R devel - Feb 2024 - [EXTERNAL] Re: NOTE: multiple local function definitions for ?fun? with different formal arguments

Because functions get called and therefore, the calling sequence matters. It?s
just protecting you from yourself, but as someone pointed out, there?s a way to
silence such notes.
G


From: Herv? Pag?s <hpages.on.github at gmail.com>
Sent: Tuesday, February 6, 2024 2:40 PM
To: Izmirlian, Grant (NIH/NCI) [E] <izmirlig at mail.nih.gov>; Duncan
Murdoch <murdoch.duncan at gmail.com>; r-devel at r-project.org
Subject: Re: [EXTERNAL] Re: [Rd] NOTE: multiple local function definitions for
?fun? with different formal arguments


On 2/6/24 11:19, Izmirlian, Grant (NIH/NCI) [E] wrote:
The note refers to the fact that the function named ?fun? appears to be defined
in two different ways.

Sure I get that. But how is that any different from a variable being defined in
two different ways like in

    if (mode == 1)
        x <- -8
    else
        x <- 55

This is such a common and perfectly fine pattern. Why would this be considered a
potential hazard when the variable is a function?

H.

From: Herv? Pag?s <hpages.on.github at
gmail.com><mailto:hpages.on.github at gmail.com>
Sent: Tuesday, February 6, 2024 2:17 PM
To: Duncan Murdoch <murdoch.duncan at gmail.com><mailto:murdoch.duncan
at gmail.com>; Izmirlian, Grant (NIH/NCI) [E] <izmirlig at
mail.nih.gov><mailto:izmirlig at mail.nih.gov>; r-devel at
r-project.org<mailto:r-devel at r-project.org>
Subject: [EXTERNAL] Re: [Rd] NOTE: multiple local function definitions for ?fun?
with different formal arguments


Thanks. Workarounds are interesting but... what's the point of the NOTE in
the first place?

H.
On 2/4/24 09:07, Duncan Murdoch wrote:
On 04/02/2024 10:55 a.m., Izmirlian, Grant (NIH/NCI) [E] via R-devel wrote:


Well you can see that yeast is exactly weekday you have.  The way out is to just
not name the result

I think something happened to your explanation...




toto <- function(mode)
{
     ifelse(mode == 1,
         function(a,b) a*b,
         function(u, v, w) (u + v) / w)
}

It's a bad idea to use ifelse() when you really want if() ... else ... .  In
this case it works, but it doesn't always.  So the workaround should be


toto <- function(mode)
{
    if(mode == 1)
        function(a,b) a*b
    else
        function(u, v, w) (u + v) / w
}






________________________________
From: Grant Izmirlian <izmirlidroid at gmail.com><mailto:izmirlidroid
at gmail.com>
Date: Sun, Feb 4, 2024, 10:44 AM
To: "Izmirlian, Grant (NIH/NCI) [E]" <izmirlig at
mail.nih.gov><mailto:izmirlig at mail.nih.gov>
Subject: Fwd: [EXTERNAL] R-devel Digest, Vol 252, Issue 2

Hi,

I just ran into this 'R CMD check' NOTE for the first time:

* checking R code for possible problems ... NOTE
toto: multiple local function definitions for ?fun? with different
   formal arguments

The "offending" code is something like this (simplified from the real
code):

toto <- function(mode)
{
     if (mode == 1)
         fun <- function(a, b) a*b
     else
         fun <- function(u, v, w) (u + v) / w
     fun
}

Is that NOTE really intended? Hard to see why this code would be
considered "wrong".

I know it's just a NOTE but still...

I agree it's a false positive, but the issue is that you have a function
object in your function which can't be called unconditionally.  The
workaround doesn't create such an object.

Recognizing that your function never tries to call fun requires global
inspection of toto(), and most of the checks are based on local inspection.

Duncan Murdoch

______________________________________________
R-devel at r-project.org<mailto:R-devel at r-project.org> mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel

--

Herv? Pag?s



Bioconductor Core Team

hpages.on.github at gmail.com<mailto:hpages.on.github at gmail.com>
CAUTION: This email originated from outside of the organization. Do not click
links or open attachments unless you recognize the sender and are confident the
content is safe.


--

Herv? Pag?s



Bioconductor Core Team

hpages.on.github at gmail.com<mailto:hpages.on.github at gmail.com>
CAUTION: This email originated from outside of the organization. Do not click
links or open attachments unless you recognize the sender and are confident the
content is safe.


	[[alternative HTML version deleted]]

Here's a dummy example that I think illustrates the problem:

toto <- function() {
  if (runif(1) < 0.5)
    function(a) a
  else
    function(a,b) a+b
}
> fcn <- toto()
> fcn(1,2)
[1] 3
> fcn <- toto()
> fcn(1,2)
[1] 3
> fcn <- toto()
> fcn(1,2)
Error in fcn(1, 2) : unused argument (2)

How can you use the returned function, if you get different arguments?

In your example, you cannot use the returned function without knowing
'mode', or by inspecting the returned function.  So, the warning is
there to alert you to a potential bug.  Anecdotally, I'm pretty sure
this R CMD check NOTE has caught at least one such bug in one of
my/our packages.

If you want to keep the current design pattern, one approach could be
to add ... to your function definitions:

toto <- function(mode)
{
     if (mode == 1)
         fun <- function(a, b, ...) a*b
     else
         fun <- function(u, v, w) (u + v) / w
     fun
}

to make sure that toto() returns functions that accept the same
minimal number of arguments.

/Henrik

On Tue, Feb 6, 2024 at 1:15?PM Izmirlian, Grant (NIH/NCI) [E] via
R-devel <r-devel at r-project.org> wrote:
>
> Because functions get called and therefore, the calling sequence matters.
It?s just protecting you from yourself, but as someone pointed out, there?s a
way to silence such notes.
> G
>
>
> From: Herv? Pag?s <hpages.on.github at gmail.com>
> Sent: Tuesday, February 6, 2024 2:40 PM
> To: Izmirlian, Grant (NIH/NCI) [E] <izmirlig at mail.nih.gov>; Duncan
Murdoch <murdoch.duncan at gmail.com>; r-devel at r-project.org
> Subject: Re: [EXTERNAL] Re: [Rd] NOTE: multiple local function definitions
for ?fun? with different formal arguments
>
>
> On 2/6/24 11:19, Izmirlian, Grant (NIH/NCI) [E] wrote:
> The note refers to the fact that the function named ?fun? appears to be
defined in two different ways.
>
> Sure I get that. But how is that any different from a variable being
defined in two different ways like in
>
>     if (mode == 1)
>         x <- -8
>     else
>         x <- 55
>
> This is such a common and perfectly fine pattern. Why would this be
considered a potential hazard when the variable is a function?
>
> H.
>
> From: Herv? Pag?s <hpages.on.github at
gmail.com><mailto:hpages.on.github at gmail.com>
> Sent: Tuesday, February 6, 2024 2:17 PM
> To: Duncan Murdoch <murdoch.duncan at
gmail.com><mailto:murdoch.duncan at gmail.com>; Izmirlian, Grant
(NIH/NCI) [E] <izmirlig at mail.nih.gov><mailto:izmirlig at
mail.nih.gov>; r-devel at r-project.org<mailto:r-devel at
r-project.org>
> Subject: [EXTERNAL] Re: [Rd] NOTE: multiple local function definitions for
?fun? with different formal arguments
>
>
> Thanks. Workarounds are interesting but... what's the point of the NOTE
in the first place?
>
> H.
> On 2/4/24 09:07, Duncan Murdoch wrote:
> On 04/02/2024 10:55 a.m., Izmirlian, Grant (NIH/NCI) [E] via R-devel wrote:
>
>
> Well you can see that yeast is exactly weekday you have.  The way out is to
just not name the result
>
> I think something happened to your explanation...
>
>
>
>
> toto <- function(mode)
> {
>      ifelse(mode == 1,
>          function(a,b) a*b,
>          function(u, v, w) (u + v) / w)
> }
>
> It's a bad idea to use ifelse() when you really want if() ... else ...
.  In this case it works, but it doesn't always.  So the workaround should
be
>
>
> toto <- function(mode)
> {
>     if(mode == 1)
>         function(a,b) a*b
>     else
>         function(u, v, w) (u + v) / w
> }
>
>
>
>
>
>
> ________________________________
> From: Grant Izmirlian <izmirlidroid at
gmail.com><mailto:izmirlidroid at gmail.com>
> Date: Sun, Feb 4, 2024, 10:44 AM
> To: "Izmirlian, Grant (NIH/NCI) [E]" <izmirlig at
mail.nih.gov><mailto:izmirlig at mail.nih.gov>
> Subject: Fwd: [EXTERNAL] R-devel Digest, Vol 252, Issue 2
>
> Hi,
>
> I just ran into this 'R CMD check' NOTE for the first time:
>
> * checking R code for possible problems ... NOTE
> toto: multiple local function definitions for ?fun? with different
>    formal arguments
>
> The "offending" code is something like this (simplified from the
real code):
>
> toto <- function(mode)
> {
>      if (mode == 1)
>          fun <- function(a, b) a*b
>      else
>          fun <- function(u, v, w) (u + v) / w
>      fun
> }
>
> Is that NOTE really intended? Hard to see why this code would be
> considered "wrong".
>
> I know it's just a NOTE but still...
>
> I agree it's a false positive, but the issue is that you have a
function object in your function which can't be called unconditionally.  The
workaround doesn't create such an object.
>
> Recognizing that your function never tries to call fun requires global
inspection of toto(), and most of the checks are based on local inspection.
>
> Duncan Murdoch
>
> ______________________________________________
> R-devel at r-project.org<mailto:R-devel at r-project.org> mailing
list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
> --
>
> Herv? Pag?s
>
>
>
> Bioconductor Core Team
>
> hpages.on.github at gmail.com<mailto:hpages.on.github at gmail.com>
> CAUTION: This email originated from outside of the organization. Do not
click links or open attachments unless you recognize the sender and are
confident the content is safe.
>
>
> --
>
> Herv? Pag?s
>
>
>
> Bioconductor Core Team
>
> hpages.on.github at gmail.com<mailto:hpages.on.github at gmail.com>
> CAUTION: This email originated from outside of the organization. Do not
click links or open attachments unless you recognize the sender and are
confident the content is safe.
>
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel