install.package() for the source packages [🇷 for BE/BA]

posted by yjlee168 Homepage – Kaohsiung, Taiwan, 2020-07-20 02:53 (868 d 20:37 ago) – Posting: # 21753
Views: 9,465

Dear mittyri,

Thank you so much for your comments.

❝ ... By the way if you install locally that's another story, since

Not used if repos = NULL...

❝ So it is not possible to install dependent packages for a locally installed package within install.package()


Exactly. Recently I just wondered: I used to do install.packages("foo") for all dependent or required packages before installing bear. That meant I used install.packages("foo",depend=NA) as default. And it was OK. Now I use install.packages("foo",depend=TRUE) and it is OK too. So I was wondering if we really need to set depend=TRUE. I also found if I set depend=TRUE R will install much more packages in that case. For example, I need to install package A. Package A depends on 3 packages (package B, C and D). Should I also install simultaneously package B, C and D? Furthermore, if package B depends on another 3 packages (package E, F & G), then will the option of depend=TRUE for package A also enable R to install package E, F & G? Therefore, I do not think we need to set depend=TRUE. I think R will check all required packages first for the package that we want to install. Then R will install these required packages first and leave the package that we need to install at last. Am I correct?

❝ There are solutions available:

❝ - use devtools::install_local; there's an option to install dependencies from CRAN


Great. I will try that.

❝ - create and use your own repo


Unless go to Github. Not a chance for Sourceforge. I will think about that too.

All the best,
-- Yung-jin Lee
bear v2.9.1:- created by Hsin-ya Lee & Yung-jin Lee
Kaohsiung, Taiwan https://www.pkpd168.com/bear
Download link (updated) -> here

Complete thread:

UA Flag
Activity
 Admin contact
22,428 posts in 4,694 threads, 1,598 registered users;
22 visitors (0 registered, 22 guests [including 15 identified bots]).
Forum time: 22:30 CET (Europe/Vienna)

Statistics is the art of never having to say you’re wrong.
Variance is what any two statisticians are at.    C.J. Bradfield

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5