Re: [O] was this intentional with the default header ":results value"

[Eric Luo]

address@hidden (Thomas S. Dye) writes:

Thanks, it's clear to me with the difference between "value" and
"output" now.

whether there is a way to tell emacs-lisp that "123456789123456789" is a
string rather than a number.

emacs-lisp handles the output as expected if the c1 has any character
other than numbers as the following.
,----
| #+BEGIN_SRC R
|   df <- data.frame(c1="c123456789123456789",c2=2)
|   df$c1 <- as.vector(df$c1)
|   df
| #+END_SRC
| 
| #+RESULTS:
| | c123456789123456789 | 2 |
| 
`----

> Aloha Eric,
>
> Eric Luo <address@hidden> writes:
>
>> Hi, 
>> Why in the following code block, c1 was printed as scientific notation
>> rather then characters.
>> ,----
>> | 
>> | #+BEGIN_SRC R :results value
>> |   df <- data.frame(c1="123456789123456789",c2=2)
>> | #+END_SRC
>> | 
>> | #+RESULTS:
>> | | 1.2345678912345678e+17 | 2 |
>> `----
>
> With :results value the results are passed from the source language into
> emacs-lisp and then displayed in the buffer.  So, the representation of
> things like very large numbers depends on how that was done in
> emacs-lisp, independent of the source language conventions.  In this
> case, I suspect the number was written with a formatting string and ‘%g’
> which uses scientific notation if that is a shorter representation.
>
>
>> But if the header :results output, the result was expected.
>> ,----
>> | 
>> | #+BEGIN_SRC R :results output 
>> |   df <- data.frame(c1="123456789123456789",c2=2)
>> |   df
>> | #+END_SRC
>> | 
>> | #+RESULTS:
>> | :                   c1 c2
>> | : 1 123456789123456789  2
>> `----
>>
>
> With :results output the results aren't translated into emacs-lisp.
> Instead, the output from the source language is collected and displayed
> in the buffer.  Thus, the output conventions of the source language are
> respected. 
>
> hth,