The theoretical assumption was that pattern matching on constants is much faster than function calls and branching (if you like some literature, this chapter of Jones' book <http://research.microsoft.com/en-us/um/people/simonpj/papers/slpj-book-1987/PAGES/078.HTM> provides some good insight). The empirical proof was running fprof <http://erlang.org/doc/man/fprof.html>, an erlang tool "used to profile a program to find out how the execution time is used".
Further analysis of the compiled bytecode or better yet, the actual machine code and CPU instructions produced by the BEAM, is left as an exercise to the reader. Max On 23 May 2016 at 19:32, Onorio Catenacci <[email protected]> wrote: > Definitely looks like empirical proof to me :) > > On Mon, May 23, 2016 at 12:35 PM, max s <[email protected]> wrote: > >> @Onorio >> >> You're right, we should not take assumptions before benchmarking even >> though I think my intuition was justified. I wrote a quick code that I >> profiled with fprof and I get 3~4 better performances using the matching on >> constants ; you can see the gist here >> https://gist.github.com/anonymous/3cf62215407b829da4ab013d111cebfe (let >> me know and correct me if there's a mistake in my profiling). >> >> @José >> >> Agreed and actually it'd be more in line with erlang philosophy of doing >> things >> >> Max >> >> On 23 May 2016 at 17:45, José Valim <[email protected]> >> wrote: >> >>> Why not use atoms? Atoms are precisely that: constants. >>> >>> >>> >>> *José Valim* >>> www.plataformatec.com.br >>> Skype: jv.ptec >>> Founder and Director of R&D >>> >>> On Sun, May 22, 2016 at 12:16 PM, Nytz12 <[email protected]> >>> wrote: >>> >>>> Hi >>>> >>>> I use the following pattern to define cross module constants >>>> >>>> defmodule Visitor do >>>> >>>> defmodule FilterTypes do >>>> def c_GREATER_THAN, do: "greater_than" >>>> def c_LESS_THAN, do: "less_than" >>>> def c_IS, do: "is" >>>> def c_IS_NOT, do: "is_not" >>>> end >>>> ... >>>> >>>> Which lets me do something like this from other modules >>>> >>>> Visitor.FilterTypes.c_GREATER_THAN # == "greater than" >>>> >>>> Problem is that I just discored that these 'constants', which are >>>> really function calls, cannot be used to do pattern match like: >>>> >>>> def filter(query, filter, field, value) do >>>> case filter do >>>> FilterTypes.c_GREATER_THAN -> greater_than(query, field, value) >>>> FilterTypes.c_LESS_THAN -> less_than(query, field, value) >>>> FilterTypes.c_IS -> is(query, field, value) >>>> FilterTypes.c_IS_NOT -> is_not(query, field, value) >>>> _ -> {:error, "unknown filter", filter} >>>> end >>>> end >>>> >>>> Is there a better approach i'm missing? >>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "elixir-lang-talk" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to [email protected]. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/elixir-lang-talk/ace6b869-dfce-4d3f-aeeb-2d9bc4583f03%40googlegroups.com >>>> <https://groups.google.com/d/msgid/elixir-lang-talk/ace6b869-dfce-4d3f-aeeb-2d9bc4583f03%40googlegroups.com?utm_medium=email&utm_source=footer> >>>> . >>>> For more options, visit https://groups.google.com/d/optout. >>>> >>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "elixir-lang-talk" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected]. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/elixir-lang-talk/CAGnRm4KmHUN5PHYgoV4hT9tyOnN--o9ysuRoofi5V%2B%3DfJSth8Q%40mail.gmail.com >>> <https://groups.google.com/d/msgid/elixir-lang-talk/CAGnRm4KmHUN5PHYgoV4hT9tyOnN--o9ysuRoofi5V%2B%3DfJSth8Q%40mail.gmail.com?utm_medium=email&utm_source=footer> >>> . >>> >>> For more options, visit https://groups.google.com/d/optout. >>> >> >> -- >> You received this message because you are subscribed to a topic in the >> Google Groups "elixir-lang-talk" group. >> To unsubscribe from this topic, visit >> https://groups.google.com/d/topic/elixir-lang-talk/l-7rK9_bT2Q/unsubscribe >> . >> To unsubscribe from this group and all its topics, send an email to >> [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/elixir-lang-talk/CAOPKHev-O0ZdDURxp144Xp4UrT-0FWECVYfNRwxyrg6HRkP0vQ%40mail.gmail.com >> <https://groups.google.com/d/msgid/elixir-lang-talk/CAOPKHev-O0ZdDURxp144Xp4UrT-0FWECVYfNRwxyrg6HRkP0vQ%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > > > -- > Onorio Catenacci > > http://onor.io > http://www.google.com/+OnorioCatenacci > > -- > You received this message because you are subscribed to the Google Groups > "elixir-lang-talk" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/elixir-lang-talk/CAP%3DvNq8U5rmL%3DxfZ2UT4VJ4hDOgPaUqzThKZoSmOO8FJjZLRhw%40mail.gmail.com > <https://groups.google.com/d/msgid/elixir-lang-talk/CAP%3DvNq8U5rmL%3DxfZ2UT4VJ4hDOgPaUqzThKZoSmOO8FJjZLRhw%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "elixir-lang-talk" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/elixir-lang-talk/CAOPKHes6Mhv%2ByU1UGKYVJOg%2BxHnW%2B-9z3YdJGDrKpZh4Y3YrOA%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
