[Rd] generic '[' for a non-exported class

2010-04-07 Thread Christophe Genolini

Hi all,
I define a S4 class 'foo'. I define '[' and '[-' for it.
I do not want to export foo, so I do not put it in NAMESPACE.
I do not want to export '[' and '[-' either (since the user can not use 
foo, no raison to give him access to '[' for foo).


But R CMD check does not agree with me and report an error:
Undocumented S4 methods:
 generic '[' and siglist 'foo'
 generic '[-' and siglist 'foo'


Any solution ?
Christophe

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


[Rd] Additions to posting guide

2010-04-07 Thread Gabor Grothendieck
One user said he would have found it useful to have textConnection be
described in the posting guide since he was having problems creating a
self-contained post.

I suggest the following section be added:


How to post commented, minimal, self-contained, reproducible code
===

The footer of every message to r-help reminds the poster that postings
should include commented, minimal, self-contained, reproducible code.

That means that

- responders can copy the code to the clipboard without further modification
  and paste it into their session to reproduce it.

- the data must be included and

- the data should be cut down to the smallest amount that will illustrate
  the problem.


To include the data:

- use dput(mydata).  That will output R source code whose value is the
  indicated data and you can then write:

mydata -

  where you place the output of dput to the right of the arrow.

- use textConnection.  An alternative in the case of data frames is to
  read in a text string like this:

  Lines - A,B,C
  1,2,x
  3,4,y
  DF - read.table(textConnection(Lines), header = TRUE)

  Using textConnection(Lines) in place of a file name causes the input to
  be taken from the indicated string and facilitates self-contained posts.

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] Additions to posting guide

2010-04-07 Thread Gabor Grothendieck
e-letter, who had suggested that something about textConnection be
added, also suggested that warning against posting code that first
wipes out all variables in the workspace.

On Wed, Apr 7, 2010 at 7:31 AM, Gabor Grothendieck
ggrothendi...@gmail.com wrote:
 One user said he would have found it useful to have textConnection be
 described in the posting guide since he was having problems creating a
 self-contained post.

 I suggest the following section be added:


 How to post commented, minimal, self-contained, reproducible code
 ===

 The footer of every message to r-help reminds the poster that postings
 should include commented, minimal, self-contained, reproducible code.

 That means that

 - responders can copy the code to the clipboard without further modification
  and paste it into their session to reproduce it.

 - the data must be included and

 - the data should be cut down to the smallest amount that will illustrate
  the problem.


 To include the data:

 - use dput(mydata).  That will output R source code whose value is the
  indicated data and you can then write:

        mydata -

  where you place the output of dput to the right of the arrow.

 - use textConnection.  An alternative in the case of data frames is to
  read in a text string like this:

  Lines - A,B,C
  1,2,x
  3,4,y
  DF - read.table(textConnection(Lines), header = TRUE)

  Using textConnection(Lines) in place of a file name causes the input to
  be taken from the indicated string and facilitates self-contained posts.


__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] generic '[' for a non-exported class

2010-04-07 Thread Seth Falcon

On 4/7/10 1:09 AM, Christophe Genolini wrote:

Hi all,
I define a S4 class 'foo'. I define '[' and '[-' for it.
I do not want to export foo, so I do not put it in NAMESPACE.
I do not want to export '[' and '[-' either (since the user can not use
foo, no raison to give him access to '[' for foo).

But R CMD check does not agree with me and report an error:
Undocumented S4 methods:
generic '[' and siglist 'foo'
generic '[-' and siglist 'foo'


Any solution ?


You can document these on an internal API Rd page.  Create an Rd file 
like yourPkg-internal-api.Rd and add the appropriate \alias{} lines to it.


+ seth

--
Seth Falcon | @sfalcon | http://userprimary.net/

__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel


Re: [Rd] generic '[' for a non-exported class

2010-04-07 Thread Henrik Bengtsson
On Wed, Apr 7, 2010 at 6:13 PM, Seth Falcon s...@userprimary.net wrote:
 On 4/7/10 1:09 AM, Christophe Genolini wrote:

 Hi all,
 I define a S4 class 'foo'. I define '[' and '[-' for it.
 I do not want to export foo, so I do not put it in NAMESPACE.
 I do not want to export '[' and '[-' either (since the user can not use
 foo, no raison to give him access to '[' for foo).

 But R CMD check does not agree with me and report an error:
 Undocumented S4 methods:
 generic '[' and siglist 'foo'
 generic '[-' and siglist 'foo'


 Any solution ?

 You can document these on an internal API Rd page.  Create an Rd file like
 yourPkg-internal-api.Rd and add the appropriate \alias{} lines to it.

...with Rd \keyword{internal} on that Rd page (so it does not show up
on the Rd help index page).

/Henrik


 + seth

 --
 Seth Falcon | @sfalcon | http://userprimary.net/

 __
 R-devel@r-project.org mailing list
 https://stat.ethz.ch/mailman/listinfo/r-devel


__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel