R help - Jul 2009 - assignments to function return values

colnames(m) = c("a", "b")

I am fairly new to R and trying to understand this language. Having
learned quite a few other programming languages the above statement
when i saw it first immediately led to two reactions:

(1) wtf?

(2) maybe they made the function return an object with overloaded
assignment operator, lets try this:

x = colnames(m)
x = c("c","d")

but this doesnt work this way and obviously there must be some other
mechanisms at work here. I opened the r-intro document
http://cran.r-project.org/doc/manuals/R-intro.html and read through
chapter 10 (Writing your own functions) but i didnt see one single
mention of this strange syntax, yet it is used in some example code
here (10.6.2) but the reader is left with a big question mark in his
mind and no single word mentions this outstanding unusualness while
all other aspects of functions that are pretty common with most other
languages in existance are dealt with in great depth.

I then searched for some hints in the description of the assignment
operator and the asign() function but didnt find anything there also.

In my oppinion the R documentation could be greatly improved by giving
each chapter a list of links to all other chapters of the
documentation needed to understand the example code of this chapter
(all things that are not immediately obvious to someone coming from a
more mainstream language) and thus making each chapter a self
contained possible entry point into the complete documentation. This
would of course lead to some redundance for the documentation as a
whole but it would IMHO greatly improve the learning experience.

And here comes my question: How can the above mentioned syntax be
explained, how can i define such functions with can work in this
strange inverse manner, where should i have searched to find the
answer myself? Is there a document like "R for programmers" which
completely leaves out all the basic stuff and concentrates on the
obvious differences to most common mainstream languages only?

TIA
Bernd

On Sat, Jul 18, 2009 at 5:47 PM, Bernd<prof7bit at googlemail.com>
wrote:
> colnames(m) = c("a", "b")
>
> I am fairly new to R and trying to understand this language. Having
> learned quite a few other programming languages the above statement
> when i saw it first immediately led to two reactions:
>
> (1) wtf?
>
> (2) maybe they made the function return an object with overloaded
> assignment operator, lets try this:
>
> x = colnames(m)
> x = c("c","d")
>
> but this doesnt work this way and obviously there must be some other
> mechanisms at work here. I opened the r-intro document
> http://cran.r-project.org/doc/manuals/R-intro.html and read through
> chapter 10 (Writing your own functions) but i didnt see one single
> mention of this strange syntax, yet it is used in some example code
> here (10.6.2) but the reader is left with a big question mark in his
> mind and no single word mentions this outstanding unusualness while
> all other aspects of functions that are pretty common with most other
> languages in existance are dealt with in great depth.
>
> I then searched for some hints in the description of the assignment
> operator and the asign() function but didnt find anything there also.
>
> In my oppinion the R documentation could be greatly improved by giving
> each chapter a list of links to all other chapters of the
> documentation needed to understand the example code of this chapter
> (all things that are not immediately obvious to someone coming from a
> more mainstream language) and thus making each chapter a self
> contained possible entry point into the complete documentation. This
> would of course lead to some redundance for the documentation as a
> whole but it would IMHO greatly improve the learning experience.
>
> And here comes my question: How can the above mentioned syntax be
> explained, how can i define such functions with can work in this
> strange inverse manner, where should i have searched to find the
> answer myself? Is there a document like "R for programmers" which
> completely leaves out all the basic stuff and concentrates on the
> obvious differences to most common mainstream languages only?
>
 It's a sprinkle of syntactic sugar:

 colnames(x)=c("a","b")

is the same as:

 x = "colnames<-"(x,c("a","b"))

So there is a function called "colnames<-". See it by typing:

get("colnames<-")

and you'll see it has args x and value, and returns the modified x.

You can write your own foo<- functions and they operate like that.

 This is probably documented somewhere...

Barry

To get help on a "replacement function" try this:

?"colnames<-"

and to see its source:

`colnames<-`


On Sat, Jul 18, 2009 at 12:47 PM, Bernd<prof7bit at googlemail.com>
wrote:
> colnames(m) = c("a", "b")
>
> I am fairly new to R and trying to understand this language. Having
> learned quite a few other programming languages the above statement
> when i saw it first immediately led to two reactions:
>
> (1) wtf?
>
> (2) maybe they made the function return an object with overloaded
> assignment operator, lets try this:
>
> x = colnames(m)
> x = c("c","d")
>
> but this doesnt work this way and obviously there must be some other
> mechanisms at work here. I opened the r-intro document
> http://cran.r-project.org/doc/manuals/R-intro.html and read through
> chapter 10 (Writing your own functions) but i didnt see one single
> mention of this strange syntax, yet it is used in some example code
> here (10.6.2) but the reader is left with a big question mark in his
> mind and no single word mentions this outstanding unusualness while
> all other aspects of functions that are pretty common with most other
> languages in existance are dealt with in great depth.
>
> I then searched for some hints in the description of the assignment
> operator and the asign() function but didnt find anything there also.
>
> In my oppinion the R documentation could be greatly improved by giving
> each chapter a list of links to all other chapters of the
> documentation needed to understand the example code of this chapter
> (all things that are not immediately obvious to someone coming from a
> more mainstream language) and thus making each chapter a self
> contained possible entry point into the complete documentation. This
> would of course lead to some redundance for the documentation as a
> whole but it would IMHO greatly improve the learning experience.
>
> And here comes my question: How can the above mentioned syntax be
> explained, how can i define such functions with can work in this
> strange inverse manner, where should i have searched to find the
> answer myself? Is there a document like "R for programmers" which
> completely leaves out all the basic stuff and concentrates on the
> obvious differences to most common mainstream languages only?
>
> TIA
> Bernd
>
> ______________________________________________
> R-help at r-project.org mailing list
> 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.
>