[R] Calling stargazer() with do.call() in R 4.2.0

[Andrew Simmons]

Hello,


I don't have the slightest clue what stargazer is supposed to be
doing, but it seems as though it's trying to create names for the ...
list using:
object.names.string <- deparse(substitute(list(...)))

It makes an error in assuming that the return value of deparse will be
a character string. For stargazer(res), object.names.string becomes:
"list(res)"

while for do.call(stargazer, list(res)), object.names.string becomes:
[1] "list(structure(list(coefficients = c(`(Intercept)` =
6.41594246095523, "
[2] "UrbanPop = 0.0209346588197249), residuals = c(Alabama =
5.56984732750071, "
[3] "Alaska = 2.57919391569797, Arizona = 0.009284833466778, Arkansas
= 1.33732459805853, "
[4] "California = 0.679003586449805, Colorado = -0.148845848893771, "
[5] "Connecticut = -4.72791119007405, Delaware = -2.02323789597542, "
[6] "Florida = 7.30928483346678, Georgia = 9.72797800986128, Hawaii =
-2.8535191429924, "
[7] "Idaho = -4.94641403722037, Illinois = 2.2464808570076, Indiana =
-0.576695284237348, "
[8] "Iowa = -5.40921801367955, Kansas = -1.79762994305707, Kentucky =
2.19545528041907, "
[9] "Louisiana = 7.60237005694293, Maine = -5.3836100607612, Maryland
= 3.4814353981232, "
 [ reached getOption("max.print") -- omitted 110 entries ]

perhaps the package maintainer could change the line from:
object.names.string <- deparse(substitute(list(...)))
to:
object.names.string <- deparse1(substitute(list(...)), collapse = "")

or you could change your code to:
do.call(stargazer, alist(res))

please note that using alist instead of list is only a workaround, you
should still let the package maintainer know of this bug. If the
maintainer asks, this is what I used to get the strings above:
fun <- \(...) deparse(substitute(list(...)))
data("USArrests")
res <- lm( Murder ~ UrbanPop, data = USArrests)
fun(res)
print(do.call("fun", list(res)), max = 9)

On Sat, May 28, 2022 at 4:41 PM Arne Henningsen
<arne.henningsen using gmail.com> wrote:
>
> On Sat, 28 May 2022 at 01:21, Uwe Ligges
> <ligges using statistik.tu-dortmund.de> wrote:
> > On 27.05.2022 17:29, Arne Henningsen wrote:
> >> Dear all  (cc Marek = maintainer of the stargazer package)
> >>
> >> We use do.call() to automatically create many LaTeX tables with
> >> stargazer but after upgrading to R 4.2.0, this no longer works. I
> >> illustrate this with a simple reproducible example:
> >>
> >> R> data("USArrests")
> >> R> res <- lm( Murder ~ UrbanPop, data = USArrests )
> >> R> library(stargazer)
> >> R> stargazer(res)  # works as expected
> >> R> do.call( stargazer, list(res) )
> >> Error in if (is.na(s)) { : the condition has length > 1
> >
> > Without looking at the code in detail: The line aboce suggests the code
> > needs an any():    if(any(is.na(x))) raher than if(is.na(x)).
>
> Yes, this is likely a problem in the 'stargazer' package.
>
> ... but why does the problem occur when using do.call( stargazer, )
> but the problem does *not* occur when using stargazer() directly?
>
> Best regards,
> Arne
>
> >> Any ideas what we can do so that the last command works with R 4.2.0?
> >>
> >> /Arne
>
> --
> Arne Henningsen
> http://www.arne-henningsen.name
>
> ______________________________________________
> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.