Hadley Wickham
2015-Apr-28 16:21 UTC
[Rd] install.packages no longer respects quiet = TRUE ?
> install.packages("plyr", quiet = T)trying URL 'http://cran.rstudio.com/bin/macosx/mavericks/contrib/3.2/plyr_1.8.2.tgz' Content type 'application/x-gzip' length 855795 bytes (835 KB) =================================================downloaded 835 KB The downloaded binary packages are in /tmp/RtmpOcorLA/downloaded_packages I don't have R 3.1.3 handy, but I'm pretty sure this used to produce no output. Hadley -- http://had.co.nz/
Dirk Eddelbuettel
2015-Apr-28 16:41 UTC
[Rd] install.packages no longer respects quiet = TRUE ?
On 28 April 2015 at 11:21, Hadley Wickham wrote: | > install.packages("plyr", quiet = T) | | trying URL 'http://cran.rstudio.com/bin/macosx/mavericks/contrib/3.2/plyr_1.8.2.tgz' | Content type 'application/x-gzip' length 855795 bytes (835 KB) | =================================================| downloaded 835 KB | | | The downloaded binary packages are in | /tmp/RtmpOcorLA/downloaded_packages | | | I don't have R 3.1.3 handy, but I'm pretty sure this used to produce no output. Canot replicate with R 3.2.0 on Ubuntu, and it doesn't matter if I use T or TRUE. Dirk -- http://dirk.eddelbuettel.com | @eddelbuettel | edd at debian.org
Prof Brian Ripley
2015-Apr-28 17:20 UTC
[Rd] install.packages no longer respects quiet = TRUE ?
> On 28 April 2015 at 11:21, Hadley Wickham wrote: > | > install.packages("plyr", quiet = T) > | > | trying URL 'http://cran.rstudio.com/bin/macosx/mavericks/contrib/3.2/plyr_1.8.2.tgz' > | Content type 'application/x-gzip' length 855795 bytes (835 KB) > | =================================================> | downloaded 835 KB > | > | > | The downloaded binary packages are in > | /tmp/RtmpOcorLA/downloaded_packages > | > | > | I don't have R 3.1.3 handy, but I'm pretty sure this used to produce no output.First, the description is quiet: logical: if true, reduce the amount of output. not 'no output'. This 'change' is seen only for installs of binary packages. The difference is that the default is now type = "both", and 3.1.3 used type = 'binary': adding type = "binary" produces what you expected in 3.2.0, and type = "both" is already quieter in R-patched. (Note that it is rather difficult to test install.packages for Mac binaries prior to release as that relies on a 'CRAN distribution' which pretty much has to be the default version on the machine.) -- Brian D. Ripley, ripley at stats.ox.ac.uk Emeritus Professor of Applied Statistics, University of Oxford 1 South Parks Road, Oxford OX1 3TG, UK