[R-pkg-devel] Package Submission: issues with R_registerRoutines', 'R_useDynamicSymbols'

2017-04-13 Thread Vineetha Warriyar Kodalore Vijayan 

Hi,


When I submitted my R package to CRAN, it didn't go through the automatic 
incoming check, resulting the following NOTE.  I tested on my macOS with R CMD 
CHECK and R CMD --as-cran, both went through with one NOTE ( Maintainer: NOTE).

* checking compiled code ... NOTE
File 'EpiILM/libs/i386/EpiILM.dll':
  Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'
File 'EpiILM/libs/x64/EpiILM.dll':
  Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'

It is good practice to register native routines and to disable symbol
search.

See 'Writing portable packages' in the 'Writing R Extensions' manual.

I did go through the manual (section 5.4), but didn't understand much. Can 
someone help to resolve this issue?


Thanks,

Vineetha


[[alternative HTML version deleted]]

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

Re: [R-pkg-devel] Package Submission: issues with R_registerRoutines', 'R_useDynamicSymbols'

2017-04-13 Thread Ege Rubak 

Hi,

This is something that has gotten quite a bit of attention recently due 
to the coming R 3.4 giving a NOTE when routines aren't declared. You 
should be able to find info on this list or elsewhere.
Basically you need to make a file called e.g. init.c in your src/ with 
the appropriate code. This file can be auto-generated for you as 
described by Dirk here:

http://dirk.eddelbuettel.com/blog/2017/03/30/#002_even_easier_package_registration

Good luck!

Ege

On 04/13/2017 11:14 PM, Vineetha Warriyar Kodalore Vijayan wrote:


Hi,


When I submitted my R package to CRAN, it didn't go through the automatic 
incoming check, resulting the following NOTE.  I tested on my macOS with R CMD 
CHECK and R CMD --as-cran, both went through with one NOTE ( Maintainer: NOTE).

* checking compiled code ... NOTE
File 'EpiILM/libs/i386/EpiILM.dll':
  Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'
File 'EpiILM/libs/x64/EpiILM.dll':
  Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'

It is good practice to register native routines and to disable symbol
search.

See 'Writing portable packages' in the 'Writing R Extensions' manual.

I did go through the manual (section 5.4), but didn't understand much. Can 
someone help to resolve this issue?


Thanks,

Vineetha


[[alternative HTML version deleted]]

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



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

Re: [R-pkg-devel] Package Submission: issues with R_registerRoutines', 'R_useDynamicSymbols'

2017-04-13 Thread Vineetha Warriyar Kodalore Vijayan 
Hi Ege,


Thank you!

In the init.c file,  do I have to change 'void' to corresponding declaration?

E.g:

extern void F77_NAME(datacon)(void *, void *, void *, void *, void *, void *, 
void *, void *, void *, void *, void *);


Also, how do I know the package passes the CRAN incoming feasibility as it does 
not show any warnings or notes on my R CMD CHECK --as-cran. Re submit and see 
again?


Regards,

Vineetha Warriyar


From: R-package-devel  on behalf of Ege 
Rubak 
Sent: Thursday, April 13, 2017 9:35:36 AM
To: r-package-devel@r-project.org
Subject: Re: [R-pkg-devel] Package Submission: issues with R_registerRoutines', 
'R_useDynamicSymbols'

Hi,

This is something that has gotten quite a bit of attention recently due
to the coming R 3.4 giving a NOTE when routines aren't declared. You
should be able to find info on this list or elsewhere.
Basically you need to make a file called e.g. init.c in your src/ with
the appropriate code. This file can be auto-generated for you as
described by Dirk here:
http://dirk.eddelbuettel.com/blog/2017/03/30/#002_even_easier_package_registration

Good luck!

Ege

On 04/13/2017 11:14 PM, Vineetha Warriyar Kodalore Vijayan wrote:
>
> Hi,
>
>
> When I submitted my R package to CRAN, it didn't go through the automatic 
> incoming check, resulting the following NOTE.  I tested on my macOS with R 
> CMD CHECK and R CMD --as-cran, both went through with one NOTE ( Maintainer: 
> NOTE).
>
> * checking compiled code ... NOTE
> File 'EpiILM/libs/i386/EpiILM.dll':
>   Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'
> File 'EpiILM/libs/x64/EpiILM.dll':
>   Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'
>
> It is good practice to register native routines and to disable symbol
> search.
>
> See 'Writing portable packages' in the 'Writing R Extensions' manual.
>
> I did go through the manual (section 5.4), but didn't understand much. Can 
> someone help to resolve this issue?
>
>
> Thanks,
>
> Vineetha
>
>
>[[alternative HTML version deleted]]
>
> __
> R-package-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel
>

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

[[alternative HTML version deleted]]

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

Re: [R-pkg-devel] Package Submission: issues with R_registerRoutines', 'R_useDynamicSymbols'

2017-04-13 Thread Ege Rubak 

Hi,

A few extra comments are below.

On 04/14/2017 12:46 AM, Zhian Kamvar wrote:

Hi,


On Apr 13, 2017, at 11:41 , Vineetha Warriyar Kodalore Vijayan
mailto:vineethawarriyar@ucalgary.ca>> wrote:

Hi Ege,


Thank you!

In the init.c file,  do I have to change 'void' to corresponding
declaration?

E.g:

extern void F77_NAME(datacon)(void *, void *, void *, void *, void *,
void *, void *, void *, void *, void *, void *);

I don't think you need to change anything in the generated file. At 
least I didn't, and I believe Brian Ripley mentioned that the script 
generated fully working init files for hundreds or thousands of packages 
with compiled code and only needed manual changes for tens of packages 
when they tested it on CRAN.


Also, how do I know the package passes the CRAN incoming feasibility
as it does not show any warnings or notes on my R CMD CHECK --as-cran.
Re submit and see again?



You can use R-Hub as a way to check your package on the development
version: https://github.com/r-hub/rhub#readme If you have your package
on a public remote git repository, you can additionally set up a
continuous integration service like
travis-ci: https://docs.travis-ci.com/user/languages/r/ and use that to
check against devel.

I agree that R-Hub is a great solution, and I have used it several times 
myself. You can even just use the website to submit a package:

https://builder.r-hub.io/advanced
But the R interface is nice, so it is recommendable to install the rhub 
package.
Alternatively, submitting to http://win-builder.r-project.org/ also lets 
you test under R-devel.


Best,
Ege


Best,
Zhian



Regards,

Vineetha Warriyar


From: R-package-devel mailto:r-package-devel-boun...@r-project.org>> on behalf of Ege Rubak
mailto:ru...@math.aau.dk>>
Sent: Thursday, April 13, 2017 9:35:36 AM
To: r-package-devel@r-project.org 
Subject: Re: [R-pkg-devel] Package Submission: issues with
R_registerRoutines', 'R_useDynamicSymbols'

Hi,

This is something that has gotten quite a bit of attention recently due
to the coming R 3.4 giving a NOTE when routines aren't declared. You
should be able to find info on this list or elsewhere.
Basically you need to make a file called e.g. init.c in your src/ with
the appropriate code. This file can be auto-generated for you as
described by Dirk here:
http://dirk.eddelbuettel.com/blog/2017/03/30/#002_even_easier_package_registration

Good luck!

Ege

On 04/13/2017 11:14 PM, Vineetha Warriyar Kodalore Vijayan wrote:


Hi,


When I submitted my R package to CRAN, it didn't go through the
automatic incoming check, resulting the following NOTE.  I tested on
my macOS with R CMD CHECK and R CMD --as-cran, both went through with
one NOTE ( Maintainer: NOTE).

* checking compiled code ... NOTE
File 'EpiILM/libs/i386/EpiILM.dll':
 Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'
File 'EpiILM/libs/x64/EpiILM.dll':
 Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'

It is good practice to register native routines and to disable symbol
search.

See 'Writing portable packages' in the 'Writing R Extensions' manual.

I did go through the manual (section 5.4), but didn't understand
much. Can someone help to resolve this issue?


Thanks,

Vineetha


  [[alternative HTML version deleted]]

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



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

[[alternative HTML version deleted]]

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




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

Re: [R-pkg-devel] Package Submission: issues with R_registerRoutines', 'R_useDynamicSymbols'

2017-04-13 Thread Vineetha Warriyar Kodalore Vijayan 
Thanks everyone!


-Vineetha


From: Ege Rubak 
Sent: Thursday, April 13, 2017 11:04:20 AM
To: Zhian Kamvar; Vineetha Warriyar Kodalore Vijayan
Cc: r-package-devel@r-project.org
Subject: Re: [R-pkg-devel] Package Submission: issues with R_registerRoutines', 
'R_useDynamicSymbols'

Hi,

A few extra comments are below.

On 04/14/2017 12:46 AM, Zhian Kamvar wrote:
> Hi,
>
>> On Apr 13, 2017, at 11:41 , Vineetha Warriyar Kodalore Vijayan
>> > > wrote:
>>
>> Hi Ege,
>>
>>
>> Thank you!
>>
>> In the init.c file,  do I have to change 'void' to corresponding
>> declaration?
>>
>> E.g:
>>
>> extern void F77_NAME(datacon)(void *, void *, void *, void *, void *,
>> void *, void *, void *, void *, void *, void *);
>>
I don't think you need to change anything in the generated file. At
least I didn't, and I believe Brian Ripley mentioned that the script
generated fully working init files for hundreds or thousands of packages
with compiled code and only needed manual changes for tens of packages
when they tested it on CRAN.
>>
>> Also, how do I know the package passes the CRAN incoming feasibility
>> as it does not show any warnings or notes on my R CMD CHECK --as-cran.
>> Re submit and see again?
>>
>
> You can use R-Hub as a way to check your package on the development
> version: https://github.com/r-hub/rhub#readme If you have your package
> on a public remote git repository, you can additionally set up a
> continuous integration service like
> travis-ci: https://docs.travis-ci.com/user/languages/r/ and use that to
> check against devel.
>
I agree that R-Hub is a great solution, and I have used it several times
myself. You can even just use the website to submit a package:
https://builder.r-hub.io/advanced
But the R interface is nice, so it is recommendable to install the rhub
package.
Alternatively, submitting to http://win-builder.r-project.org/ also lets
you test under R-devel.

Best,
Ege

> Best,
> Zhian
>
>>
>> Regards,
>>
>> Vineetha Warriyar
>>
>> 
>> From: R-package-devel > > on behalf of Ege Rubak
>> mailto:ru...@math.aau.dk>>
>> Sent: Thursday, April 13, 2017 9:35:36 AM
>> To: r-package-devel@r-project.org 
>> Subject: Re: [R-pkg-devel] Package Submission: issues with
>> R_registerRoutines', 'R_useDynamicSymbols'
>>
>> Hi,
>>
>> This is something that has gotten quite a bit of attention recently due
>> to the coming R 3.4 giving a NOTE when routines aren't declared. You
>> should be able to find info on this list or elsewhere.
>> Basically you need to make a file called e.g. init.c in your src/ with
>> the appropriate code. This file can be auto-generated for you as
>> described by Dirk here:
>> http://dirk.eddelbuettel.com/blog/2017/03/30/#002_even_easier_package_registration
>>
>> Good luck!
>>
>> Ege
>>
>> On 04/13/2017 11:14 PM, Vineetha Warriyar Kodalore Vijayan wrote:
>>>
>>> Hi,
>>>
>>>
>>> When I submitted my R package to CRAN, it didn't go through the
>>> automatic incoming check, resulting the following NOTE.  I tested on
>>> my macOS with R CMD CHECK and R CMD --as-cran, both went through with
>>> one NOTE ( Maintainer: NOTE).
>>>
>>> * checking compiled code ... NOTE
>>> File 'EpiILM/libs/i386/EpiILM.dll':
>>>  Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'
>>> File 'EpiILM/libs/x64/EpiILM.dll':
>>>  Found no calls to: 'R_registerRoutines', 'R_useDynamicSymbols'
>>>
>>> It is good practice to register native routines and to disable symbol
>>> search.
>>>
>>> See 'Writing portable packages' in the 'Writing R Extensions' manual.
>>>
>>> I did go through the manual (section 5.4), but didn't understand
>>> much. Can someone help to resolve this issue?
>>>
>>>
>>> Thanks,
>>>
>>> Vineetha
>>>
>>>
>>>   [[alternative HTML version deleted]]
>>>
>>> __
>>> R-package-devel@r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-package-devel
>>>
>>
>> __
>> R-package-devel@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-package-devel
>>
>> [[alternative HTML version deleted]]
>>
>> __
>> R-package-devel@r-project.org mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-package-devel
>

[[alternative HTML version deleted]]

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

Re: [R-pkg-devel] Package Submission: issues with R_registerRoutines', 'R_useDynamicSymbols'

2017-04-13 Thread Tim Keitt 
On Thu, Apr 13, 2017 at 10:14 AM, Vineetha Warriyar Kodalore Vijayan <
vineethawarriyar@ucalgary.ca> wrote:

> Can someone help to resolve this issue?


I wrote a short R script that will read your RcppExports.cpp and emit the
corresponding init.c file. I call it from my configure script, but if your
code does not change, then you can just call it once (in your package
directory) and be done with it.

The R script is here:
https://github.com/thk686/rpg/blob/master/src/gen_init_c.R

THK

http://www.keittlab.org/

[[alternative HTML version deleted]]

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

[R-pkg-devel] Warning of DDRTree

2017-04-13 Thread Xiaojie Qiu 
We are planning to update the DDRTree package but found the following
warnings:

  
d:/RCompile/CRANpkg/lib/3.4/BH/include/boost/unordered/detail/buckets.hpp:586:29:
warning: ISO C++ 1998 does not support 'long long' [-Wlong-long]
  
d:/RCompile/CRANpkg/lib/3.4/BH/include/boost/unordered/detail/buckets.hpp:591:38:
warning: ISO C++ 1998 does not support 'long long' [-Wlong-long]

 Those warnings seem to relate to boost but not to our package. Can anyone
have some idea on how to fix this?
The check log is here:
https://win-builder.r-project.org/incoming_pretest/170414_025235_DDRTree_015/00check.log
The windows install log is here:
https://win-builder.r-project.org/incoming_pretest/170414_025235_DDRTree_015/00install.out

Thanks in advance!

Best,
Xiaojie

[[alternative HTML version deleted]]

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

Re: [R-pkg-devel] Warning of DDRTree

2017-04-13 Thread Kevin Ushey 
You can try placing this in your src/Makevars[.win] file:

CXXFLAGS = -DBOOST_NO_LONG_LONG

The dplyr package does this (and more) to avoid 'long long' leaking in
from Boost headers. See e.g.

https://github.com/tidyverse/dplyr/blob/ff719f53f88f2f03a298fa0ea489b902527ee2f1/src/Makevars#L2

The other alternative would be to just enforce the use of C++11 in
your package, by adding e.g.

SystemRequirements: C++11

to your DESCRIPTION file, or

CXX_STD = CXX11

to your src/Makevars[.win]. See
https://cran.r-project.org/doc/manuals/r-release/R-exts.html#Using-C_002b_002b11-code
for more information.

Best,
Kevin

On Thu, Apr 13, 2017 at 6:49 PM, Xiaojie Qiu  wrote:
>
> We are planning to update the DDRTree package but found the following
> warnings:
>
>   
> d:/RCompile/CRANpkg/lib/3.4/BH/include/boost/unordered/detail/buckets.hpp:586:29:
> warning: ISO C++ 1998 does not support 'long long' [-Wlong-long]
>   
> d:/RCompile/CRANpkg/lib/3.4/BH/include/boost/unordered/detail/buckets.hpp:591:38:
> warning: ISO C++ 1998 does not support 'long long' [-Wlong-long]
>
>  Those warnings seem to relate to boost but not to our package. Can anyone
> have some idea on how to fix this?
> The check log is here:
> https://win-builder.r-project.org/incoming_pretest/170414_025235_DDRTree_015/00check.log
> The windows install log is here:
> https://win-builder.r-project.org/incoming_pretest/170414_025235_DDRTree_015/00install.out
>
> Thanks in advance!
>
> Best,
> Xiaojie
>
> [[alternative HTML version deleted]]
>
> __
> R-package-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-package-devel

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

Re: [R-pkg-devel] Warning of DDRTree

2017-04-13 Thread Dirk Eddelbuettel 

On 14 April 2017 at 01:49, Xiaojie Qiu wrote:
| We are planning to update the DDRTree package but found the following
| warnings:
| 
|   
d:/RCompile/CRANpkg/lib/3.4/BH/include/boost/unordered/detail/buckets.hpp:586:29:
| warning: ISO C++ 1998 does not support 'long long' [-Wlong-long]
|   
d:/RCompile/CRANpkg/lib/3.4/BH/include/boost/unordered/detail/buckets.hpp:591:38:
| warning: ISO C++ 1998 does not support 'long long' [-Wlong-long]
| 
|  Those warnings seem to relate to boost but not to our package. Can anyone
| have some idea on how to fix this?

Yes. Compile the program as a C++11 program.

My favourite way of saying that is via

   CXX_STD = CXX11

in src/Makevars. Now that all core systems (Win, OS X, Linux) used with R
have proper C++11 compiler support this is no longer an issue.

[ The underlying issue is boring. The powers that be decided a long time ago
that only C++98 was real C++ in the context of R. And that old standard has
no 'long long', hence the issue. Ironically, intermediate standards added
long long but we were not aloud to play with it because someone else knew
what's best for us. Luckily this is a thing of the past. ]

Dirk

-- 
http://dirk.eddelbuettel.com | @eddelbuettel | e...@debian.org

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