Data Science. TileDB. Open Source. Quant Research. R. C++. Debian. Linux. Adjunct Clinical Professor, University of Illinois. Lots of coffee. And some running.
It's the final coercion that does it: somewhat known and documented fact that the S3 class object gets dropped by #rstats. You can assemble proper (one row) data.frame objects (that you can then `rbind`), or you can avoid fate by going to character. Inject some `print` to see.
Just use package RPostgreSQL (older, solid) or RPostgres (newer, 'tidy', more depends) and connect directly. Either package uses the binary connector to Postgres so you don't have to worry about details. #rstats
#r2u going from strength to strength: over 100 @github stars, and the shell loop `grep -c`ing access in 15 min intervals saw new high of nearly 20k downloads in 15 mins.
#r2u: CRAN packages as binaries for Ubuntu: fast, reliable, and cheap.
More at eddelbuettel.github.io/r2u/
Indeed a much under-appreciated aspect because they are encoded scripts, with proper dependencies within the distribution of the underlying image, and as well as with proper existence-tests proofing by actually building images. Code as infrastructure, infrastructure as code.
Character variables are internally hashed so you save on use of repeated values; that was an important change many (many!!) moons ago. We can all bow in the general direction of Iowa and thank @LukeTierney4 for this (and so many other internal improvements). #rstats
Congratulations -- and as usual, uploaded the @Debian package for 1.29, update my PPA for @Ubuntu, verified that the RQuantLib package builds and tests unchanged,and updated the qlcal package for the calendaring changes.
Just a regular long weekend Sunday, yet #r2u already served another 35.9k @Ubuntu binary CRAN packages for #Rstats. As a reminder of how much this rocks, a recent shell demo 'video' of installing all of #tidyverse in 18 seconds, fail safe, with all dependencies, and fast.
Some people have analysed the output from `history()` (also in a file) that way to see which #Rstats functions they hit more / most.
As for your NA question: It depends. But there is an entire CRAN Task View on missing data and its treatment:
cran.r-project.org/web/views…