Good morning, Based on feedback from folks, changes since the initial 0.2.1 announcement:
* Added the ability to suppress zero or more advice functions via a `without-advice` macro. * Changed the keyword used to indicate that a function should never be advised to be namespaced. https://github.com/thunknyc/richelieu Below is the example code from the README. (require '[richelieu.core :refer :all]) ;;; Here are some simple functions. (defn add [& xs] (apply + xs)) (defn mult [& xs] (apply * xs)) (defn sum-squares [& xs] (apply add (map #(mult % %) xs))) ;;; `defadvice` is just a way to use `defn` with ':richelieu/no-advice`;;; metadata to prevent crazy infinite advice loops. (defadvice plus1 "Adds one to each incoming argument, does nothing to the output." [f & xs] (apply f (map inc xs))) (defadvice times2 "Multiplies each incoming argument by two, does nothing to the output." [f & xs] (apply f (map (partial * 2) xs))) ;;; You can advise raw functions. (def add* (-> add (advise plus1) (advise times2))) (add* 1 1) ;;; But more often, you'll want to trace vars, which is what the rest;;; of the example deals with. ;;; This tracing advice shows how to get the current advised object,;;; which can either be a var or a function value, depending on the;;; context in which the advice was added. (def ^:dynamic *trace-depth* 0) (defn- ^:richelieu.core/no-advice trace-indent [] (apply str (repeat *trace-depth* \space))) (defadvice trace "Writes passed arguments and passes them to underlying function. Writes resulting value before returning it as result." [f & args] (printf "%s> %s %s\n" (trace-indent) *current-advised* args) (let [res (binding [*trace-depth* (inc *trace-depth*)] (apply f args))] (printf "%s< %s %s\n" (trace-indent) *current-advised* res) res)) (advise-var #'add trace) (unadvise-var #'add trace) ;;; This is safe because we used `defadvice` to prevent trace from;;; advising itself--or other advice functions. (advise-ns 'user trace) (sum-squares 1 2 3 4);;; The above invocation produces the following output: ;; > #'user/sum-squares (1 2 3 4);; > #'user/mult (1 1);; < #'user/mult 1;; > #'user/mult (2 2);; < #'user/mult 4;; > #'user/mult (3 3);; < #'user/mult 9;; > #'user/mult (4 4);; < #'user/mult 16;; > #'user/add (1 4 9 16);; < #'user/add 30;; < #'user/sum-squares 30 ;;; You can also suppress the evalutation of advice with the;;; `without-advice` macro. For example, the following will produce;;; no tracing output, but will allow any other advice that (possibly;;; someone else) attached to any function. (without-advice [trace] (sum-squares (1 2 3 4))) ;; ==> 30, no tracing ;;; Finally, it will often be a good idea to refer to advice functions;;; via var quoting instead of by a simple reference. This will allow;;; you to redefine them during development and still add or remove old;;; versions of attached advice functions, because they will be;;; associated with the var, not the particular function value that the;;; var pointed to at the time. (advise-ns 'user trace) ;; This works great until you re-eval ;; your `(defadvice trace ...)` form. (advise-ns 'user #'trace) ;; Infinitesimally slower but highly ;; recommended. On Tuesday, December 2, 2014 9:39:45 PM UTC-5, Edwin Watkeys wrote: > > I think you're right; I was going back and forth on that. > > -- > Edwin Watkeys, 917-324-2435 > > > On Dec 2, 2014, at 20:52, Atamert Ölçgen <mu...@muhuk.com> wrote: > > Just a small suggestion; I would make :unadvisable namespaced since it's > richelieu specific. > > On Wed, Dec 3, 2014 at 6:01 AM, Gary Verhaegen <gary.verhae...@gmail.com> > wrote: > >> At a very superficial glance, it looks like dire also sort of fits into >> the same space: https://github.com/MichaelDrogalis/dire >> >> >> On Tuesday, 2 December 2014, Edwin Watkeys <e...@poseur.com> wrote: >> >>> Phillip, >>> >>> Of Robert Hooke's features, I think the ability to suppress advice >>> temporarily (its `with-hooks-disabled`) as well to advise a function within >>> a particular dynamic scope (`with-scope`) are most relevant to Richelieu. >>> Since one of the major goals of Richelieu is to serve as a generic basis >>> for advising, I'll probably implement a `with-advice-disabled` form that >>> takes a sequence of advisedf-and-advicef-set values to temporarily suppress. >>> >>> R.H. and Richelieu, while they do much the same thing, seem to be >>> orthogonal to each other in terms of intent: Richelieu exists to allow >>> folks to write arbitrary advice-based facilities that are oblivious to each >>> others' existences, decorating functions that weren't written with being >>> advised in mind—think tracing and profiling. Phil, on the other hand, >>> focused on providing a facility for developers who anticipate that their >>> code might be advised à la the Emacs hooks mechanism. Or not. I'm totally >>> speculating. >>> >>> As for the name, I guess I'm willing to overlook some semantic >>> quibbles—especially since something very similar to Emacs's normal hooks >>> could easily built atop R.H.—in pursuit of a charming allusion. >>> >>> Edwin >>> >>> On Tuesday, December 2, 2014 8:39:37 AM UTC-5, Phillip Lord wrote: >>>> >>>> >>>> I think yours might be nicer, to be honest, though, although Robert >>>> Hooke has some features yours doesn't. Advising entire namespaces is an >>>> interesting addition for sure. >>>> >>>> I still don't understand why Robert Hooke has this name though. I can't >>>> have been the only person expecting it to implements hooks. >>>> >>>> Phil >>>> >>>> Edwin Watkeys <e...@poseur.com> writes: >>>> >>>> > Phillip, >>>> > >>>> > I’d cry if it weren’t so funny; I’ve just begun to make my way >>>> through the >>>> > lastest Read Eval Print λove and the first page or two dwells on >>>> reinvention. >>>> > At least mine wasn’t intentional. >>>> > >>>> > Edwin >>>> >>>> -- >>>> Phillip Lord, Phone: +44 (0) 191 208 7827 >>>> Lecturer in Bioinformatics, Email: >>>> philli...@newcastle.ac.uk >>>> School of Computing Science, http://homepages.cs.ncl.ac. >>>> uk/phillip.lord >>>> Room 914 Claremont Tower, skype: russet_apples >>>> Newcastle University, twitter: phillord >>>> NE1 7RU >>>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "Clojure" group. >>> To post to this group, send email to clojure@googlegroups.com >>> Note that posts from new members are moderated - please be patient with >>> your first post. >>> To unsubscribe from this group, send email to >>> clojure+unsubscr...@googlegroups.com >>> For more options, visit this group at >>> http://groups.google.com/group/clojure?hl=en >>> --- >>> You received this message because you are subscribed to the Google >>> Groups "Clojure" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to clojure+unsubscr...@googlegroups.com. >>> For more options, visit https://groups.google.com/d/optout. >>> >> -- >> You received this message because you are subscribed to the Google >> Groups "Clojure" group. >> To post to this group, send email to clojure@googlegroups.com >> Note that posts from new members are moderated - please be patient with >> your first post. >> To unsubscribe from this group, send email to >> clojure+unsubscr...@googlegroups.com >> For more options, visit this group at >> http://groups.google.com/group/clojure?hl=en >> --- >> You received this message because you are subscribed to the Google Groups >> "Clojure" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to clojure+unsubscr...@googlegroups.com. >> For more options, visit https://groups.google.com/d/optout. >> > > > > -- > Kind Regards, > Atamert Ölçgen > > -+- > --+ > +++ > > www.muhuk.com > > -- > You received this message because you are subscribed to the Google > Groups "Clojure" group. > To post to this group, send email to clojure@googlegroups.com > Note that posts from new members are moderated - please be patient with > your first post. > To unsubscribe from this group, send email to > clojure+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/clojure?hl=en > --- > You received this message because you are subscribed to a topic in the > Google Groups "Clojure" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/clojure/ycw4pZQBFfs/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > clojure+unsubscr...@googlegroups.com. > For more options, visit https://groups.google.com/d/optout. > > -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en --- You received this message because you are subscribed to the Google Groups "Clojure" group. To unsubscribe from this group and stop receiving emails from it, send an email to clojure+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
