[Phylobase-devl] Aug. virtual hackathon
Ben Bolker
bolker at ufl.edu
Thu Aug 13 19:37:00 CEST 2009
The problem with the vignette is in the following lines:
err1 <- try(g2 <- phylo4d(g1, geospiza_raw$data), silent=TRUE)
cat(as.character(err1))
where g1 is a phylo4 object containing the geospiza tree
and geospiza_raw$data is a data frame of tip data.
The problem is that the tip labels don't match:
> setdiff(labels(g1),rownames(geospiza_raw$data))
[1] "platyspiza" "pinaroloxias" "olivacea"
> setdiff(rownames(geospiza_raw$data),labels(g1))
[1] "Platyspiza" "Pinaroloxias"
(2 problems with capitalization and one missing taxon)
This used to be grounds for throwing an error: now
the Platyspiza and Pinaroloxias data are thrown away,
and platyspiza/pinaroloxias/olivacea end up with NA
tip data.
Therefore, since err1() is no longer of class "try-error",
R no longer knows how to coerce it to character.
If we hack this bit out, we get in trouble a little
bit later with
g2 <- phylo4d(g1, gdata,missing.tip.data="OK")
since there is no longer a missing.tip.data argument
to the constructor.
Basically, this indicates that the behavior of
the constructor when there are data/tree mismatches
has changed. We need to:
(1) figure out what has happened, decide whether to
switch back to the old way, and document accordingly;
(2) write unit tests that will test for appropriate
errors being thrown in various bad situations.
I haven't looked into this closely, but my initial
feeling is that I am strongly in favor of mismatches
provoking errors as the default, which can then be
overridden by the user as they desire and see fit.
Have committed these changes. R CMD check fm-branch
works with a couple of warnings about undocumented objects/
S4 methods.
cheers
Ben
François Michonneau wrote:
> Hi all -
>
> I just committed the last changes in my branch and a PDF document [1]
> that describes what I have been doing on my branch. I wrote it fairly
> quickly, so I hope there aren't too many mistakes.
>
> I know that there are still a couple of bugs here and there and a
> couple of functions that need more attention but you should be able to
> play with my modifications with not too many woes.
>
> I didn't update the documentation at all, so (unfortunately) you have
> to look at the code to understand what I changed. I didn't update the
> vignette either.
>
> It builds without the vignette (R CMD build fm-branch --no-vignettes).
> It doesn't check however, but I don't think it's related to my changes
> as it fails with the following error message:
> * checking PDF version of manual without index ... ERROR
> It looks that it comes from a typo in the vignette. I'll try to
> investigate the problem.
>
> Concerning the hackathlon, Aug 18th, 11am EDT works for me.
>
> Concerning what needs to be done during the hackathlon, I'd prefer to
> use the tracker this time instead of the wiki.
>
> Feel free to ask questions about the document or the changes I made in
> my branch.
>
> Cheers,
> -- François
>
> [1] you can find it at www/misc/fm-branch.pdf in the subversion tree
>
>
>
> On Thu, 2009-08-13 at 11:01 -0400, Ben Bolker wrote:
>> I would also prefer not a weekend (but not impossible either).
>>
>> Just to start the ball rolling, I propose Tuesday August 18
>> at 11 AM EDT (a little early for west-coast folks, but not
>> horrendous -- since Marguerite
>> can't play right now, we don't have to worry about Hawai'i)
>>
>> ??
>>
>> cheers
>> Ben
>>
>> Peter Cowan wrote:
>>> If I gather correctly, most folks are available after the 13th and
>>> before the 24th, except for Thibaut who is currently far away from a
>>> computer.
>>>
>>> What days in that period work best? Last time did we schedule it over
>>> a weekend, or did it just feel like that because I was traveling? I
>>> prefer it not be a weekend, but can do it if need be. Proposals for a
>>> specific starting day and duration welcome.
>>>
>>> Before, we actually start coding, it's probably worthwhile having at
>>> least some discussion about the changes that Francois has made, and
>>> any other major changes people would like. It would also be good to
>>> get the RUnit stuff checked in (if I can help at all with that let me
>>> know Jim) so that we can make a good dent in writing unit tests (which
>>> will make any future changes much easier to carry out).
>>>
>>> Cheers
>>>
>>> Peter
>>>
>>> On Aug 3, 2009, at 4:48 PM, Marguerite Butler wrote:
>>>
>>>> Hi Folks,
>>>>
>>>> This does sound great. Some of the students I've worked with would
>>>> really appreciate a more stable package that was also easier to
>>>> install.
>>>>
>>>> But I won't be able to participate this time. Need to work on my
>>>> tenure packet (ugg) and upcoming classes.
>>>>
>>>> Marguerite
>>>>
>>>> On Aug 3, 2009, at 11:11 AM, Jim Regetz wrote:
>>>>
>>>>> This sounds great - I'm definitely interested. I don't expect to have
>>>>> any major schedule conflicts in the next month, so this is a good
>>>>> time.
>>>>>
>>>>> I wasn't able to commit the RUnit tests before embarking on a busy
>>>>> July
>>>>> of traveling, so that's still on my plate. I also have some notes
>>>>> on a
>>>>> few unresolved issues I've encountered while working with phylo4d
>>>>> objects, but I'll have to dust them off and remind myself of the
>>>>> details
>>>>> before saying anything useful about that.
>>>>>
>>>>> Cheers,
>>>>> Jim
>>>>>
>>>>> Peter Cowan wrote:
>>>>>> phylobase devs,
>>>>>>
>>>>>> Last December, we had a very successful virtual hackathon. However,
>>>>>> development has stalled since then. Would folks be interested/
>>>>>> willing
>>>>>> to participate in a hackathon in the next month, focused on
>>>>>> preparing
>>>>>> the package for a release to CRAN? The school year is looming so I
>>>>>> suspect for many the sooner the better, though ESA is this week,
>>>>>> which
>>>>>> will keep at least Steve occupied.
>>>>>>
>>>>>> * If you are able, what are the best days for you?
>>>>>>
>>>>>> * What needs to be done before we can submit the package to CRAN?
>>>>>> How
>>>>>> long do you expect they will take?
>>>>>>
>>>>>>
>>>>>> Personally I'm free anytime. I need to finish up some changes I've
>>>>>> started with the plotting functions, I suspect it will take me a day
>>>>>> or two wrap them up.
>>>>>>
>>>>>> Peter
>>>>>> _______________________________________________
>>>>>> Phylobase-devl mailing list
>>>>>> Phylobase-devl at lists.r-forge.r-project.org
>>>>>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/phylobase-devl
>>>>>>
>>>>> _______________________________________________
>>>>> Phylobase-devl mailing list
>>>>> Phylobase-devl at lists.r-forge.r-project.org
>>>>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/phylobase-devl
>>>> ____________________________________________
>>>> Marguerite A. Butler
>>>> Assistant Professor
>>>> Department of Zoology
>>>> University of Hawaii
>>>> 2538 McCarthy Mall, Snyder 405
>>>> Honolulu, HI 96822
>>>>
>>>> Dept: 808-956-8617
>>>> Lab: 808-956-5867
>>>> FAX: 808-956-9812
>>>> http://www.hawaii.edu/zoology/faculty/butler.html
>>>> http://www2.hawaii.edu/~mbutler
>>>> http://www.hawaii.edu/zoology/
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Phylobase-devl mailing list
>>>> Phylobase-devl at lists.r-forge.r-project.org
>>>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/phylobase-devl
>>> _______________________________________________
>>> Phylobase-devl mailing list
>>> Phylobase-devl at lists.r-forge.r-project.org
>>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/phylobase-devl
>>
>> _______________________________________________
>> Phylobase-devl mailing list
>> Phylobase-devl at lists.r-forge.r-project.org
>> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/phylobase-devl
--
Ben Bolker
Associate professor, Biology Dep't, Univ. of Florida
bolker at ufl.edu / www.zoology.ufl.edu/bolker
GPG key: www.zoology.ufl.edu/bolker/benbolker-publickey.asc
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 261 bytes
Desc: OpenPGP digital signature
Url : http://lists.r-forge.r-project.org/pipermail/phylobase-devl/attachments/20090813/07052513/attachment-0001.pgp
More information about the Phylobase-devl
mailing list