MacroPackage: exclToCDocOverviewCGDocRelNotesFAQIndexPermutedIndex
Allegro CL version 8.2
Unrevised from 8.1 to 8.2.
8.1 version

advise

Arguments: fspec class name position &rest forms

Starting in Allegro CL release 6.0, the new fwrapper facility, described in fwrappers-and-advice.htm, replaces the advice facility, and continued use of the advice facility is deprecated.

None of the arguments are evaluated. This macro advises the function fspec appropriately with code defined by the forms.

fspec specifies the function (or macro) to advise. This argument is normally a symbol but it can be any function name (see implementation.htm). See fwrappers-and-advice.htm for information on advising macros. The information in the remainder of this definition relates to functions only.

class specifies the type of advice. The three choices are :before, :around, and :after. :before advice will be evaluated before the advised function is called; :around advice will be evaluated around the calling of the advised function (in the sense that the call to the function will be embedded within the :around code); :after advice will be evaluated after the advised function has been called. See below for more information on classes of advice.

name specifies the name of the advice. Its value may be nil, but it is convenient to refer to advice by name. Note that there can be only one piece of advice of each class with a given name. (But there may be as many as you like without a name, i.e. with nil as the specified name.) If more advice is defined with the same name and class as existing advice, the new advice will replace the existing advice, thus taking its place.

position specifies the location of advice relative to other advice of the same class on the same function. The important point about this argument is that it is used only when the advice is first defined and is then forgotten. Therefore, the position is calculated with respect to already existing advice of the same class, whose positions are 0, 1, 2... regardless of the value of the position argument when they were defined. See fwrappers-and-advice.htm for examples that further explain this point.

There can be many pieces of advice of each class on a function. position specifies where the piece of advice currently being defined will go. position may be a non-negative integer, a name or nil. If it is a number, then the lower the number, the farther the advice will be from the actual evaluation of the forms defining the function itself. Thus for :before and :around advice, advice 0 will be run first, then 1, 2, 3, and so on, while for :after advice, advice 0 will be run last, preceded by 1, 2, 3, and so on. Say there are n pieces of advice of the class of the advice being defined. The existing advice is numbered 0, 1, 2, ..., n-1, inclusive. Then a position of 0 will result in the new advice being placed farther from the function call than all the rest of the advice, a position of k where k is between 1 and n-1 will be placed farther then the kth piece of existing advice, and a position n or larger will be placed nearest the function call.

If position is nil, the advice is put farther from the function call then all currently-defined advice of the same class. If position is a name of existing advice, the advice will be placed next farther from the function call (before for :before and :around, after for :after) than the named advice. Note that if new advice has the same name and class as existing advice, it will replace the existing advice and take the position of the existing advice. In that case, the position argument will be ignored.

forms are the actual forms that will be evaluated. All advice on a function is combined in an enclosing cl:prog, so a return placed anywhere in any type of advice will cause the advised function to return with the values specified with the return, without any further advice code being evaluated and (if it has not already done so) without the advised function being evaluated. Syntactically, :around advice should be a single form. If multiple forms are specified, however, they are automatically wrapped in a cl:progn. Warning: forms cannot be the single form nil. If you want the single form nil (in, for example, :around advice to not call the function and return nil), specify (progn nil) instead.

Advice of class :before

:before advice will be evaluated before the function call. The argument list with which the function was called is available to :before advice. The argument list is stored as the value of the variable excl:arglist (this symbol also names a function which returns the arguments list of an operator). You may check the values in this list, change the values or replace the list altogether with a new list. (If you replace the list, be careful that it have the correct format for the function -- number and types of arguments in the list -- or you may get an error, or worse, a wrong result but no error.) :before advice is used only for its side effects. Any value returned by :before advice code is ignored.

Advice of class :around

:around advice places the function call inside the code of the advice. The keyword :do-it signals where the function should be called within the advice code. When Lisp encounters the :do-it, it calls the next piece of :around advice, if there is more, or the function. When the function returns, the :around advice code continues execution. :do-it may appear several times in :around advice. Normally, it is placed in conditional code, as shown in this code fragment:

(advise myfunction :around example nil
  (... [do stuff before function call in :around advice code]
   (if (zerop (car excl:arglist)) (+ 5 :do-it) (* 7 :do-it))
   ... [do stuff after function call] ))

In that case, the system will encounter one or the other :do-it, but not both. However, it is allowed to have several :do-its all of which are evaluated. In that case, the succeeding :around advice and the advised function are evaluated more than once. :after advice is still evaluated only once, however. :around advice can work with excl:arglist before the :do-it. Since the advised function is run at the location specified by the :do-it, the values the function return are available to :around advice just like with any function call. The list bound to the variable values is not set up until after all :around advice is run, then values is bound to the list of values returned by the final piece of :around advice. Note that if the function returns multiple values, these should be caught with multiple-value-bind or multiple-value-list or some similar function if you are interested in the values beyond the first.

Advice of class :after

advice is evaluated after the function has completed, but before control is passed back to whatever called the function. :after advice may examine and change the list of returned values from the last piece of :around advice (or the function, if there is no :around advice) stored in the variable values. :after advice is used only for its side effects. Any value returned by the :after advice code is ignored. The list bound to values is returned from the now completed function call.

See also fwrappers-and-advice.htm for general information on the new fwrapper facility and the older, now deprecated, advice facility in Allegro CL.


Copyright (c) 1998-2016, Franz Inc. Oakland, CA., USA. All rights reserved.
This page was not revised from the 8.1 page.
Created 2010.1.21.

ToCDocOverviewCGDocRelNotesFAQIndexPermutedIndex
Allegro CL version 8.2
Unrevised from 8.1 to 8.2.
8.1 version