Rainer


Rainer M Krug <Rainer@krugs.de> writes:

> "Charles C. Berry" <ccberry@ucsd.edu> writes:
>
>> On Wed, 8 Oct 2014, Rainer M Krug wrote:
>>
>>> "Charles C. Berry" <ccberry@ucsd.edu> writes:
>>>
>>>> On Wed, 8 Oct 2014, Rainer M Krug wrote:
>>>>
>>>>> "Charles C. Berry" <ccberry@ucsd.edu> writes:
>>>>>
>>>>>> On Mon, 6 Oct 2014, Rainer M Krug wrote:
>>>>>>
>>>>>>> Hi
>>>>>>>
>>>>>>> The variable transfer of tables from org to R caused sometimes 'could
>>>>>>> not find function "read.table"' errors (e.g. when the file was tangled
>>>>>>> into a ./data directory which was loaded by the function
>>>>>>> devtools::load_all("./")). This can easily be fixed by adding the package
>>>>>>> name to the call in R, i.e. replacing =read.table()= with
>>>>>>> =utils::read.table()= which is done in this patch.
>>>>>>
>>>>>> It does fix that one case.
>>>>>>
>>>>>> But I wonder if that is the best way.
>>>>>>
>>>>>> The heart of the matter is that load_all eventually calls sys.source,
>>>>>> which can be persnickety about finding objects on the search path. See
>>>>>> ?sys.source.
>>>>>>
>>>>>> If the src block you tangle to ./data/ has any code that uses any
>>>>>> other objects from utils, stats, datasets or whatever, you will be in
>>>>>> the same pickle.
>>>>>
>>>>> Exactly - that is true. But it is the same when putting this in a
>>>>> package (as far as I am aware).
>>>>>
>>>>
>>>> Do you mean that putting `x <- rnorm(10)' into a data/*.R file will
>>>> fail when you try to build and check?
>>>>
>>>> In fact, `R CMD build' will execute it and save the result as a
>>>> data/*.rda file. And check will go through.
>>>>
>>>> devtools::load_all (calling load_data) fails to do that. Which is why
>>>> I think this is a devtools issue.
>>>
>>> OK - point taken. But I still think that the =utils::read.table()= would
>>> not hurt, rather make the variable transfer safer.
>>>
>>
>> What you want to change is in a defconst. So, the user can override
>> with a file-local version.
>>
>> So, making the change really is harmless.
>>
>> Maybe add a note to the docstring to say that using
>> utils::read.table' assures that `read.table' always can be found just
>> in case anyone ever asks.
>
> OK - done in attached patch.
>
> Rainer
> From 1eadbf6b44b8fc2fa5af29ea483383e9d137d19e Mon Sep 17 00:00:00 2001
> From: "Rainer M. Krug" <R.M.Krug@gmail.com>
> Date: Mon, 6 Oct 2014 13:48:49 +0200
> Subject: [PATCH] ob-R.el: Add package name to read.table call
>
> * lisp/ob-R.el:
> (ob-R-transfer-variable-table-with-header): Add package name to call
> of R function read.table (now utils::read.table).  This clarifies the
> call as well as avoids 'could not find function' error in R under
> certain conditions.
> ---
>  lisp/ob-R.el | 9 ++++++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/lisp/ob-R.el b/lisp/ob-R.el
> index ea33031..a64b647 100644
> --- a/lisp/ob-R.el
> +++ b/lisp/ob-R.el
> @@ -101,7 +101,7 @@ this variable.")
>       con <- textConnection(
>         %S
>       )
> -     res <- read.table(
> +     res <- utils::read.table(
>         con,
>         header    = %s,
>         row.names = %s,
> @@ -112,14 +112,17 @@ this variable.")
>       res
>     })"
>    "R code used to transfer a table defined as a variable from org to R.
> -This function is used when the table contains a header.")
> +This function is used when the table contains a header. The usage
> +of utils::read.table() ensures that the command read.table() can
> +be found even in circumstances when the utils package is not in
> +the search path from R.")
>  
>  (defconst ob-R-transfer-variable-table-without-header
>    "%s <- local({
>       con <- textConnection(
>         %S
>       )
> -     res <- read.table(
> +     res <- utils::read.table(
>         con,
>         header    = %s,
>         row.names = %s,
> -- 
> 2.1.2
>
>
>>
>> Chuck
>>
>>

-- 
Rainer M. Krug
email: Rainer<at>krugs<dot>de
PGP: 0x0F52F982