Skip Menu |

This queue is for tickets about the PDL-Stats CPAN distribution.

Report information
The Basics
Id: 54499
Status: resolved
Priority: 0/
Queue: PDL-Stats

People
Owner: MAGGIEXYZ [...] cpan.org
Requestors: djerius [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: 0.4.1
Fixed in: 0.4.2



Subject: PDL::Stats::Basic exports non-statistics functions
Hi, Thanks very much for your work on PDL::Stats. It's a great addition to PDL. Might I make a few suggestions about a few functions in PDL::Stats::Basic, namely get_data, which_id, and n_pair? These routines seem to be somewhat peripheral to the other functions in PDL:Stats::Basic, and I think they might be better placed in other modules. For example, get_data() is a data input routine which appears similar in nature to PDL::IO::Misc::rcols(). If its functionality can't be rolled into rcols(), then perhaps it should be moved into that package. Similarly, which_id is a generic indexing routine; it might be grouped with the other indexing routines in PDL::Primitive. n_pair() is a generic bad value routine; it would probably be better placed in a PDL::Bad::Misc package. Currently these routines are automatically exported, and because they are named somewhat generically, there is a good chance that they will conflict with existing code. For example, I often have a get_data() subroutine. Because PDL::Stats::Basic exports get_data by default, Perl complains about get_data being redefined. n_pair is perhaps better named "n_good_pairs". Might I suggest not exporting this routines? Thanks, Diab
Thank you for the feedback! Please watch for a fix regarding the automatic export in the next release. PDL-Stats is being developed independing of PDL, so there hasn't been much co-ordination regarding the functions. I have been meaning to talk more with PDL folks about some functions and PDL::GSL::CDF, but nerds are communication shy folks by default... Best, Maggie
Subject: Re: [rt.cpan.org #54499] PDL::Stats::Basic exports non-statistics functions
Date: Fri, 05 Mar 2010 11:29:31 -0500
To: bug-PDL-Stats [...] rt.cpan.org
From: Diab Jerius <dj [...] head.cfa.harvard.edu>
On Thu, 2010-02-11 at 06:34 -0500, Maggie J. Xiong via RT wrote: Show quoted text
> <URL: https://rt.cpan.org/Ticket/Display.html?id=54499 > > > Thank you for the feedback! Please watch for a fix regarding the > automatic export in the next release. PDL-Stats is being developed > independing of PDL, so there hasn't been much co-ordination regarding > the functions. I have been meaning to talk more with PDL folks about > some functions and PDL::GSL::CDF, but nerds are communication shy folks > by default... > > Best, > Maggie
Maggie, Sorry for the late reply! Coordination would be great; PDL suffers a bit from not having an overall structure where it's obvious to put things, so sometimes things go in the weirdest of places. I'm still trying to figure out the difference between PDL::Basic and PDL::Primitive... It can be quite hard to find things. Diab
Diab, I'm transitioning to using the new rcols()--PDL-2.4.6_004 or higher--for data IO. I added rtable() to PDL::Stats::Basic and put in a note about deprecating get_data(). At the moment I'm still letting PDL::PP auto exporting get_data(). The easiest way to solve your problem of conflicting get_data() is to "use PDL::Stats ();". Before the transition to rtable() is done this is probably the best compromise. I mentioned the functions which_id, n_pair on the PDL dev list. Apparently these functions are more specific to stats than pdl in general. They are therefore staying in PDL-Stats. FYI--the new release 0.4.2 will also include perldl online help support for PDL-Stats :) Best, Maggie On 2010-03-05 11:30:34, dj@head.cfa.harvard.edu wrote: Show quoted text
> On Thu, 2010-02-11 at 06:34 -0500, Maggie J. Xiong via RT wrote:
> > <URL: https://rt.cpan.org/Ticket/Display.html?id=54499 > > > > > Thank you for the feedback! Please watch for a fix regarding the > > automatic export in the next release. PDL-Stats is being developed > > independing of PDL, so there hasn't been much co-ordination regarding > > the functions. I have been meaning to talk more with PDL folks about > > some functions and PDL::GSL::CDF, but nerds are communication shy folks > > by default... > > > > Best, > > Maggie
> > Maggie, > > Sorry for the late reply! Coordination would be great; PDL suffers a > bit from not having an overall structure where it's obvious to put > things, so sometimes things go in the weirdest of places. I'm still > trying to figure out the difference between PDL::Basic and > PDL::Primitive... It can be quite hard to find things. > > Diab > >
Waiting for next major PDL release to use rcols() for data IO. That timing is beyond my control. Closing the ticket for now.