURCE-ARGLIST to TARGET-ARGLIST.
|
|
The arguments supplied to TARGET-ARGLIST will be taken from SOURCE-ARGLIST just
|
|
as if they had been supplied to a function with TARGET-ARGLIST directly.
|
|
Excess source arguments will be neglected, missing source arguments will be
|
|
supplied as nil. Returns a `funcall' or `apply' form with the second element
|
|
being `function' which has to be replaced by an actual function argument.
|
|
Example: `(ad-map-arglists '(a &rest args) '(w x y z))' will return
|
|
`(funcall function a (car args) (car (cdr args)) (nth 2 args))'."
|
|
(let* ((parsed-source-arglist (ad-parse-arglist source-arglist))
|
|
(source-reqopt-args (append (nth 0 parsed-source-arglist)
|
|
(nth 1 parsed-source-arglist)))
|
|
(source-rest-arg (nth 2 parsed-source-arglist))
|
|
(parsed-target-arglist (ad-parse-arglist target-arglist))
|
|
(target-reqopt-args (append (nth 0 parsed-target-arglist)
|
|
(nth 1 parsed-target-arglist)))
|
|
(target-rest-arg (nth 2 parsed-target-arglist))
|
|
(need-apply (and source-rest-arg target-rest-arg))
|
|
(target-arg-index -1))
|
|
;; This produces ``error-proof'' target function calls with the exception
|
|
;; of a case like (&rest a) mapped onto (x &rest y) where the actual args
|
|
;; supplied to A might not be enough to supply the required target arg X
|
|
(append (list (if need-apply 'apply 'funcall) 'function)
|
|
(cond (need-apply
|
|
;; `apply' can take care of that directly:
|
|
(append source-reqopt-args (list source-rest-arg)))
|
|
(t (mapcar (function
|
|
(lambda (arg)
|
|
(setq target-arg-index (1+ target-arg-index))
|
|
(ad-get-argument
|
|
source-arglist target-arg-index)))
|
|
(append target-reqopt-args
|
|
(and target-rest-arg
|
|
;; If we have a rest arg gobble up
|
|
;; remaining source args:
|
|
(nthcdr (length target-reqopt-args)
|
|
source-reqopt-args)))))))))
|
|
|
|
(defun ad-make-mapped-call (source-arglist target-arglist target-function)
|
|
"Make form to call TARGET-FUNCTION with args from SOURCE-ARGLIST."
|
|
(let ((mapped-form (ad-map-arglists source-arglist target-arglist)))
|
|
(if (eq (car mapped-form) 'funcall)
|
|
(cons target-function (cdr (cdr mapped-form)))
|
|
(prog1 mapped-form
|
|
(setcar (cdr mapped-form) (list 'quote target-function))))))
|
|
|
|
;; @@@ Making an advised documentation string:
|
|
;; ===========================================
|
|
;; New policy: The documentation string for an advised function will be built
|
|
;; at the time the advised `documentation' function is called. This has the
|
|
;; following advantages:
|
|
;; 1) command-key substitutions will automatically be correct
|
|
;; 2) No wasted string space due to big advised docstrings in caches or
|
|
;; compiled files that contain preactivations
|
|
;; The overall overhead for this should be negligible because people normally
|
|
;; don't lookup documentation for the same function over and over again.
|
|
|
|
(defun ad-make-single-advice-docstring (advice class &optional style)
|
|
(let ((advice-docstring (ad-docstring (ad-advice-definition advice))))
|
|
(cond ((eq style 'plain)
|
|
advice-docstring)
|
|
((eq style 'freeze)
|
|
(format "Permanent %s-advice `%s':%s%s"
|
|
class (ad-advice-name advice)
|
|
(if advice-docstring "\n" "")
|
|
(or advice-docstring "")))
|
|
(t (if advice-docstring
|
|
(format "%s-advice `%s':\n%s"
|
|
(capitalize (symbol-name class))
|
|
(ad-advice-name advice)
|
|
advice-docstring)
|
|
(format "%s-advice `%s'."
|
|
(capitalize (symbol-name class))
|
|
(ad-advice-name advice)))))))
|
|
|
|
(require 'help-fns) ;For help-split-fundoc and help-add-fundoc-usage.
|
|
|
|
(defun ad-make-advised-docstring (function &optional style)
|
|
"Construct a documentation string for the advised FUNCTION.
|
|
It concatenates the original documentation with the documentation
|
|
strings of the individual pieces of advice which will be formatted
|
|
according to STYLE. STYLE can be `plain' or `freeze', everything else
|
|
will be interpreted as `default'. The order of the advice documentation
|
|
strings corresponds to before/around/after and the individual ordering
|
|
in any of these classes."
|
|
(let* ((origdef (ad-real-orig-definition function))
|
|
(origtype (symbol-name (ad-definition-type origdef)))
|
|
(origdoc
|
|
;; Retrieve raw doc, key substitution will be taken care of later:
|
|
(ad-real-documentation origdef t))
|
|
(usage (help-split-fundoc origdoc function))
|
|
paragraphs advice-docstring ad-usage)
|
|
(setq usage (if (null usage) t (setq origdoc (cdr usage)) (car usage)))
|
|
(if origdoc (setq paragraphs (list origdoc)))
|
|
(unless (eq style 'plain)
|
|
(push (propertize (concat "This " origtype " is advised.")
|
|
'face 'font-lock-warning-face)
|
|
paragraphs))
|
|
(ad-dolist (class ad-advice-classes)
|
|
(ad-dolist (advice (ad-get-enabled-advices function class))
|
|
(setq advice-docstring
|
|
(ad-make-single-advice-docstring advice class style))
|
|
(if advice-docstring
|
|
(push advice-docstring paragraphs))))
|
|
(setq origdoc (if paragraphs
|
|
(propertize
|
|
;; separate paragraphs with blank lines:
|
|
(mapconcat 'identity (nreverse paragraphs) "\n\n")
|
|
'ad-advice-info function)))
|
|
(help-add-fundoc-usage origdoc usage)))
|
|
|
|
(defun ad-make-plain-docstring (function)
|
|
(ad-make-advised-docstring function 'plain))
|
|
(defun ad-make-freeze-docstring (function)
|
|
(ad-make-advised-docstring function 'freeze))
|
|
|
|
;; @@@ Accessing overriding arglists and interactive forms:
|
|
;; ========================================================
|
|
|
|
(defun ad-advised-arglist (function)
|
|
"Find first defined arglist in FUNCTION's redefining advices."
|
|
(ad-dolist (advice (append (ad-get-enabled-advices function 'before)
|
|
(ad-get-enabled-advices function 'around)
|
|
(ad-get-enabled-advices function 'after)))
|
|
(let ((arglist (ad-arglist (ad-advice-definition advice))))
|
|
(if arglist
|
|
;; We found the first one, use it:
|
|
(ad-do-return arglist)))))
|
|
|
|
(defun ad-advised-interactive-form (function)
|
|
"Find first interactive form in FUNCTION's redefining advices."
|
|
(ad-dolist (advice (append (ad-get-enabled-advices function 'before)
|
|
(ad-get-enabled-advices function 'around)
|
|
(ad-get-enabled-advices function 'after)))
|
|
(let ((interactive-form
|
|
(ad-interactive-form (ad-advice-definition advice))))
|
|
(if interactive-form
|
|
;; We found the first one, use it:
|
|
(ad-do-return interactive-form)))))
|
|
|
|
;; @@@ Putting it all together:
|
|
;; ============================
|
|
|
|
(defun ad-make-advised-definition (function)
|
|
"Generate an advised definition of FUNCTION from its advice info."
|
|
(if (and (ad-is-advised function)
|
|
(ad-has-redefining-advice function))
|
|
(let* ((origdef (ad-real-orig-definition function))
|
|
(origname (ad-get-advice-info-field function 'origname))
|
|
(orig-interactive-p (commandp origdef))
|
|
(orig-subr-p (ad-subr-p origdef))
|
|
(orig-special-form-p (ad-special-form-p origdef))
|
|
(orig-macro-p (ad-macro-p origdef))
|
|
;; Construct the individual pieces that we need for assembly:
|
|
(orig-arglist (ad-arglist origdef function))
|
|
(advised-arglist (or (ad-advised-arglist function)
|
|
orig-arglist))
|
|
(advised-interactive-form (ad-advised-interactive-form function))
|
|
(interactive-form
|
|
(cond (orig-macro-p nil)
|
|
(advised-interactive-form)
|
|
((interactive-form origdef)
|
|
(interactive-form
|
|
(if (and (symbolp function) (get function 'elp-info))
|
|
(aref (get function 'elp-info) 2)
|
|
origdef)))))
|
|
(orig-form
|
|
(cond ((or orig-special-form-p orig-macro-p)
|
|
;; Special forms and macros will be advised into macros.
|
|
;; The trick is to construct an expansion for the advised
|
|
;; macro that does the correct thing when it gets eval'ed.
|
|
;; For macros we'll just use the expansion of the original
|
|
;; macro and return that. This way compiled advised macros
|
|
;; will be expanded into something useful. Note that after
|
|
;; advices have full control over whether they want to
|
|
;; evaluate the expansion (the value of `ad-return-value')
|
|
;; at macro expansion time or not. For special forms there
|
|
;; is no solution that interacts reasonably with the
|
|
;; compiler, hence we just evaluate the original at macro
|
|
;; expansion time and return the result. The moral of that
|
|
;; is that one should always deactivate advised special
|
|
;; forms before one byte-compiles a file.
|
|
`(,(if orig-macro-p 'macroexpand 'eval)
|
|
(cons ',origname
|
|
,(ad-get-arguments advised-arglist 0))))
|
|
((and orig-subr-p
|
|
orig-interactive-p
|
|
(not interactive-form)
|
|
(not advised-interactive-form))
|
|
;; Check whether we were called interactively
|
|
;; in order to do proper prompting:
|
|
`(if (called-interactively-p)
|
|
(call-interactively ',origname)
|
|
,(ad-make-mapped-call advised-arglist
|
|
orig-arglist
|
|
origname)))
|
|
;; And now for normal functions and non-interactive subrs
|
|
;; (or subrs whose interactive behavior was advised):
|
|
(t (ad-make-mapped-call
|
|
advised-arglist orig-arglist origname)))))
|
|
|
|
;; Finally, build the sucker:
|
|
(ad-assemble-advised-definition
|
|
(cond (orig-macro-p 'macro)
|
|
(orig-special-form-p 'special-form)
|
|
(t 'function))
|
|
advised-arglist
|
|
(ad-make-advised-definition-docstring function)
|
|
interactive-form
|
|
orig-form
|
|
(ad-get-enabled-advices function 'before)
|
|
(ad-get-enabled-advices function 'around)
|
|
(ad-get-enabled-advices function 'after)))))
|
|
|
|
(defun ad-assemble-advised-definition
|
|
(type args docstring interactive orig &optional befores arounds afters)
|
|
|
|
"Assembles an original and its advices into an advised function.
|
|
It constructs a function or macro definition according to TYPE which has to
|
|
be either `macro', `function' or `special-form'. ARGS is the argument list
|
|
that has to be used, DOCSTRING if non-nil defines the documentation of the
|
|
definition, INTERACTIVE if non-nil is the interactive form to be used,
|
|
ORIG is a form that calls the body of the original unadvised function,
|
|
and BEFORES, AROUNDS and AFTERS are the lists of advices with which ORIG
|
|
should be modified. The assembled function will be returned."
|
|
|
|
(let (before-forms around-form around-form-protected after-forms definition)
|
|
(ad-dolist (advice befores)
|
|
(cond ((and (ad-advice-protected advice)
|
|
before-forms)
|
|
(setq before-forms
|
|
`((unwind-protect
|
|
,(ad-prognify before-forms)
|
|
,@(ad-body-forms
|
|
(ad-advice-definition advice))))))
|
|
(t (setq before-forms
|
|
(append before-forms
|
|
(ad-body-forms (ad-advice-definition advice)))))))
|
|
|
|
(setq around-form `(setq ad-return-value ,orig))
|
|
(ad-dolist (advice (reverse arounds))
|
|
;; If any of the around advices is protected then we
|
|
;; protect the complete around advice onion:
|
|
(if (ad-advice-protected advice)
|
|
(setq around-form-protected t))
|
|
(setq around-form
|
|
(ad-substitute-tree
|
|
(function (lambda (form) (eq form 'ad-do-it)))
|
|
(function (lambda (form) around-form))
|
|
(ad-prognify (ad-body-forms (ad-advice-definition advice))))))
|
|
|
|
(setq after-forms
|
|
(if (and around-form-protected before-forms)
|
|
`((unwind-protect
|
|
,(ad-prognify before-forms)
|
|
,around-form))
|
|
(append before-forms (list around-form))))
|
|
(ad-dolist (advice afters)
|
|
(cond ((and (ad-advice-protected advice)
|
|
after-forms)
|
|
(setq after-forms
|
|
`((unwind-protect
|
|
,(ad-prognify after-forms)
|
|
,@(ad-body-forms
|
|
(ad-advice-definition advice))))))
|
|
(t (setq after-forms
|
|
(append after-forms
|
|
(ad-body-forms (ad-advice-definition advice)))))))
|
|
|
|
(setq definition
|
|
`(,@(if (memq type '(macro special-form)) '(macro))
|
|
lambda
|
|
,args
|
|
,@(if docstring (list docstring))
|
|
,@(if interactive (list interactive))
|
|
(let (ad-return-value)
|
|
,@after-forms
|
|
,(if (eq type 'special-form)
|
|
'(list 'quote ad-return-value)
|
|
'ad-return-value))))
|
|
|
|
(ad-insert-argument-access-forms definition args)))
|
|
|
|
;; This is needed for activation/deactivation hooks:
|
|
(defun ad-make-hook-form (function hook-name)
|
|
"Make hook-form from FUNCTION's advice bodies in class HOOK-NAME."
|
|
(let ((hook-forms
|
|
(mapcar (function (lambda (advice)
|
|
(ad-body-forms (ad-advice-definition advice))))
|
|
(ad-get-enabled-advices function hook-name))))
|
|
(if hook-forms
|
|
(ad-prognify (apply 'append hook-forms)))))
|
|
|
|
|
|
;; @@ Caching:
|
|
;; ===========
|
|
;; Generating an advised definition of a function is moderately expensive,
|
|
;; hence, it makes sense to cache it so we can reuse it in appropriate
|
|
;; circumstances. Of course, it only makes sense to reuse a cached
|
|
;; definition if the current advice and function definition state is the
|
|
;; same as it was at the time when the cached definition was generated.
|
|
;; For that purpose we associate every cache with an id so we can verify
|
|
;; if it is still valid at a certain point in time. This id mechanism
|
|
;; makes it possible to preactivate advised functions, write the compiled
|
|
;; advised definitions to a file and reuse them during the actual
|
|
;; activation without having to risk that the resulting definition will be
|
|
;; incorrect, well, almost.
|
|
;;
|
|
;; A cache id is a list with six elements:
|
|
;; 1) the list of names of enabled before advices
|
|
;; 2) the list of names of enabled around advices
|
|
;; 3) the list of names of enabled after advices
|
|
;; 4) the type of the original function (macro, subr, etc.)
|
|
;; 5) the arglist of the original definition (or t if it was equal to the
|
|
;; arglist of the cached definition)
|
|
;; 6) t if the interactive form of the original definition was equal to the
|
|
;; interactive form of the cached definition
|
|
;;
|
|
;; Here's how a cache can get invalidated or be incorrect:
|
|
;; A) a piece of advice used in the cache gets redefined
|
|
;; B) the current list of enabled advices is different from the ones used
|
|
;; for the cache
|
|
;; C) the type of the original function changed, e.g., a function became a
|
|
;; macro, or a subr became a function
|
|
;; D) the arglist of the original function changed
|
|
;; E) the interactive form of the original function changed
|
|
;; F) a piece of advice used in the cache got redefined before the
|
|
;; defadvice with the cached definition got loaded: This is a PROBLEM!
|
|
;;
|
|
;; Cases A and B are the normal ones. A is taken care of by `ad-add-advice'
|
|
;; which clears the cache in such a case, B is easily checked during
|
|
;; verification at activation time.
|
|
;;
|
|
;; Cases C, D and E have to be considered if one is slightly paranoid, i.e.,
|
|
;; if one considers the case that the original function could be different
|
|
;; from the one available at caching time (e.g., for forward advice of
|
|
;; functions that get redefined by some packages - such as `eval-region' gets
|
|
;; redefined by edebug). All these cases can be easily checked during
|
|
;; verification. Element 4 of the id lets one check case C, element 5 takes
|
|
;; care of case D (using t in the equality case saves some space, because the
|
|
;; arglist can be recovered at validation time from the cached definition),
|
|
;; and element 6 takes care of case E which is only a problem if the original
|
|
;; was actually a function whose interactive form was not overridden by a
|
|
;; piece of advice.
|
|
;;
|
|
;; Case F is the only one which will lead to an incorrect advised function.
|
|
;; There is no way to avoid this without storing the complete advice definition
|
|
;; in the cache-id which is not feasible.
|
|
;;
|
|
;; The cache-id of a typical advised function with one piece of advice and
|
|
;; no arglist redefinition takes 7 conses which is a small price to pay for
|
|
;; the added efficiency. The validation itself is also pretty cheap, certainly
|
|
;; a lot cheaper than reconstructing an advised definition.
|
|
|
|
(defmacro ad-get-cache-definition (function)
|
|
`(car (ad-get-advice-info-field ,function 'cache)))
|
|
|
|
(defmacro ad-get-cache-id (function)
|
|
`(cdr (ad-get-advice-info-field ,function 'cache)))
|
|
|
|
(defmacro ad-set-cache (function definition id)
|
|
`(ad-set-advice-info-field
|
|
,function 'cache (cons ,definition ,id)))
|
|
|
|
(defun ad-clear-cache (function)
|
|
"Clears a previously cached advised definition of FUNCTION.
|
|
Clear the cache if you want to force `ad-activate' to construct a new
|
|
advised definition from scratch."
|
|
(interactive
|
|
(list (ad-read-advised-function "Clear cached definition of")))
|
|
(ad-set-advice-info-field function 'cache nil))
|
|
|
|
(defun ad-make-cache-id (function)
|
|
"Generate an identifying image of the current advices of FUNCTION."
|
|
(let ((original-definition (ad-real-orig-definition function))
|
|
(cached-definition (ad-get-cache-definition function)))
|
|
(list (mapcar (function (lambda (advice) (ad-advice-name advice)))
|
|
(ad-get-enabled-advices function 'before))
|
|
(mapcar (function (lambda (advice) (ad-advice-name advice)))
|
|
(ad-get-enabled-advices function 'around))
|
|
(mapcar (function (lambda (advice) (ad-advice-name advice)))
|
|
(ad-get-enabled-advices function 'after))
|
|
(ad-definition-type original-definition)
|
|
(if (equal (ad-arglist original-definition function)
|
|
(ad-arglist cached-definition))
|
|
t
|
|
(ad-arglist original-definition function))
|
|
(if (eq (ad-definition-type original-definition) 'function)
|
|
(equal (interactive-form original-definition)
|
|
(interactive-form cached-definition))))))
|
|
|
|
(defun ad-get-cache-class-id (function class)
|
|
"Return the part of FUNCTION's cache id that identifies CLASS."
|
|
(let ((cache-id (ad-get-cache-id function)))
|
|
(if (eq class 'before)
|
|
(car cache-id)
|
|
(if (eq class 'around)
|
|
(nth 1 cache-id)
|
|
(nth 2 cache-id)))))
|
|
|
|
(defun ad-verify-cache-class-id (cache-class-id advices)
|
|
(ad-dolist (advice advices (null cache-class-id))
|
|
(if (ad-advice-enabled advice)
|
|
(if (eq (car cache-class-id) (ad-advice-name advice))
|
|
(setq cache-class-id (cdr cache-class-id))
|
|
(ad-do-return nil)))))
|
|
|
|
;; There should be a way to monitor if and why a cache verification failed
|
|
;; in order to determine whether a certain preactivation could be used or
|
|
;; not. Right now the only way to find out is to trace
|
|
;; `ad-cache-id-verification-code'. The code it returns indicates where the
|
|
;; verification failed. Tracing `ad-verify-cache-class-id' might provide
|
|
;; some additional useful information.
|
|
|
|
(defun ad-cache-id-verification-code (function)
|
|
(let ((cache-id (ad-get-cache-id function))
|
|
(code 'before-advice-mismatch))
|
|
(and (ad-verify-cache-class-id
|
|
(car cache-id) (ad-get-advice-info-field function 'before))
|
|
(setq code 'around-advice-mismatch)
|
|
(ad-verify-cache-class-id
|
|
(nth 1 cache-id) (ad-get-advice-info-field function 'around))
|
|
(setq code 'after-advice-mismatch)
|
|
(ad-verify-cache-class-id
|
|
(nth 2 cache-id) (ad-get-advice-info-field function 'after))
|
|
(setq code 'definition-type-mismatch)
|
|
(let ((original-definition (ad-real-orig-definition function))
|
|
(cached-definition (ad-get-cache-definition function)))
|
|
(and (eq (nth 3 cache-id) (ad-definition-type original-definition))
|
|
(setq code 'arglist-mismatch)
|
|
(equal (if (eq (nth 4 cache-id) t)
|
|
(ad-arglist original-definition function)
|
|
(nth 4 cache-id) )
|
|
(ad-arglist cached-definition))
|
|
(setq code 'interactive-form-mismatch)
|
|
(or (null (nth 5 cache-id))
|
|
(equal (interactive-form original-definition)
|
|
(interactive-form cached-definition)))
|
|
(setq code 'verified))))
|
|
code))
|
|
|
|
(defun ad-verify-cache-id (function)
|
|
"True if FUNCTION's cache-id is compatible with its current advices."
|
|
(eq (ad-cache-id-verification-code function) 'verified))
|
|
|
|
|
|
;; @@ Preactivation:
|
|
;; =================
|
|
;; Preactivation can be used to generate compiled advised definitions
|
|
;; at compile time without having to give up the dynamic runtime flexibility
|
|
;; of the advice mechanism. Preactivation is a special feature of `defadvice',
|
|
;; it involves the following steps:
|
|
;; - remembering the function's current state (definition and advice-info)
|
|
;; - advising it with the defined piece of advice
|
|
;; - clearing its cache
|
|
;; - generating an interpreted advised definition by activating it, this will
|
|
;; make use of all its current active advice and its current definition
|
|
;; - saving the so generated cached definition and id
|
|
;; - resetting the function's advice and definition state to what it was
|
|
;; before the preactivation
|
|
;; - Returning the saved definition and its id to be used in the expansion of
|
|
;; `defadvice' to assign it as an initial cache, hence it will be compiled
|
|
;; at time the `defadvice' gets compiled.
|
|
;; Naturally, for preactivation to be effective it has to be applied/compiled
|
|
;; at the right time, i.e., when the current state of advices and function
|
|
;; definition exactly reflects the state at activation time. Should that not
|
|
;; be the case, the precompiled definition will just be discarded and a new
|
|
;; advised definition will be generated.
|
|
|
|
(defun ad-preactivate-advice (function advice class position)
|
|
"Preactivate FUNCTION and returns the constructed cache."
|
|
(let* ((function-defined-p (fboundp function))
|
|
(old-definition
|
|
(if function-defined-p
|
|
(symbol-function function)))
|
|
(old-advice-info (ad-copy-advice-info function))
|
|
(ad-advised-functions ad-advised-functions))
|
|
(unwind-protect
|
|
(progn
|
|
(ad-add-advice function advice class position)
|
|
(ad-enable-advice function class (ad-advice-name advice))
|
|
(ad-clear-cache function)
|
|
(ad-activate function -1)
|
|
(if (and (ad-is-active function)
|
|
(ad-get-cache-definition function))
|
|
(list (ad-get-cache-definition function)
|
|
(ad-get-cache-id function))))
|
|
(ad-set-advice-info function old-advice-info)
|
|
;; Don't `fset' function to nil if it was previously unbound:
|
|
(if function-defined-p
|
|
(ad-safe-fset function old-definition)
|
|
(fmakunbound function)))))
|
|
|
|
|
|
;; @@ Freezing:
|
|
;; ============
|
|
;; Freezing transforms a `defadvice' into a redefining `defun/defmacro'
|
|
;; for the advised function without keeping any advice information. This
|
|
;; feature was jwz's idea: It generates a dumpable function definition
|
|
;; whose documentation can be written to the DOC file, and the generated
|
|
;; code does not need any Advice runtime support. Of course, frozen advices
|
|
;; cannot be undone.
|
|
|
|
;; Freezing only considers the advice of the particular `defadvice', other
|
|
;; already existing advices for the same function will be ignored. To ensure
|
|
;; proper interaction when an already advised function gets redefined with
|
|
;; a frozen advice, frozen advices always use the actual original definition
|
|
;; of the function, i.e., they are always at the core of the onion. E.g., if
|
|
;; an already advised function gets redefined with a frozen advice and then
|
|
;; unadvised, the frozen advice remains as the new definition of the function.
|
|
|
|
;; While multiple freeze advices for a single function or freeze-advising
|
|
;; of an already advised function are possible, they are better avoided,
|
|
;; because definition/compile/load ordering is relevant, and it becomes
|
|
;; incomprehensible pretty quickly.
|
|
|
|
(defun ad-make-freeze-definition (function advice class position)
|
|
(if (not (ad-has-proper-definition function))
|
|
(error
|
|
"ad-make-freeze-definition: `%s' is not yet defined"
|
|
function))
|
|
(let* ((name (ad-advice-name advice))
|
|
;; With a unique origname we can have multiple freeze advices
|
|
;; for the same function, each overloading the previous one:
|
|
(unique-origname
|
|
(intern (format "%s-%s-%s" (ad-make-origname function) class name)))
|
|
(orig-definition
|
|
;; If FUNCTION is already advised, we'll use its current origdef
|
|
;; as the original definition of the frozen advice:
|
|
(or (ad-get-orig-definition function)
|
|
(symbol-function function)))
|
|
(old-advice-info
|
|
(if (ad-is-advised function)
|
|
(ad-copy-advice-info function)))
|
|
(real-docstring-fn
|
|
(symbol-function 'ad-make-advised-definition-docstring))
|
|
(real-origname-fn
|
|
(symbol-function 'ad-make-origname))
|
|
(frozen-definition
|
|
(unwind-protect
|
|
(progn
|
|
;; Make sure we construct a proper docstring:
|
|
(ad-safe-fset 'ad-make-advised-definition-docstring
|
|
'ad-make-freeze-docstring)
|
|
;; Make sure `unique-origname' is used as the origname:
|
|
(ad-safe-fset 'ad-make-origname (lambda (x) unique-origname))
|
|
;; No we reset all current advice information to nil and
|
|
;; generate an advised definition that's solely determined
|
|
;; by ADVICE and the current origdef of FUNCTION:
|
|
(ad-set-advice-info function nil)
|
|
(ad-add-advice function advice class position)
|
|
;; The following will provide proper real docstrings as
|
|
;; well as a definition that will make the compiler happy:
|
|
(ad-set-orig-definition function orig-definition)
|
|
(ad-make-advised-definition function))
|
|
;; Restore the old advice state:
|
|
(ad-set-advice-info function old-advice-info)
|
|
;; Restore functions:
|
|
(ad-safe-fset
|
|
'ad-make-advised-definition-docstring real-docstring-fn)
|
|
(ad-safe-fset 'ad-make-origname real-origname-fn))))
|
|
(if frozen-definition
|
|
(let* ((macro-p (ad-macro-p frozen-definition))
|
|
(body (cdr (if macro-p
|
|
(ad-lambdafy frozen-definition)
|
|
frozen-definition))))
|
|
`(progn
|
|
(if (not (fboundp ',unique-origname))
|
|
(fset ',unique-origname
|
|
;; avoid infinite recursion in case the function
|
|
;; we want to freeze is already advised:
|
|
(or (ad-get-orig-definition ',function)
|
|
(symbol-function ',function))))
|
|
(,(if macro-p 'defmacro 'defun)
|
|
,function
|
|
,@body))))))
|
|
|
|
|
|
;; @@ Activation and definition handling:
|
|
;; ======================================
|
|
|
|
(defun ad-should-compile (function compile)
|
|
"Return non-nil if the advised FUNCTION should be compiled.
|
|
If COMPILE is non-nil and not a negative number then it returns t.
|
|
If COMPILE is a negative number then it returns nil.
|
|
If COMPILE is nil then the result depends on the value of
|
|
`ad-default-compilation-action' (which see)."
|
|
(if (integerp compile)
|
|
(>= compile 0)
|
|
(if compile
|
|
compile
|
|
(cond ((eq ad-default-compilation-action 'never)
|
|
nil)
|
|
((eq ad-default-compilation-action 'always)
|
|
t)
|
|
((eq ad-default-compilation-action 'like-original)
|
|
(or (ad-subr-p (ad-get-orig-definition function))
|
|
(ad-compiled-p (ad-get-orig-definition function))))
|
|
;; everything else means `maybe':
|
|
(t (featurep 'byte-compile))))))
|
|
|
|
(defun ad-activate-advised-definition (function compile)
|
|
"Redefine FUNCTION with its advised definition from cache or scratch.
|
|
The resulting FUNCTION will be compiled if `ad-should-compile' returns t.
|
|
The current definition and its cache-id will be put into the cache."
|
|
(let ((verified-cached-definition
|
|
(if (ad-verify-cache-id function)
|
|
(ad-get-cache-definition function))))
|
|
(ad-safe-fset function
|
|
(or verified-cached-definition
|
|
(ad-make-advised-definition function)))
|
|
(if (ad-should-compile function compile)
|
|
(ad-compile-function function))
|
|
(if verified-cached-definition
|
|
(if (not (eq verified-cached-definition (symbol-function function)))
|
|
;; we must have compiled, cache the compiled definition:
|
|
(ad-set-cache
|
|
function (symbol-function function) (ad-get-cache-id function)))
|
|
;; We created a new advised definition, cache it with a proper id:
|
|
(ad-clear-cache function)
|
|
;; ad-make-cache-id needs the new cached definition:
|
|
(ad-set-cache function (symbol-function function) nil)
|
|
(ad-set-cache
|
|
function (symbol-function function) (ad-make-cache-id function)))))
|
|
|
|
(defun ad-handle-definition (function)
|
|
"Handle re/definition of an advised FUNCTION during de/activation.
|
|
If FUNCTION does not have an original definition associated with it and
|
|
the current definition is usable, then it will be stored as FUNCTION's
|
|
original definition. If no current definition is available (even in the
|
|
case of undefinition) nothing will be done. In the case of redefinition
|
|
the action taken depends on the value of `ad-redefinition-action' (which
|
|
see). Redefinition occurs when FUNCTION already has an original definition
|
|
associated with it but got redefined with a new definition and then
|
|
de/activated. If you do not like the current redefinition action change
|
|
the value of `ad-redefinition-action' and de/activate again."
|
|
(let ((original-definition (ad-get-orig-definition function))
|
|
(current-definition (if (ad-real-definition function)
|
|
(symbol-function function))))
|
|
(if original-definition
|
|
(if current-definition
|
|
(if (and (not (eq current-definition original-definition))
|
|
;; Redefinition with an advised definition from a
|
|
;; different function won't count as such:
|
|
(not (ad-advised-definition-p current-definition)))
|
|
;; we have a redefinition:
|
|
(if (not (memq ad-redefinition-action '(accept discard warn)))
|
|
(error "ad-handle-definition (see its doc): `%s' %s"
|
|
function "invalidly redefined")
|
|
(if (eq ad-redefinition-action 'discard)
|
|
(ad-safe-fset function original-definition)
|
|
(ad-set-orig-definition function current-definition)
|
|
(if (eq ad-redefinition-action 'warn)
|
|
(message "ad-handle-definition: `%s' got redefined"
|
|
function))))
|
|
;; either advised def or correct original is in place:
|
|
nil)
|
|
;; we have an undefinition, ignore it:
|
|
nil)
|
|
(if current-definition
|
|
;; we have a first definition, save it as original:
|
|
(ad-set-orig-definition function current-definition)
|
|
;; we don't have anything noteworthy:
|
|
nil))))
|
|
|
|
|
|
;; @@ The top-level advice interface:
|
|
;; ==================================
|
|
|
|
;;;###autoload
|
|
(defun ad-activate (function &optional compile)
|
|
"Activate all the advice information of an advised FUNCTION.
|
|
If FUNCTION has a proper original definition then an advised
|
|
definition will be generated from FUNCTION's advice info and the
|
|
definition of FUNCTION will be replaced with it. If a previously
|
|
cached advised definition was available, it will be used.
|
|
The optional COMPILE argument determines whether the resulting function
|
|
or a compilable cached definition will be compiled. If it is negative
|
|
no compilation will be performed, if it is positive or otherwise non-nil
|
|
the resulting function will be compiled, if it is nil the behavior depends
|
|
on the value of `ad-default-compilation-action' (which see).
|
|
Activation of an advised function that has an advice info but no actual
|
|
pieces of advice is equivalent to a call to `ad-unadvise'. Activation of
|
|
an advised function that has actual pieces of advice but none of them are
|
|
enabled is equivalent to a call to `ad-deactivate'. The current advised
|
|
definition will always be cached for later usage."
|
|
(interactive
|
|
(list (ad-read-advised-function "Activate advice of")
|
|
current-prefix-arg))
|
|
(if ad-activate-on-top-level
|
|
;; avoid recursive calls to `ad-activate':
|
|
(ad-with-auto-activation-disabled
|
|
(if (not (ad-is-advised function))
|
|
(error "ad-activate: `%s' is not advised" function)
|
|
(ad-handle-definition function)
|
|
;; Just return for forward advised and not yet defined functions:
|
|
(if (ad-get-orig-definition function)
|
|
(if (not (ad-has-any-advice function))
|
|
(ad-unadvise function)
|
|
;; Otherwise activate the advice:
|
|
(cond ((ad-has-redefining-advice function)
|
|
(ad-activate-advised-definition function compile)
|
|
(ad-set-advice-info-field function 'active t)
|
|
(eval (ad-make-hook-form function 'activation))
|
|
function)
|
|
;; Here we are if we have all disabled advices:
|
|
(t (ad-deactivate function)))))))))
|
|
|
|
(defalias 'ad-activate-on 'ad-activate)
|
|
|
|
(defun ad-deactivate (function)
|
|
"Deactivate the advice of an actively advised FUNCTION.
|
|
If FUNCTION has a proper original definition, then the current
|
|
definition of FUNCTION will be replaced with it. All the advice
|
|
information will still be available so it can be activated again with
|
|
a call to `ad-activate'."
|
|
(interactive
|
|
(list (ad-read-advised-function "Deactivate advice of" 'ad-is-active)))
|
|
(if (not (ad-is-advised function))
|
|
(error "ad-deactivate: `%s' is not advised" function)
|
|
(cond ((ad-is-active function)
|
|
(ad-handle-definition function)
|
|
(if (not (ad-get-orig-definition function))
|
|
(error "ad-deactivate: `%s' has no original definition"
|
|
function)
|
|
(ad-safe-fset function (ad-get-orig-definition function))
|
|
(ad-set-advice-info-field function 'active nil)
|
|
(eval (ad-make-hook-form function 'deactivation))
|
|
function)))))
|
|
|
|
(defun ad-update (function &optional compile)
|
|
"Update the advised definition of FUNCTION if its advice is active.
|
|
See `ad-activate' for documentation on the optional COMPILE argument."
|
|
(interactive
|
|
(list (ad-read-advised-function
|
|
"Update advised definition of" 'ad-is-active)))
|
|
(if (ad-is-active function)
|
|
(ad-activate function compile)))
|
|
|
|
(defun ad-unadvise (function)
|
|
"Deactivate FUNCTION and then remove all its advice information.
|
|
If FUNCTION was not advised this will be a noop."
|
|
(interactive
|
|
(list (ad-read-advised-function "Unadvise function")))
|
|
(cond ((ad-is-advised function)
|
|
(if (ad-is-active function)
|
|
(ad-deactivate function))
|
|
(ad-clear-orig-definition function)
|
|
(ad-set-advice-info function nil)
|
|
(ad-pop-advised-function function))))
|
|
|
|
(defun ad-recover (function)
|
|
"Try to recover FUNCTION's original definition, and unadvise it.
|
|
This is more low-level than `ad-unadvise' in that it does not do
|
|
deactivation, which might run hooks and get into other trouble.
|
|
Use in emergencies."
|
|
;; Use more primitive interactive behavior here: Accept any symbol that's
|
|
;; currently defined in obarray, not necessarily with a function definition:
|
|
(interactive
|
|
(list (intern
|
|
(completing-read "Recover advised function: " obarray nil t))))
|
|
(cond ((ad-is-advised function)
|
|
(cond ((ad-get-orig-definition function)
|
|
(ad-safe-fset function (ad-get-orig-definition function))
|
|
(ad-clear-orig-definition function)))
|
|
(ad-set-advice-info function nil)
|
|
(ad-pop-advised-function function))))
|
|
|
|
(defun ad-activate-regexp (regexp &optional compile)
|
|
"Activate functions with an advice name containing a REGEXP match.
|
|
This activates the advice for each function
|
|
that has at least one piece of advice whose name includes a match for REGEXP.
|
|
See `ad-activate' for documentation on the optional COMPILE argument."
|
|
(interactive
|
|
(list (ad-read-regexp "Activate via advice regexp")
|
|
current-prefix-arg))
|
|
(ad-do-advised-functions (function)
|
|
(if (ad-find-some-advice function 'any regexp)
|
|
(ad-activate function compile))))
|
|
|
|
(defun ad-deactivate-regexp (regexp)
|
|
"Deactivate functions with an advice name containing REGEXP match.
|
|
This deactivates the advice for each function
|
|
that has at least one piece of advice whose name includes a match for REGEXP."
|
|
(interactive
|
|
(list (ad-read-regexp "Deactivate via advice regexp")))
|
|
(ad-do-advised-functions (function)
|
|
(if (ad-find-some-advice function 'any regexp)
|
|
(ad-deactivate function))))
|
|
|
|
(defun ad-update-regexp (regexp &optional compile)
|
|
"Update functions with an advice name containing a REGEXP match.
|
|
This reactivates the advice for each function
|
|
that has at least one piece of advice whose name includes a match for REGEXP.
|
|
See `ad-activate' for documentation on the optional COMPILE argument."
|
|
(interactive
|
|
(list (ad-read-regexp "Update via advice regexp")
|
|
current-prefix-arg))
|
|
(ad-do-advised-functions (function)
|
|
(if (ad-find-some-advice function 'any regexp)
|
|
(ad-update function compile))))
|
|
|
|
(defun ad-activate-all (&optional compile)
|
|
"Activate all currently advised functions.
|
|
See `ad-activate' for documentation on the optional COMPILE argument."
|
|
(interactive "P")
|
|
(ad-do-advised-functions (function)
|
|
(ad-activate function compile)))
|
|
|
|
(defun ad-deactivate-all ()
|
|
"Deactivate all currently advised functions."
|
|
(interactive)
|
|
(ad-do-advised-functions (function)
|
|
(ad-deactivate function)))
|
|
|
|
(defun ad-update-all (&optional compile)
|
|
"Update all currently advised functions.
|
|
With prefix argument, COMPILE resulting advised definitions."
|
|
(interactive "P")
|
|
(ad-do-advised-functions (function)
|
|
(ad-update function compile)))
|
|
|
|
(defun ad-unadvise-all ()
|
|
"Unadvise all currently advised functions."
|
|
(interactive)
|
|
(ad-do-advised-functions (function)
|
|
(ad-unadvise function)))
|
|
|
|
(defun ad-recover-all ()
|
|
"Recover all currently advised functions. Use in emergencies.
|
|
To recover a function means to try to find its original (pre-advice)
|
|
definition, and delete all advice.
|
|
This is more low-level than `ad-unadvise' in that it does not do
|
|
deactivation, which might run hooks and get into other trouble."
|
|
(interactive)
|
|
(ad-do-advised-functions (function)
|
|
(condition-case nil
|
|
(ad-recover function)
|
|
(error nil))))
|
|
|
|
|
|
;; Completion alist of valid `defadvice' flags
|
|
(defvar ad-defadvice-flags
|
|
'(("protect") ("disable") ("activate")
|
|
("compile") ("preactivate") ("freeze")))
|
|
|
|
;;;###autoload
|
|
(defmacro defadvice (function args &rest body)
|
|
"Define a piece of advice for FUNCTION (a symbol).
|
|
The syntax of `defadvice' is as follows:
|
|
|
|
\(defadvice FUNCTION (CLASS NAME [POSITION] [ARGLIST] FLAG...)
|
|
[DOCSTRING] [INTERACTIVE-FORM]
|
|
BODY...)
|
|
|
|
FUNCTION ::= Name of the function to be advised.
|
|
CLASS ::= `before' | `around' | `after' | `activation' | `deactivation'.
|
|
NAME ::= Non-nil symbol that names this piece of advice.
|
|
POSITION ::= `first' | `last' | NUMBER. Optional, defaults to `first',
|
|
see also `ad-add-advice'.
|
|
ARGLIST ::= An optional argument list to be used for the advised function
|
|
instead of the argument list of the original. The first one found in
|
|
before/around/after-advices will be used.
|
|
FLAG ::= `protect'|`disable'|`activate'|`compile'|`preactivate'|`freeze'.
|
|
All flags can be specified with unambiguous initial substrings.
|
|
DOCSTRING ::= Optional documentation for this piece of advice.
|
|
INTERACTIVE-FORM ::= Optional interactive form to be used for the advised
|
|
function. The first one found in before/around/after-advices will be used.
|
|
BODY ::= Any s-expression.
|
|
|
|
Semantics of the various flags:
|
|
`protect': The piece of advice will be protected against non-local exits in
|
|
any code that precedes it. If any around-advice of a function is protected
|
|
then automatically all around-advices will be protected (the complete onion).
|
|
|
|
`activate': All advice of FUNCTION will be activated immediately if
|
|
FUNCTION has been properly defined prior to this application of `defadvice'.
|
|
|
|
`compile': In conjunction with `activate' specifies that the resulting
|
|
advised function should be compiled.
|
|
|
|
`disable': The defined advice will be disabled, hence, it will not be used
|
|
during activation until somebody enables it.
|
|
|
|
`preactivate': Preactivates the advised FUNCTION at macro-expansion/compile
|
|
time. This generates a compiled advised definition according to the current
|
|
advice state that will be used during activation if appropriate. Only use
|
|
this if the `defadvice' gets actually compiled.
|
|
|
|
`freeze': Expands the `defadvice' into a redefining `defun/defmacro' according
|
|
to this particular single advice. No other advice information will be saved.
|
|
Frozen advices cannot be undone, they behave like a hard redefinition of
|
|
the advised function. `freeze' implies `activate' and `preactivate'. The
|
|
documentation of the advised function can be dumped onto the `DOC' file
|
|
during preloading.
|
|
|
|
See Info node `(elisp)Advising Functions' for comprehensive documentation.
|
|
usage: (defadvice FUNCTION (CLASS NAME [POSITION] [ARGLIST] FLAG...)
|
|
[DOCSTRING] [INTERACTIVE-FORM]
|
|
BODY...)"
|
|
(declare (doc-string 3))
|
|
(if (not (ad-name-p function))
|
|
(error "defadvice: Invalid function name: %s" function))
|
|
(let* ((class (car args))
|
|
(name (if (not (ad-class-p class))
|
|
(error "defadvice: Invalid advice class: %s" class)
|
|
(nth 1 args)))
|
|
(position (if (not (ad-name-p name))
|
|
(error "defadvice: Invalid advice name: %s" name)
|
|
(setq args (nthcdr 2 args))
|
|
(if (ad-position-p (car args))
|
|
(prog1 (car args)
|
|
(setq args (cdr args))))))
|
|
(arglist (if (listp (car args))
|
|
(prog1 (car args)
|
|
(setq args (cdr args)))))
|
|
(flags
|
|
(mapcar
|
|
(function
|
|
(lambda (flag)
|
|
(let ((completion
|
|
(try-completion (symbol-name flag) ad-defadvice-flags)))
|
|
(cond ((eq completion t) flag)
|
|
((assoc completion ad-defadvice-flags)
|
|
(intern completion))
|
|
(t (error "defadvice: Invalid or ambiguous flag: %s"
|
|
flag))))))
|
|
args))
|
|
(advice (ad-make-advice
|
|
name (memq 'protect flags)
|
|
(not (memq 'disable flags))
|
|
`(advice lambda ,arglist ,@body)))
|
|
(preactivation (if (memq 'preactivate flags)
|
|
(ad-preactivate-advice
|
|
function advice class position))))
|
|
;; Now for the things to be done at evaluation time:
|
|
(if (memq 'freeze flags)
|
|
;; jwz's idea: Freeze the advised definition into a dumpable
|
|
;; defun/defmacro whose docs can be written to the DOC file:
|
|
(ad-make-freeze-definition function advice class position)
|
|
;; the normal case:
|
|
`(progn
|
|
(ad-add-advice ',function ',advice ',class ',position)
|
|
,@(if preactivation
|
|
`((ad-set-cache
|
|
',function
|
|
;; the function will get compiled:
|
|
,(cond ((ad-macro-p (car preactivation))
|
|
`(ad-macrofy
|
|
(function
|
|
,(ad-lambdafy
|
|
(car preactivation)))))
|
|
(t `(function
|
|
,(car preactivation))))
|
|
',(car (cdr preactivation)))))
|
|
,@(if (memq 'activate flags)
|
|
`((ad-activate ',function
|
|
,(if (memq 'compile flags) t))))
|
|
',function))))
|
|
|
|
|
|
;; @@ Tools:
|
|
;; =========
|
|
|
|
(defmacro ad-with-originals (functions &rest body)
|
|
"Binds FUNCTIONS to their original definitions and execute BODY.
|
|
For any members of FUNCTIONS that are not currently advised the rebinding will
|
|
be a noop. Any modifications done to the definitions of FUNCTIONS will be
|
|
undone on exit of this macro."
|
|
(let* ((index -1)
|
|
;; Make let-variables to store current definitions:
|
|
(current-bindings
|
|
(mapcar (function
|
|
(lambda (function)
|
|
(setq index (1+ index))
|
|
(list (intern (format "ad-oRiGdEf-%d" index))
|
|
`(symbol-function ',function))))
|
|
functions)))
|
|
`(let ,current-bindings
|
|
(unwind-protect
|
|
(progn
|
|
,@(progn
|
|
;; Make forms to redefine functions to their
|
|
;; original definitions if they are advised:
|
|
(setq index -1)
|
|
(mapcar
|
|
(function
|
|
(lambda (function)
|
|
(setq index (1+ index))
|
|
`(ad-safe-fset
|
|
',function
|
|
(or (ad-get-orig-definition ',function)
|
|
,(car (nth index current-bindings))))))
|
|
functions))
|
|
,@body)
|
|
,@(progn
|
|
;; Make forms to back-define functions to the definitions
|
|
;; they had outside this macro call:
|
|
(setq index -1)
|
|
(mapcar
|
|
(function
|
|
(lambda (function)
|
|
(setq index (1+ index))
|
|
`(ad-safe-fset
|
|
',function
|
|
,(car (nth index current-bindings)))))
|
|
functions))))))
|
|
|
|
(if (not (get 'ad-with-originals 'lisp-indent-hook))
|
|
(put 'ad-with-originals 'lisp-indent-hook 1))
|
|
|
|
|
|
;; @@ Advising `documentation':
|
|
;; ============================
|
|
;; Use the advice mechanism to advise `documentation' to make it
|
|
;; generate proper documentation strings for advised definitions:
|
|
|
|
;; This makes sure we get the right arglist for `documentation'
|
|
;; during bootstrapping.
|
|
(ad-define-subr-args 'documentation '(function &optional raw))
|
|
|
|
;; @@ Starting, stopping and recovering from the advice package magic:
|
|
;; ===================================================================
|
|
|
|
(defun ad-start-advice ()
|
|
"Start the automatic advice handling magic."
|
|
(interactive)
|
|
;; Advising `ad-activate-internal' means death!!
|
|
(ad-set-advice-info 'ad-activate-internal nil)
|
|
(ad-safe-fset 'ad-activate-internal 'ad-activate))
|
|
|
|
(defun ad-stop-advice ()
|
|
"Stop the automatic advice handling magic.
|
|
You should only need this in case of Advice-related emergencies."
|
|
(interactive)
|
|
;; Advising `ad-activate-internal' means death!!
|
|
(ad-set-advice-info 'ad-activate-internal nil)
|
|
(ad-safe-fset 'ad-activate-internal 'ad-activate-internal-off))
|
|
|
|
(defun ad-recover-normality ()
|
|
"Undo all advice related redefinitions and unadvises everything.
|
|
Use only in REAL emergencies."
|
|
(interactive)
|
|
;; Advising `ad-activate-internal' means death!!
|
|
(ad-set-advice-info 'ad-activate-internal nil)
|
|
(ad-safe-fset 'ad-activate-internal 'ad-activate-internal-off)
|
|
(ad-recover-all)
|
|
(setq ad-advised-functions nil))
|
|
|
|
(ad-start-advice)
|
|
|
|
(provide 'advice)
|
|
|
|
;; arch-tag: 29f8c9a1-8c88-471f-95d7-e28541c6b7c0
|
|
;;; advice.el ends here
|