push to CRAN

[AndreaCirilloAC]

@RobertMyles as you previously suggested I think this useful package is now ready to be submitted to CRAN.
What do you think about it?

[RobertMyles]

I think it's good to go, Andrea! Maybe we could run some checks with the devtools package before we submit it, but other than that I think it's good to go :-)

[AndreaCirilloAC]

I run the devtools check on the package. everything went smoothly except for the following error:

* checking package dependencies ... ERROR
Namespace dependency not required: ‘magrittr’

See section ‘The DESCRIPTION file’ in the ‘Writing R Extensions’
manual.
* DONE
Status: 1 ERROR
checking package dependencies ... ERROR
Namespace dependency not required: ‘magrittr’

going to look into this.

[RobertMyles]

Ah, it must be the line "@importFrom magrittr '%>%' "; it appears to be unnecessary; my mistake!

[AndreaCirilloAC]

I was looking at that as well. Nevertheless I was wondering: if we remove the dependency from magrittr from when will the package import the pipe operator?

…

Il giorno 28 mar 2017, alle ore 14:38, Robert Myles McDonnell ***@***.***> ha scritto: Ah, it must be the line ***@***.*** magrittr '%>%' "; it appears to be unnecessary; my mistake! — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.

[RobertMyles]

Yeah, that's a strange one. We use the pipe, so I think it's better to explicitly import it, and I'm not sure why the check would throw an error like that. Looking at the code, I think we should probably also change the example to a "Not run" example. Perhaps we could try changing back to importing dplyr instead of magrittr and see if it throws this error?

[RobertMyles]

Hi Andrea, I just pushed a commit that should fix some of these problems. I didn't change the example, maybe you might like to do that, but it is pretty much ready to go to CRAN.

[AndreaCirilloAC]

thank you @RobertMyles , just submitted.
going to wait for their review for (hopefully) closing the issue.

[AndreaCirilloAC]

nearly done, but still not there, got the following two errors from pre-test:

• checking examples ...
  ** running examples for arch 'i386' ... ERROR
  Error: .Platform$OS.type == "unix" is not TRUE
  Execution halted
  ** running examples for arch 'x64' ... ERROR

full log at:https://win-builder.r-project.org/incoming_pretest/170404_051110_updateR_01/00check.log

this is due to platform limitation I guess. going to look how to overcome this, which I think is possible since packages like our inspiration InstallR from Tal Galili are for Windows only.

[AndreaCirilloAC]

added os_type on description with fad7ff3