Chris
In this case, you could write an auxiliary third function that takes
an additional Bool parameter. Both your functions call the third
function with this Bool parameter.
An alternative solution is to make this a Val{Bool} parameter, which
would likely specialize the functions at build time. This might
improve performance if the functions are called in a
performance-critial region.
-erik
On Mon, Jan 18, 2016 at 4:36 PM, Cedric St-Jean <[email protected]> wrote:
> my_func(fcn1::Function, passedIn::Float64) =
> my_func(fcn1, (y, z, passedin)->default_fcn(0.0, y, z, passedin),
> passedIn)
>
> You could achieve the same effect in one definition if you put fcn2 as a
> keyword argument. Also check out FastAnonymous.jl if performance matters.
>
>
> On Monday, January 18, 2016 at 4:25:20 PM UTC-5, Christopher Alexander
> wrote:
>>
>> Thanks for the response! As a follow-up, what would I do in a situation
>> where the passed-in second function (fcn2) and the default function take a
>> different number of arguments?
>>
>> Thanks!
>>
>> Chris
>>
>> On Monday, January 18, 2016 at 4:06:41 PM UTC-5, Erik Schnetter wrote:
>>>
>>> Define the second function like this:
>>> ```
>>> my_func(fcn1::Function, passedIn::Float64) = my_func(fcn1,
>>> default_fcn, passedIn)
>>> ```
>>>
>>> -erik
>>>
>>>
>>> On Mon, Jan 18, 2016 at 4:02 PM, Christopher Alexander
>>> <[email protected]> wrote:
>>> > Hello all, I had a question concerning a best practice in a particular
>>> > case
>>> > of multiple dispatch which is as follows.
>>> >
>>> > Let's say I have a function with two different methods.
>>> >
>>> > function my_func(fcn1::Function,fcn2::Function, passedIn::Float64)
>>> > x = 0.0
>>> > y = 1.0
>>> > z = 2.0
>>> > val1 = fcn(x, y, passedIn)
>>> > val2 = fcn2(y, z, passedIn)
>>> > return val1, val2
>>> > end
>>> >
>>> > function my_func(fcn1::Function, passedIn::Float64)
>>> > x = 0.0
>>> > y = 1.0
>>> > z = 2.0
>>> > val1 = fcn(x, y, passedIn)
>>> > val2 = default_fcn(x, y, z, passedIn)
>>> > return val1, val2
>>> > end
>>> >
>>> > My question is basically, what would be the best way to do this without
>>> > massive code duplication? The actual situation I am working with has
>>> > much
>>> > more going on in the function, so it's not like I could create some
>>> > init
>>> > function to set up x, y, & z. But literally the only different
>>> > behavior
>>> > between the two methods is whether or not a second function is passed
>>> > in.
>>> >
>>> > Thanks!
>>> >
>>> > Chris
>>> >
>>>
>>>
>>>
>>> --
>>> Erik Schnetter <[email protected]>
>>> http://www.perimeterinstitute.ca/personal/eschnetter/
--
Erik Schnetter <[email protected]>
http://www.perimeterinstitute.ca/personal/eschnetter/
