[PATCH] Write org variables into own R environment instead of .GlobalEnv

[PATCH] Write org variables into own R environment instead of .GlobalEnv

[Rainer M Krug]

[-- Attachment #1: Type: text/plain, Size: 4119 bytes --]


* (org-babel-R-assign-elisp): replaced simple assignments of variables
to variables in R with assignments into an environment called 'org'
and lock these so that they can not be modified anymore.

* (org-babel-expand-body:R): wrap the call
  of `org-babel-variable-assignments:R' into
  - detachment of org environment
  - creation of new environment org to delete the old environment
  - locking of the org environment to prevent adding of bindings
  - appending of environment to search path of R

These patch implements the writing of org variables into a separate R
environment and attaches it to the search path. This solves several
not so nice issues:

1) The org variables are now grouped and can be seen via 'ls(org)' in
R
2) As the environment org and all bindings are locked, the variables
can not be accidentally deleted. They can be overwritten, but they can
be =restored by simply deleting the variable in R or by using
`org$VARIABLE' instead of `VARIABLE'
3) All variables can be saved by simply calling `save(org, FILENAME'
in R which makes it possible to store all variable definitions for
tangling in one file.
---
 lisp/ob-R.el | 47 +++++++++++++++++++++++++++++++----------------
 1 file changed, 31 insertions(+), 16 deletions(-)

diff --git a/lisp/ob-R.el b/lisp/ob-R.el
index 62aa7f2..da2aea2 100644
--- a/lisp/ob-R.el
+++ b/lisp/ob-R.el
@@ -103,8 +103,12 @@ this variable.")
             (append
              (when (cdr (assoc :prologue params))
                (list (cdr (assoc :prologue params))))
-             (org-babel-variable-assignments:R params)
-             (list body)
+	     '("try(detach(org), silent=TRUE)")
+             '("org <- new.env()")
+	     (org-babel-variable-assignments:R params)
+	     '("lockEnvironment(org)")
+	     '("attach(org)")
+	     (list body)
              (when (cdr (assoc :epilogue params))
                (list (cdr (assoc :epilogue params)))))))
        (if graphics-file
@@ -203,20 +207,31 @@ This function is called by `org-babel-execute-src-block'."
 			  "TRUE" "FALSE"))
 	      (row-names (if rownames-p "1" "NULL")))
 	  (if (= max min)
-	      (format "%s <- read.table(\"%s\",
-                      header=%s,
-                      row.names=%s,
-                      sep=\"\\t\",
-                      as.is=TRUE)" name file header row-names)
-	    (format "%s <- read.table(\"%s\",
-                   header=%s,
-                   row.names=%s,
-                   sep=\"\\t\",
-                   as.is=TRUE,
-                   fill=TRUE,
-                   col.names = paste(\"V\", seq_len(%d), sep =\"\"))"
-		    name file header row-names max))))
-    (format "%s <- %s" name (org-babel-R-quote-tsv-field value))))
+	      (format "assign(
+                         '%s', 
+                         read.table(\"%s\",
+                           header=%s,
+                           row.names=%s,
+                           sep=\"\\t\",
+                           as.is=TRUE
+                         ),
+                         envir = org
+                       )
+                       lockBinding('%s', org)" name file header row-names name)
+	    (format "assign(
+                       '%s', 
+                       read.table(\"%s\",
+                         header=%s,
+                         row.names=%s,
+                         sep=\"\\t\",
+                         as.is=TRUE,
+                         fill=TRUE,
+                         col.names = paste(\"V\", seq_len(%d), sep =\"\")
+                       ),
+                       envir = org
+                     )
+                     lockBinding('%s', org)" name file header row-names max name))))
+    (format "assign('%s', %s, envir = org); lockBinding('%s', org)" name (org-babel-R-quote-tsv-field value) name)))
 
 (defvar ess-ask-for-ess-directory) ; dynamically scoped
 (defun org-babel-R-initiate-session (session params)
-- 
1.8.5.2 (Apple Git-48)


-- 
Rainer M. Krug

email: RMKrug<at>gmail<dot>com

PGP: 0x0F52F982

[-- Attachment #2: Type: application/pgp-signature, Size: 494 bytes --]

Re: [PATCH] Write org variables into own R environment instead of .GlobalEnv

[Bastien]

Hi Rainer,

did you sign the FSF papers?  We need this before considering applying
the patch.  Also, the formatting of the Changelog entry is not correct,
please carefully read http://orgmode.org/worg/org-contribute.html

Thanks,

-- 
 Bastien

Re: [PATCH] Write org variables into own R environment instead of .GlobalEnv

[Rainer M Krug]

[-- Attachment #1: Type: text/plain, Size: 549 bytes --]

Bastien <bzg@gnu.org> writes:

> Hi Rainer,
>
> did you sign the FSF papers?  We need this before considering applying
> the patch.  

I mailed them on Tuesday, using the form
[[http://orgmode.org/request-assign-future.txt]] but did not get a
response.

> Also, the formatting of the Changelog entry is not correct,
> please carefully read http://orgmode.org/worg/org-contribute.html

OK - I will do so and re-send the patch.

Cheers,

Rainer

>
> Thanks,

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

[-- Attachment #2: Type: application/pgp-signature, Size: 494 bytes --]

Re: [PATCH] Write org variables into own R environment instead of .GlobalEnv

[Bastien]

Hi Rainer,

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

> I mailed them on Tuesday, using the form
> [[http://orgmode.org/request-assign-future.txt]] but did not get a
> response.

This usually take 4-6 weeks, so let's be patient.

>> Also, the formatting of the Changelog entry is not correct,
>> please carefully read http://orgmode.org/worg/org-contribute.html
>
> OK - I will do so and re-send the patch.

Thanks in advance!

-- 
 Bastien

Re: [PATCH] Write org variables into own R environment instead of .GlobalEnv

[Rainer M Krug]

[-- Attachment #1: Type: text/plain, Size: 714 bytes --]

Bastien <bzg@altern.org> writes:

> Hi Rainer,
>
> Rainer M Krug <Rainer@krugs.de> writes:
>
>> I mailed them on Tuesday, using the form
>> [[http://orgmode.org/request-assign-future.txt]] but did not get a
>> response.
>
> This usually take 4-6 weeks, so let's be patient.

OK - I'll let you know when I hear something.

>
>>> Also, the formatting of the Changelog entry is not correct,
>>> please carefully read http://orgmode.org/worg/org-contribute.html
>>
>> OK - I will do so and re-send the patch.
>
> Thanks in advance!

Patch is in a separate email - hope the commit message is better now.

Thanks,

Rainer

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

[-- Attachment #2: Type: application/pgp-signature, Size: 494 bytes --]

Re: [PATCH] Write org variables into own R environment instead of .GlobalEnv

[Bastien]

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

> OK - I'll let you know when I hear something.

We should both hear about it.

Please ping me if you don't hear anything in a month,
I'll then ask again to the FSF copyright clerk.

>>>> Also, the formatting of the Changelog entry is not correct,
>>>> please carefully read http://orgmode.org/worg/org-contribute.html
>>>
>>> OK - I will do so and re-send the patch.
>>
>> Thanks in advance!
>
> Patch is in a separate email - hope the commit message is better
> now.

Looks good, thanks!

-- 
 Bastien

Re: [PATCH] Write org variables into own R environment instead of .GlobalEnv

[Rainer M Krug]

[-- Attachment #1: Type: text/plain, Size: 724 bytes --]

Bastien <bzg@gnu.org> writes:

> Rainer M Krug <Rainer@krugs.de> writes:
>
>> OK - I'll let you know when I hear something.
>
> We should both hear about it.
>
> Please ping me if you don't hear anything in a month,
> I'll then ask again to the FSF copyright clerk.

I'll do - thanks.

>
>>>>> Also, the formatting of the Changelog entry is not correct,
>>>>> please carefully read http://orgmode.org/worg/org-contribute.html
>>>>
>>>> OK - I will do so and re-send the patch.
>>>
>>> Thanks in advance!
>>
>> Patch is in a separate email - hope the commit message is better
>> now.
>
> Looks good, thanks!

Thanks,

Rainer

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

[-- Attachment #2: Type: application/pgp-signature, Size: 494 bytes --]

[PATCH] Write org variables into own R environment

[Rainer M Krug]

[-- Attachment #1: Type: text/plain, Size: 3296 bytes --]


* lisp/ob-R.el (org-babel-expand-body:R): Create empty environment
  called `org' in R before adding variables and afterwards lock it and
  add it to the R search path.
  (org-babel-variable-assignments:R): Assign variables into own
  `org' environment in R instead of .GlobalEnv

These patch implements the writing of org variables into a separate R
environment and attaches it to the search path.  For the usage of these
variables, nothing changes in R, but:

1) The org variables are now grouped and can be seen via `ls(org)' in
R and are not shown anymore in the .GlobalEnv when using `ls()'
2) As the environment `org' and all bindings are locked, the variables
can not be accidentally deleted. They can be overwritten, but they can
be restored by simply deleting the variable in R or by using
`org$VARIABLE' instead of `VARIABLE'
3) All variables can be saved by simply calling `save(org, FILENAME'
in R which makes it possible to store all variable definitions for
tangling in one file.
---
 lisp/ob-R.el | 25 +++++++++----------------
 1 file changed, 9 insertions(+), 16 deletions(-)

diff --git a/lisp/ob-R.el b/lisp/ob-R.el
index 62aa7f2..82971de 100644
--- a/lisp/ob-R.el
+++ b/lisp/ob-R.el
@@ -103,8 +103,12 @@ this variable.")
             (append
              (when (cdr (assoc :prologue params))
                (list (cdr (assoc :prologue params))))
-             (org-babel-variable-assignments:R params)
-             (list body)
+	     '("try(detach(org), silent=TRUE)")
+             '("org <- new.env()")
+	     (org-babel-variable-assignments:R params)
+	     '("lockEnvironment(org)")
+	     '("attach(org)")
+	     (list body)
              (when (cdr (assoc :epilogue params))
                (list (cdr (assoc :epilogue params)))))))
        (if graphics-file
@@ -203,20 +207,9 @@ This function is called by `org-babel-execute-src-block'."
 			  "TRUE" "FALSE"))
 	      (row-names (if rownames-p "1" "NULL")))
 	  (if (= max min)
-	      (format "%s <- read.table(\"%s\",
-                      header=%s,
-                      row.names=%s,
-                      sep=\"\\t\",
-                      as.is=TRUE)" name file header row-names)
-	    (format "%s <- read.table(\"%s\",
-                   header=%s,
-                   row.names=%s,
-                   sep=\"\\t\",
-                   as.is=TRUE,
-                   fill=TRUE,
-                   col.names = paste(\"V\", seq_len(%d), sep =\"\"))"
-		    name file header row-names max))))
-    (format "%s <- %s" name (org-babel-R-quote-tsv-field value))))
+	      (format "assign( '%s', read.table(\"%s\", header=%s, row.names=%s, sep=\"\\t\", as.is=TRUE ), envir = org ); lockBinding('%s', org)" name file header row-names name)
+	    (format "assign( '%s', read.table(\"%s\", header=%s, row.names=%s, sep=\"\\t\", as.is=TRUE, fill=TRUE, col.names = paste(\"V\", seq_len(%d), sep =\"\") ), envir = org ); lockBinding('%s', org)" name file header row-names max name))))
+    (format "assign('%s', %s, envir = org); lockBinding('%s', org)" name (org-babel-R-quote-tsv-field value) name)))
 
 (defvar ess-ask-for-ess-directory) ; dynamically scoped
 (defun org-babel-R-initiate-session (session params)
-- 
1.8.5.2 (Apple Git-48)

[-- Attachment #2: Type: application/pgp-signature, Size: 494 bytes --]

Re: [PATCH] Write org variables into own R environment

[Aaron Ecay]

Hi Rainer,

This patch looks very promising.  (I have not had a chance to test the
implementation yet).  One issue might be the name of the “org” variable
– maybe this should instead be “.org” or “_org” to lessen the chance of
stomping on pre-existing variables named “org” in people’s code.

Thanks (also for your previous patch for handling in graphics blocks),

--
Aaron Ecay

Re: [PATCH] Write org variables into own R environment

[Rainer M Krug]

[-- Attachment #1: Type: text/plain, Size: 1482 bytes --]

Aaron Ecay <aaronecay@gmail.com> writes:

> Hi Rainer,
>
> This patch looks very promising.  (I have not had a chance to test the
> implementation yet).  

Thanks - please let me know as soon as you tested it.

> One issue might be the name of the “org” variable
> – maybe this should instead be “.org” or “_org” to lessen the chance of
> stomping on pre-existing variables named “org” in people’s code.

Agreed here. .org would be a tricky solution, because it is hidden -
advantage or disadvantage? Probably a disadvantage. I will rename it to 
"_org_variables_" which is likely to be unique.

Do you think it would be worth to have a check along the lines of 

if(class(_org_variables_)!="environment") {
  stop("_org_variables_ is already defined ")
} else {
  ...
}

or to make the name user definable?

A different thing which I am using at the moment is to save the
environment so that it can be easily distributed along the tangled code
*if all variables are defined at the beginning and not per code block*.

Another option I am thinking about is to provide the possibility to
store not only variables, but also R function - useful? Effectively a
kind of "R package light for org-mode"?

>
> Thanks (also for your previous patch for handling in graphics blocks),

Pleasure - Thanks that they are useful.

Rainer
>
> --
> Aaron Ecay
>
>

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

[-- Attachment #2: Type: application/pgp-signature, Size: 494 bytes --]