It will be a little while before it's available, but I bit the bullet and submitted RVector into CPAN. Even though it's minus a few things I wanted to add before 0.1 went publically distributed, I figured it best to go ahead make it available and get started working toward 0.2. I've already started making changes to the master branch of my Git repo, including trying to finish adding entry naming support. That part should be fixed up soon, and I've already started working on the first RVector feature that is a statistical function that I'm not sure is available in R, but I'll add more about that later. As far as CPAN is concerned, I put in the request to register the Statistics::RVector namespace, but that may not be resolved until the first of the week. Once it is, RVector should be widely available to everyone via CPAN.
RVector aka Statistics::RVector soft-release
Well, I finally have enough stuff in place to at least soft-release RVector. As I mentioned in my previous post, to fall inline with Perl module naming guidelines, RVector alone wasn't going to cut it so I had to decide on a CPAN-appropriate module name. The one I decided was best, and is now the official name of RVector is Statistics::RVector, though I'll often just call it RVector since that's what it's been since conception. While I haven't pushed Statistics::RVector into CPAN yet, it is now available with download/install instructions on the RVector page. Once I've completed a few sets of module tests for the installer process and looked over the code and doc a little more closely, I'll get it pushed into CPAN, making for the official release of RVector 0.1. Provided all goes well and I don't lose power for too long in the upcoming Snow-maggeddon, I anticipate making this happen by weeks end. Go scoop it up and let me know how you fare. Should you identify any issues, please report them on the Github project issue tracker.
Prepping RVector for CPAN
As I've been looking over getting testing and distribution stuff taken care of for RVector, I've been reading through Perldoc's perlnewmod tutorial as this is the first time I'll be distributing a module that I've built. In reading through there, it's given me some things to think about. One thing that I've come to realize is that having this module keep the namespace RVector is probably not within the realm of acceptable naming conventions for Perl modules in CPAN.
Perl modules are expected to follow a heirarchy for module naming, and putting my module in the root of that heirarchy is pretty presumptuous. As I've looked into the proper heirarchy, it seems the best possible names for RVector would be Math::Vector::RStats or Statistics::RVector. I'm leaning toward the latter, but would welcome more suggestions. It won't matter too much because of the availability of rv() as an equilvalent for $class->new().
Also, I've found that there is an existing module that offers an extreme subset of RVector's vector math capabilities. Math::Vector::Real offers vector math and functions similarly as far as vector creation, but doesn't offer entry naming, matrices, statistical functions and so forth, some of which are already done in RVector, and some of which are still on my to do list for upcoming releases after 0.1.
In the end, RVector is coming soon, and I'll still be calling it RVector in all my posts and such, it'll just have an officialized Perl module name. But, I need to hurry up and pick the official module name so that I can get all of the testing code and Makefile.pl and such built. If I can make up my mind, or if you have any ideas, I'll announce the official name and an 0.1 release soon. Thanks for keeping an eye out.
One Step Closer to RVector 0.1
It's taken a little longer than I would like, but I'm finally getting recovered from losing my colo box, and creeping slowly closer to have an installable and well documented version of RVector - the statistical/mathematical vector module for Perl which mimics the function of R Stats. At this point, I have a module that is fully capable of the base mathematical functions across the whole of the vector, as well as things like minimums, maximums, averages, variance, and standard deviation. I have Perldoc written for all of this as well and am working to get a good hosted version of it.
The only thing remaining to be done is to build a proper Makefile.pl and so forth so that you can easily install RVector into your Perl libraries without having to do manual copying. I hope to knock this out soon and offer up an "official" RVector v0.1 for download. If you're feeling particularly ready to play, go for it. You can go ahead and grab a copy of RVector.pm file from Github and toss it in your perl libs. Good luck, and good stats!
On the Road to Recovery
As I was just getting started getting the new code repo website rolling, my colo box failed, as I noted on my blog. I was just about to announce the official release of RVector for Perl, but things happen. So here's the first new post on my new code site that I'm working to relaunch in the new year. RVector will be official soon. Until then, you can still have a look at it currently on Github. I'll have more to post soon.
