Search Results: "Dirk Eddelbuettel"

29 October 2013

Dirk Eddelbuettel: Rcpp 0.10.6

A new maintenance release 0.10.6 of Rcpp is now on the CRAN network for GNU R; binaries for Debian have been uploaded as well. This version ties up a number of smaller loose ends, but also adds a few new things, particularly John's new exposeClass. Details are as usual in the NEWS extract below, and also in the ChangeLog file in the package and on the Rcpp Changelog page.
Changes in Rcpp version 0.10.6 (2013-10-27)
  • Changes in Rcpp API:
    • The function exposeClass takes a description of the constructors, fields and methods to be exposed from a C++ class, and writes C++ and R files in the package. Inherited classes can be dealt with, but require data type information. This approach avoids hand-coding module files.
    • Two missing is<>() templates for CharacterVector and CharacterMatrix have been added, and some tests for is_na() and is_finite() have been corrected thanks to Thomas Tse.
  • Changes in R code:
    • Export linking helper function LdFlags as well as RcppLdFlags.
    • Function Rcpp.package.skeleton() no longer passes a namespace argument on to package.skeleton()
  • Changes in R setup:
    • Raise requirement for R itself to be version 3.0.0 or later as needed by the vignette processing
  • Changes in Rcpp attributes:
    • sourceCpp now correctly binds to Rtools 3.0 and 3.1
Thanks to CRANberries, you can also look at a diff to the previous release 0.10.5. As always, even fuller details are on the Rcpp Changelog page and the Rcpp page which also leads to the downloads, the browseable doxygen docs and zip files of doxygen output for the standard formats. A local directory has source and documentation too. Questions, comments etc should go to the rcpp-devel mailing list off the R-Forge page

This post by Dirk Eddelbuettel originated on his Thinking inside the box blog. Please report excessive re-aggregation in third-party for-profit settings.

28 October 2013

Dirk Eddelbuettel: Kurt Elling at Dominican University

Some catchup blogging: Kurt Elling was back in town on October 19, and not just in Chicago but quite literally in our little 'burb as the headliner for this season's concerts at Domiminican University. They had done us the favour of inviting Dianne Reeves back in March 2009, but this was clearly going to be a real treat. And I managed to snatch four front row (!!) tickets as soon as I heard about it, and the show once again didn't disappoint. Elling was as usual accompanied by his long-time collaborator Laurence Hobgood on piano, as well as local heros Clark Sommers on bass and John McLean on guitar. This time, Quincy Davies was on drums. And this band is realiably excellent, as is Elling in any live setting. The set was dominated by pieces from his most recent record and augmented by a few other standout pieces. There is a pretty rich set of live music by Elling on YouTube, see for example On Broadway (which was part of the set in a wonderfully fast and funked-up variant), Golden Lady (from the previous album, and also played live this time) or for example this wonderful version of You Send Me (also part of the concert). I snatched a photo or two (while at least turning my flash off, unlike the weird gal to my left, but I digress) and posted one on Google+. Kurt Elling will be back in Chicago at the Green Mill in early January. Expect to me there, and I hope to see you too!

This post by Dirk Eddelbuettel originated on his Thinking inside the box blog. Please report excessive re-aggregation in third-party for-profit settings.

24 October 2013

Dirk Eddelbuettel: Introducing the CRAN Repository Policy Watch

CRAN is the repository network for R. It is a resounding success with (as of right now) almost 5000 packages, and growth rate which has been estimated (cf John Fox's keynote at useR! a few years ago) to be near 40% per year. We as R community members owe a great deal of thanks to the CRAN maintainers, and the R Core team. The tight integration between the language and the contributed code repository is unique among programming languages, and one of the many reasons why CRAN has been such a success and driver of growth and adoption for R. And the amount of work the CRAN maintainers put into this is substantial, and we owe them. Yet there is some friction between the repo maintainers, and the community of developers. There have numerous discussions on the main developer list about various aspect of how CRAN maintains the impeccable quality of the code in the archive. One particular aspect which has bugging (at least me) is the lack of communication when policy changes are made. It would be trivial to posts a set of changes to the developer list, and I suggested as much. Only to be soundly ignored. But changes to text (or html) files can monitored rather easily, and when such changes occur an alert can be sent. So I cooked up rather simple system for this which I called the CRAN Policy Watch. In essence, a simple cronjob monitors changes, and records new versions in a Github repo (which you can follow or star). Alternatively, the cronjob now also tweets from the @CRANPolicyWatch account which I invite everyone to subscribe to as well. If someone knows of a simple tool to summarize diffs of html or text files in static html pages, I'd be interested to expand the service to some github.io pages. Alernatively I could also just commit to a single file too and let Github summarize the changes. This was an itch I needed to scratch, and I hope some other people will fine this useful too.

19 October 2013

Dirk Eddelbuettel: New BH release 1.51.1-3

A new release of the BH package is now on CRAN and its mirrors. BH provides (a subset of) the Boost library for C++, particularly the (large) parts delivered as pure template headers not requiring linking. When this package is installed, an R package developer can deploy it at build-time via a simple LinkingTo: BH declaration (which, despite the title affects only compilation, not linking; don't ask...). This frees developers from having to include the (sizeable) Boost headers in their packages, and provides (parts of) Boost as part of the R build system. A short example of using this BH package with Rcpp is provided in this Rcpp Gallery post; a number of other Boost-related posts are also available. This release expands the scope of the package by a quite bit as shown in the NEWS entry:
Changes in version 1.51.0-3 (2013-10-19)
The other change is that I am now acting as maintainer taking over from Jay who has headed the initial creation and first releases after he, Mike and I had talked about this for way too long without actually doing anything about it. Thanks for getting everything rolling, Jay! Comments and suggestions are welcome via the mailing list or issue tracker available via the package page at R-Forge.

Dirk Eddelbuettel: Yes another Herbie Hancock concert

Hadn't mentioned yet that I saw Herbie Hancock last Friday at the CSO / Symphony Center. That's about the fourth time I have him according to this blog (September 2007,March 2005, April 2004) and I think I saw him two more times when I lived in France, and maybe once while I was in Canada. This time, the format was a horn-free quintet with Lionel Loueke on guitar, James Genus on bass, Vinnie Colaiuta on drums and Zakir Hussain on tabla. Hancock himself played mostly keyboards / synthesizer, even when he used the full piano as input. Overall the concert was little uneven--Hancock himself alluded to the fact that the five hadn't really practised together. At times they were rolling: the opening was a very rhythmic version of his very old standard Watermelon Man; the encore was a very rocking version of his best-selling pop-fusion hit Rockit. In between, it was sometimes wanting. Still, all five are tremendous artists and I also had a fabulous seat as shown in this Google+ post with a picture of Hancock's setup, including multiple screens. But they did not really connect with audience as a whole, and by the time the encore rolled around, the house was pretty empty.

13 October 2013

Dirk Eddelbuettel: Chicago Marathon 2013

A gorgeous day in Chicago for the 37th annual Chicago Marathon. Today was the seventh time I ran this race. I had sat out last fall, and ran a smaller local race (well enough for a BQ). But the last time I ran Chicago I mused that maybe next time I'd train more. I tried, but it didn't quite work with a minor injury dragging on a little. Just like in 2010. And so it continues: it is my home course, and the course I have run the most and know the best. Yet it is the one where I have the greatest difficulty coming in with a time that is good enough for Boston. I knew that today wasn't going to be that day, so I aimed lower to just run even splits (ie running about equal halfs). But even that failed: After 1:44 for the first half I lost a good ten minutes to finish at 3:40:42. Oh well. There is always next year... But it was a simply sunningly fabulous day, and the race was once again very, very well put together. We were worried about extra security: not a real issue. The streets were lined with people and even Lisa and Julia managed to cheer me on at mile 13 (still smiles), mile 19 (grimacing, about to fall apart) and 25 (hanging in). And Anna was working the mile 18 water stop. The worst where miles 20 to 22 when I walked a fair bit. The race results have details; maybe I'll update the chart I made in prior years. One minor FAIL was a wait at the end for gear check. That was pathetic. I guess the folks manning the stand didn't keep bags sorted well enough so lines were long, and slow. Oh well -- the rest was fine, and it remains one of the nicest marathon races. Oh, and the other FAIL was that my Nexus 4 phone couldn't hold a charge til I was done. What's up with that? The running tracker traced me til around mile 20, and then the phone shut down for lack of juice. Disappointing too. But hey, it spares you, dear reader, the look of some bad post-race selfies...

30 September 2013

Dirk Eddelbuettel: RcppArmadillo 0.3.920.1

Along with the Rcpp 0.10.5 release yesterday, a new minor release 0.3.920.1 of RcppArmadillo came out. It is based on Conrad's Armadillo 3.920.0 plus a minor fix, and uses some of the new Rcpp features. Both package is now on CRAN and also in Debian. This releases contains both a nice set of new Armadillo features as well as some nice additions to RcppArmadillo, due again mostly to Romain. Some of the changes tie into the changes changes in Rcpp 0.10.5 as for example the ability to pass const and const ref more efficiently (and we seem to have forgotten an entry in the NEWS file). The complete list of changes is below.
Changes in RcppArmadillo version 0.3.920.1 (2013-09-27)
  • Upgraded to Armadillo release Version 3.920.1 (Agencia Nacional Stasi)
    • faster .zeros()
    • faster round(), exp2() and log2() when using C++11
    • added signum function: sign()
    • added move constructors when using C++11
    • added 2D fast Fourier transform: fft2()
    • added .tube() for easier extraction of vectors and subcubes from cubes
    • added specification of a fill type during construction of Mat, Col, Row and Cube classes, eg. mat X(4, 5, fill::zeros)
  • Initial implementation of wrap<subview>
  • Improved implementation of as<>() and wrap() for sparse matrices
  • Converted main vignette from LaTeX style minted to lstlisting which permits builds on CRAN; removed set BuildVignettes: FALSE.
Courtesy of CRANberries, there is also a diffstat report for the most recent release As always, more detailed information is on the RcppArmadillo page. Questions, comments etc should go to the rcpp-devel mailing list off the R-Forge page.

29 September 2013

Dirk Eddelbuettel: Rcpp 0.10.5

A new version of Rcpp is now on the CRAN network for GNU R; binaries for Debian have been uploaded as well. Once more, this release brings a large number of exciting changes to Rcpp. Some concern usability, some bring new features, some increase performance; see below for the detailed list. We have now released three updates on a quarterly cycle; if we keep this up the next version ought to be ready at the end of December. As in the past, we tested the release rather rigorously by checking against all packages I could (relatively easily) built on my server: this time it successfully passed \code R CMD check for all 107 packages I can build locally out of a total of 136 packages. (Two failed: one for an error in \code Makevars , and one for the need of an X11 server during tests; this may get addressed in the in test script next time). As all of these 107 packages passed, we do not expect any issues with dependent packages. Should there be issues we would appreciate a note, preferably with reproducible code, to the rcpp-devel mailing list. The complete NEWS entry for 0.10.4 is below; more details are in the ChangeLog file in the package and on the Rcpp Changelog page.
Changes in Rcpp version 0.10.5 (2013-09-28)
  • Changes in R code:
    • New R function demangle that calls the DEMANGLE macro.
    • New R function sizeof to query the byte size of a type. This returns an object of S3 class bytes that has a print method showing bytes and bits.
  • Changes in Rcpp API:
    • Add defined(__sun) to lists of operating systems to test for when checking for lack of backtrace() needed for stack traces.
    • as<T*>, as<const T*>, as<T&> and as<const T&> are now supported, when T is a class exposed by modules, i.e. with RCPP_EXPOSED_CLASS
    • DoubleVector as been added as an alias to NumericVector
    • New template function is<T> to identify if an R object can be seen as a T. For example is<DataFrame>(x). This is a building block for more expressive dispatch in various places (modules and attributes functions).
    • wrap can now handle more types, i.e. types that iterate over std::pair<const KEY, VALUE> where KEY can be converted to a String and VALUE is either a primitive type (int, double) or a type that wraps. Examples :
      • std::map<int, double> : we can make a String from an int, and double is primitive
      • boost::unordered_map<double, std::vector<double> >: we can make a String from a double and std::vector<double> can wrap itself
      Other examples of this are included at the end of the wrap unit test file (runit.wrap.R and wrap.cpp).
    • wrap now handles containers of classes handled by modules. e.g. if you expose a class Foo via modules, then you can wrap vector<Foo>, ... An example is included in the wrap unit test file.
    • RcppLdFlags(), often used in Makevars files of packages using Rcpp, is now exported from the package namespace.
  • Changes in Attributes:
    • Objects exported by a module (i.e. by a RCPP_MODULE call in a file that is processed by sourceCpp) are now directly available in the environment. We used to make the module object available, which was less useful.
    • A plugin for openmp has been added to support use of OpenMP.
    • Rcpp::export now takes advantage of the more flexible as<>, handling constness and referenceness of the input types. For users, it means that for the parameters of function exported by modules, we can now use references, pointers and const versions of them. The file Module.cpp file has an example.
    • No longer call non-exported functions from the tools package
    • No longer search the inline package as a fallback when loading plugins for the the Rcpp::plugins attribute.
  • Changes in Modules:
    • We can now expose functions and methods that take T& or const T& as arguments. In these situations objects are no longer copied as they used to be.
  • Changes in sugar:
    • is_na supports classes DatetimeVector and DateVector
  • Changes in Rcpp documentation:
    • The vignettes have been moved from inst/doc/ to the vignettes directory which is now preferred.
    • The appearance of the vignettes has been refreshed by switching to the Bistream Charter font, and microtype package.
  • Deprecation of RCPP_FUNCTION_*:
    • The macros from the preprocessor_generated.h file have been deprecated. They are still available, but they print a message in addition to their expected behavior.
    • The macros will be permanently removed in the first Rcpp release after July 2014.
    • Users of these macros should start replacing them with more up-to-date code, such as using 'Rcpp attributes' or 'Rcpp modules'.
Thanks to CRANberries, you can also look at a diff to the previous release 0.10.4. As always, even fuller details are on the Rcpp Changelog page and the Rcpp page which also leads to the downloads, the browseable doxygen docs and zip files of doxygen output for the standard formats. A local directory has source and documentation too. Questions, comments etc should go to the rcpp-devel mailing list off the R-Forge page

15 September 2013

Dirk Eddelbuettel: RProtoBuf 0.3.1

Two months to the day after the previous 0.3.0 release of RProtoBuf, we are happy to announce a maintenance release 0.3.1 which has now arrived on CRAN. RProtoBuf provides GNU R bindings for the Google Protobuf data encoding library used and released by Google. Once again, Murray lead this release and added a number improvements. The NEWS file entry follows:
Changes in RProtoBuf version 0.3.1 (2013-09-13)
  • Added support for setting and getting 64-bit integer types as character strings of decimal integers to work around R's lack of native 64-bit integer types.
  • Added better error handling, documentation, and tests to the extensions support (getExtension and setExtension).
  • Add support to P for returning extension descriptors.
  • Improved error messages to include field names when invalid fields are specified to protocol buffer messages with new, update, etc.
  • Improved configure to detect and pass -std=c++0x if it is available to enable long long 64-bit integer support in Rcpp.
  • Improved configure in the case when pkg-config is not available or does not know about the google protocol buffer includes.
  • Replaced newly deprecated Rcpp macros with a simplified macro functionality to avoid warnings on the latest development version of Rcpp.
CRANberries also provides a diff to the previous release 0.3.0. More information is at the RProtoBuf page which has a draft package vignette, a 'quick' overview vignette and a unit test summary vignette. Questions, comments etc should go to the rprotobuf mailing list off the RProtoBuf page at R-Forge.

19 August 2013

Dirk Eddelbuettel: Slides from Rcpp talk in Chicago

A couple of days ago, I gave a talk to the Chicago R Users Group which is run ever-so-smoothly by Paul Teetor and Chase Carpenter. The talk provided a brief introduction to Rcpp for R and C++ integration. Slides are now up on my talks / presentations page.

18 August 2013

Dirk Eddelbuettel: Slides from Rcpp talk in Chicagp

A couple of days ago, I gave a talk to the Chicago R Users Group which is run ever-so-smoothly by Paul Teetor and Chase Carpenter. The talk provided a brief introduction to Rcpp for R and C++ integration. Slides are now up on my talks / presentations page.

16 August 2013

Dirk Eddelbuettel: RcppArmadillo 0.3.910.0

A new minor release 3.910.0 of Armadillo came out a few days ago. A new RcppArmadillo release 0.3.910.0 was provided rightaway, and after a brief back-and-forth with CRAN (mostly having to do with the non-standard vignette corresponding to our CSDA paper and its use of LaTeX minted) is now on CRAN and also in Debian. Conrad made some standard cross-products faster, and added vectorise(), any() and all() functions. We also added an extension header for more automated conversion of sparse matrices (which is already used by a post on the Rcpp Gallery). The complete list of changes is below.
Changes in RcppArmadillo version 0.3.910.0 (2013-08-12)
  • Upgraded to Armadillo release Version 3.910.0 (Pyrenees)
    • faster multiplication of a matrix with a transpose of itself, ie. X*X.t() and X.t()*X
    • added vectorise() for reshaping matrices into vectors
    • added all() and any() for indicating presence of elements satisfying a relational condition
  • Added conversion support for sparse matrices (of type double) created by the Matrix package as class dgCMatrix
  • Moved vignette sources from inst/doc to vignettes; set BuildVignettes: FALSE as the minted mode for LaTeX upsets the CRAN builders.
Courtesy of CRANberries, there is also a diffstat report for the most recent release As always, more detailed information is on the RcppArmadillo page. Questions, comments etc should go to the rcpp-devel mailing list off the R-Forge page.

9 August 2013

Dirk Eddelbuettel: inline 0.3.13

A minor maintenance release of inline is now on CRAN, and has just been already included in Debian. This release contains a patch kindly contributed by Mikhail Umorin which fixes the of \code cfunction with lists of signatures and function bodies. The complete NEWS entry is below.
Changes in inline version 0.3.13 (2013-08-08)
  • Applied contributed patch by Mikhail Umorin which corrects cfunction() in the case of a list of signature and body arguments.
Courtesy of CRANberries, there is also a diffstat report for the most recent release. A few more details are available at the R-Forge page.

4 August 2013

Dirk Eddelbuettel: RcppArmadillo 0.3.900.7

A bug-fix release 3.900.7 of Armadillo was provided by Conrad a few days ago. It has been rolled into a new RcppArmadillo release 0.3.900.7 which, after a brief delay, is now on CRAN and in Debian. The main bug fix is a change in the in-place version of reshape(). This in turn caused the CRAN package Amelia to fail, but some diligent debugging by Conrad (which was much appreciated) revealed that the error was in fact in RcppArmadillo where we did not set the matrix flag properly in the constructor. The corresponding one-character fix had been made in this release. Another package exhibiting an issue is unmarked. It needs to make the target variable of reshape a matrix object in a handful of places. I sent the authors a small patch addressing this but have not yet heard back. The complete list of changes is below. In the meantime, I have also done some work on a new as() and wrap() converters for sparse matrices; a first implementation is in SVN.
Changes in RcppArmadillo version 0.3.900.7 (2013-08-02)
  • Upgraded to Armadillo release Version 3.900.7 (Bavarian Sunflower)
    • minor fix for inplace reshape()
    • minor corrections for compilation issues under GCC 4.8+ and MSVC
  • Corrected setting of vec_stats in intialization of row, vector and matrix objects
  • The inline package is no longer used in the examples and unit tests which have all been converted to using Rcpp attributes
Courtesy of CRANberries, there is also a diffstat report for the most recent release As always, more detailed information is on the RcppArmadillo page. Questions, comments etc should go to the rcpp-devel mailing list off the R-Forge page.

16 July 2013

Dirk Eddelbuettel: RProtoBuf 0.3.0

A new release of RProtoBuf -- now at version 0.3.0 -- arrived earlier on CRAN. RProtoBuf provides GNU R bindings for the Google Protobuf data encoding library used and released by Google. Murray added a lot of new features which merited the increase in the minor number. The NEWS file entry follows below:
Changes in RProtoBuf version 0.3 (2013-07-13)
  • Added support for Protocol Buffer extensions through the new getExtension and setExtension methods for Message objects, and added top-level extensions to the descriptor pool.
  • Added more user-friendly show() methods for Messages, Descriptors, and other RProtoBuf objects. A common source of confusion on the mailing lists is to confuse Messages and Descriptors. The new show() methods succinctly display more information about e.g. the number of set fields in a message to make it a little clearer to users what type of object they are working with.
  • Added has method for EnumDescriptor objects to test existance of a named constant in an enumeration, and fix a bug in the length method for these objects.
  • Added a number method for EnumValueDescriptor objects to return the number of a named enumeration constant directly.
  • Improved documentation and unit tests.
CRANberries also provides a diff to the previous release 0.2.6. More information is at the RProtoBuf page which has a draft package vignette, a 'quick' overview vignette and a unit test summary vignette. Questions, comments etc should go to the rprotobuf mailing list off the RProtoBuf page at R-Forge.

14 July 2013

Dirk Eddelbuettel: Go home, Feedly, you're drunk!

With the demise of Google Reader, many of us have scrambled to find alternatives which cover many/most of the features we were used to. For me, synchronised access from web and mobile counts quite high, and I want the (Android) mobile experience to be pretty pleasant too. So with that, and like many other folks, I had shifted over to Feedly. And before I carry on, let me say that Feedly actually does pretty well. They handled the onslaught of new users; they also listened and changed their UI a little to be more compact and Reader-alike And most importantly, it mostly just works. Until it doesn't. Around the time Feedly shifted folks to their own cloud-based backend, posts from my very own CRANberries aggregator of CRAN package changes for R started to show only truncated posts. See this screenshot from reading Feedly: Feedly truncates the CRANberries post and compare it with this one from TheOldReader: OldReader shows the full CRANberries post which contains the diffstat output, nicely marked up and all. Now, it so happens that I am the creator of the very RSS feed I am consuming here, and I wrote it so that I could read the very diffstat output that is now missing. And I know full well that neither the code, nor the hosting (on my own box), nor any other aspect changed. Before confirming this via TheOldReader, I looked at the RSS output, and I tried other frontend and apps --- and it became clear that the bug seems to be at the Feedly cloud storage level. And trying to be a good sport, I submitted a bug report / suggestion, but to no avail (apart from two other chiming in that they see this elsewhere too). Given that I am the coder behind the feed that is displayed in a truncated manner, I am aware that I am using a code stack that is stale (the original Blosxom static web generator) yet which has not posed another problem anywhere in the decade I used it. Nor does the CRANberries feed pose a problem when aggregated and viewed via the TheOldReader code path, or for that matter via Planet R which also carries (parts of) CRANberries. So for now, I have to either pivot out of the CRANberries RSS reading in Feedly and go directly to the webpage (fine, but cumbersome and in need of a working connection) or use a second subscription mechanism. I appreciate what the TheOldReader folks are doing---presumably with a minor fraction of the resources available to Feedly---and may hang with them for now. But it would be awfully nice if Feedly could sleep off its hangover and fix this. In which case I'd be much happier recommending its service.

Dirk Eddelbuettel: Slides from Rcpp talk in Sydney

The Sydney Users of R Forum (SURF) were kind enough to host me two days (well, three with the traveling ...) ago for an hour-long talk on Rcpp. Apparently, it set a new attendance record for this R user group. My thanks to Louise and Eugene for hosting a terrific meeting, and taking me out for a bite and drink afterwards. I have now put up my slides on my talks / presentations page for anyone else to peruse.

25 June 2013

Dirk Eddelbuettel: Rcpp 0.10.4

A new version of Rcpp is now on the CRAN network for GNU R; binaries for Debian have been uploaded as well. This release brings a fairly large number of fixes and improvements across a number of Rcpp features, see below for the detailed list. We are also announcing with this release that we plan to phase out the RCPP_FUNCTION_* macros. Not only have they been superceded by Rcpp Modules and Rcpp Atributes (each of which has its own pdf vignette in the Rcpp package), but they also appear to be at best lightly used. We are for example not aware of any CRAN packages deploying them. To provide a smooth transition, we are aiming to keep them around for another twelve months, but plan to remove them with the first release after that time window has passed. As before, we tested the release rather rigorously by checking against all packages I could (relatively easily) built on my server: this time it covered 91 of the 124 CRAN packages depending on Rcpp. As all of these 91 packages passed their checks, we do not expect any issues with dependent packages. The complete NEWS entry for 0.10.4 is below; more details are in the ChangeLog file in the package and on the Rcpp Changelog page.
Changes in Rcpp version 0.10.4 (2013-06-23)
  • Changes in R code: None beyond those detailed for Rcpp Attributes
  • Changes in Rcpp attributes:
    • Fixed problem whereby the interaction between the gc and the RNGScope destructor could cause a crash.
    • Don't include package header file in generated C++ interface header files.
    • Lookup plugins in inline package if they aren't found within the Rcpp package.
    • Disallow compilation for files that don't have extensions supported by R CMD SHLIB
  • Changes in Rcpp API:
    • The DataFrame::create set of functions has been reworked to just use List::create and feed to the DataFrame constructor
    • The operator-() semantics for Date and Datetime are now more inline with standard C++ behaviour; with thanks to Robin Girard for the report.
    • RNGScope counter now uses unsigned long rather than int.
    • Vector<*>::erase(iterator, iterator) was fixed. Now it does not remove the element pointed by last (similar to what is done on stl types and what was intended initially). Reported on Rcpp-devel by Toni Giorgino.
    • Added equality operator between elements of CharacterVectors.
  • Changes in Rcpp sugar:
  • Changes in Rcpp build tools:
    • Fix by Martyn Plummer for Solaris in handling of SingleLogicalResult.
    • The src/Makevars file can now optionally override the path for /usr/bin/install_name_tool which is used on OS X.
    • Vignettes are trying harder not to be built in parallel.
  • Changes in Rcpp documentation:
    • Updated the bibliography in Rcpp.bib (which is also sourced by packages using Rcpp).
    • Updated the THANKS file.
  • Planned Deprecation of RCPP_FUNCTION_*:
    • The set of macros RCPP_FUNCTION_ etc ... from the preprocessor_generated.h file will be deprecated in the next version of Rcpp, i.e they will still be available but will generate some warning in addition to their expected behavior.
    • In the first release that is at least 12 months after this announcement, the macros will be removed from Rcpp.
    • Users of these macros (if there are any) should start replacing them with more up to date code, such as using Rcpp attributes or Rcpp modules.
Thanks to CRANberries, you can also look at a diff to the previous release 0.10.3. As always, even fuller details are on the Rcpp Changelog page and the Rcpp page which also leads to the downloads, the browseable doxygen docs and zip files of doxygen output for the standard formats. A local directory has source and documentation too. Questions, comments etc should go to the rcpp-devel mailing list off the R-Forge page

21 June 2013

Dirk Eddelbuettel: Upcoming Rcpp talk in Sydney

The Sydney Users of R Forum (SURF) will be hosting me for a talk on July 10. The focus will be Rcpp for R and C++ integration, and the intent is to have this be really applied with lots of motivating examples. Organizers Louise and Eugene were able to move this to a slightly larger room as the initial capacity of 50 was filled almost immediately. As of right now, the talk page shows a few available slots. If you're in the Sydney are in early July, why not register and swing by? If you're not in the area, the Rcpp Events page lists other upcoming talks too.

13 June 2013

Dirk Eddelbuettel: inline 0.3.12

Another maintenance release of inline is now on CRAN and in already included in Debian. This release was triggered by a change in the development version of R which removed an argument to package.skeleton(). The complete NEWS entry is below.
Changes in inline version 0.3.12 (2013-06-12)
  • Align package.skeleton with the R-devel version of the function which no longer has a NAMESPACE argument.
  • Remove copy of LGPL in source archive as R CMD check now complains about it. License status continues to be specified in file DESCRIPTION.
Courtesy of CRANberries, there is also a diffstat report for the most recent release. A few more details are available at the R-Forge page.

Next.

Previous.