I have an R package Bond Lab which actually supports a book Investing in MBS 
using R and Open Source Computing.  The Bond Lab beta is stable.  I have also 
created a package companion to investing in MBS.  Both are on my Github site

https://github.com/glennmschultz/

I wrote the companion as functions for each chapter calling source code.  I 
really don't like the way it works presentation wise but the user could see the 
code used.  Further, the user must create a directory and copy the source into 
the directory.  Further the source runs all examples at once which seems a 
little confusion.  However, to illustrate a function for chapter example I 
think I have to make a function to call a function which does not make sense.

My first question, any suggestions on illustrating function call aside from the 
strategy I am using.  I am not a fan of what I am doing.

My second question regards on.load() vs. on.attach() Bond Lab needs to create 
the BondLab directory which installs in the BondLab library to the ~/users 
directory, the same for the companion.  

I think I should use on.load(), correct?
I am not sure how to do this and will the same function work for both Windows 
and MAC?  
Do I need different copy from path for Windows and MAC or can it be generic 
since the copy happens at load.
I would like to get Bond Lab on CRAN.  I read CRAN rules require a yes/no from 
the user to write a directory.  So, I need to write an on.load() function for 
both Bond Lab and the Companion.  Also, CRAN requires that a package works on 
at least two systems so I need to figure out the Windows build.  This leads to 
another question.

Can I somehow make a windows binary on a MAC?
If not, can I emulate Windows on a MAC and make a binary?  Seems like this may 
not be 100%.
Will I actually have to buy a windows machine to build the windows binary for 
CRAN?

Glenn  
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to