I need to install R package fOpitons for a learning module required by society of actuaries. This package seems to be no longer available on CRAN. It was taken down in April 2022 for misrepresentation of authorship and ownership of copyright. None of the links to former versions of the package works. How can I install this package?
https://cran.r-project.org/web/packages/fOptions/index.html
None of the links in this site works.
https://cran-archive.r-project.org/web/checks/2022/2022-04-29_check_results_fOptions.html
(Note for eager "close as duplicate": the Archive/ link for this package is (currently?) broken which is very unusual for CRAN so I spelled out alternative approaches below. That part is likely not a duplicate. Cheers. -D.)
CRAN packages are open source, and even when removed from the current CRAN index for reasons deemed necessary by the CRAN team will always remain available in the Archive/ section of the package.
So the link you provide has as its second line
Formerly available versions can be obtained from the archive.
where the archive link should provide you access to various versions.
However, it is not working, which looks like an error. So I just alerted CRAN to that.
I happen to have maintained fOptions in Debian for a number of years. So you still get the Debian binary packages -- and sources. Most recent versions are at https://packages.debian.org/sid/r-cran-foptions point for example to the source archive: http://deb.debian.org/debian/pool/main/f/foptions/foptions_3042.86.orig.tar.gz
If you can install from source this should do.
Otherwise, there is also a CRAN mirror at GitHub so
https://github.com/cran/fOptions is another choice.
Related
I have troubles with installing the package fwildclusterboot. It says that it is no longer available in CRAN. Also didn't succeed even when I exploited it from the archive. I was searching for some new version but could not find it.
Can you please help me how to do it?
I am the author of fwildclusterboot.
Due to issues with a broken dependency I did not manage to fix on time, fwildclusterboot is temporarily not available on CRAN. The long story is that a dependencies' dependency failed a test, then the dependency was not fixed, therefore archived on CRAN, and I did not manage to a) drop the dependency on time and pass all CRAN notes within the given two-three week time limit. Now I am trying to bring the package back, but CRAN is not very responsive towards me.
In the meantime, you can download the package from R universe (a CRAN alternative, set up by rOpensci) by running install.packages('fwildclusterboot', repos ='https://s3alfisc.r-universe.dev'). I hope the package will be back on CRAN soon!
I'm looking for an updated version of this:
https://cran.r-project.org/web/packages/lfe/index.html
package. It seems to have been removed from CRAN. How could I find out about the history of this process to maybe find out if it was migrated to a new package or just isn't maintained anymore?
The package was back on CRAN on 11 January 2021. It has been updated several times since.
The page you linked provides some information:
Package ‘lfe’ was removed from the CRAN repository.
Formerly available versions can be obtained from the archive.
Archived on 2020-12-04 as check problems were not corrected in time.
My conclusions:
as of today (11 Dec 2020), it was removed only a week ago (Dec 4)
it appears to have been archived by CRAN maintainers, not withdrawn by package authors/maintainers.
if we go to the page and retrieve the most recent version (2.8-5.1, 2020-07-08) and open the DESCRIPTION file in the archive, we can find the URL for the package on github: https://github.com/sgaure/lfe (we could also get there by googling "lfe package github"). You can also go to the issues list on that repo to find this discussion. It's long, but it looks like the maintainers are hard at work trying to get the package back on CRAN.
Not all packages have their source code/issue-tracking lists on a public site, bot those that do will generally have it listed in the URL: field of the DESCRIPTION file.
You can probably install the latest archived version via remotes::install_version("lfe", version="2.8-5.1") (you might need development tools [compilers etc.] installed)
As Ben noted, the package has been resubmitted (I am one of the contributors) and should hopefully be back on CRAN soon.
But, FWIW and in the interim, I've created a simple package that takes existing lfe::felm scripts and converts them to the fixest::feols equivalent. I expect most people will move to the latter over time anyway, given its advantages. https://github.com/grantmcdermott/lfe2fixest
Was unable to install the package ‘subgraphMining’. The error says "is not available (for R version 3.4.3)". What steps are needed?
There is a package named subgraphMining which you can find by searching with Google using that term. (It's not a CRAN or Github hosted package.) It's found at a book website, http://www.csc.ncsu.edu/faculty/samatova/practical-graph-mining-with-R/PracticalGraphMiningWithR.html . It does require an additional package named igraph0 and there is an answered question on SO describing that issue. The igraph0 package is likewise not "available" for the current version of R but it is in the CRAN archives. So you also need the development tools for your OS (Mac in my case, so XCode and the Command Line Tools).
siteURL <- "https://www.csc2.ncsu.edu/faculty/nfsamato/practical-graph-mining-with-R/R-code/FrequentSubgraphMining.zip"
# Since I'm not a windoze user I unzip to a local disk.
install.packages("https://cran.r-project.org/src/contrib/Archive/igraph0/igraph0_0.5.7.tar.gz",
repo=NULL, type="source")
# I downloaded the package from the chapter: Frequent Subgraph Mining
install.packages("~/Downloads/FrequentSubgraphMining/subgraphMining_1.0.tar.gz",
repo=NULL, type="source")
library(subgraphMining)
It's not written particularly well. It doesn't, for instance, list any Imports or Depends in the package DESCRIPTION file and the author of igraph thinks that the package authors should have rewritten it to use the igraph package which is currently maintained. But I think that there is quite a bit of potential value in having its installation be possible in support of what appears to be a rather interesting set of methods described in hte book.
I am currently running R on mac osx but am looking to purchase a linux server for more power. Is there any way that I can check for specific R packages whether they will also work on linux? (before, of course, I actually buy the server and try to install and run the given packages). Also, is there any way to determine if a given package would run on certain linux distributions but not others (e.g. Ubuntu vs. Debian)?
Assuming the package is on CRAN, go to the package's CRAN page, e.g. https://cran.r-project.org/package=zoo and then click on the link to the right of CRAN checks which in this example would be labelled zoo results. It would take you to this page: https://cran.r-project.org/web/checks/check_results_zoo.html showing the results of checking that package on various different platforms.
If the package is not on CRAN but is on github and the developer checks it with Travis-CI then you can view the check by clicking on the Travis-CI icon. For example, the klmr modules package is not on CRAN (there is a CRAN package of the same name but it's different); however, if you look at its github home page at https://github.com/klmr/modules and click on the icon which currently is black and green and reads build passing (but could read something else if there are changes to the package or R that breaks tests) then you will be taken to the Travis-CI tests at https://travis-ci.org/klmr/modules .
tl;dr Slightly opinion/personal experience based, but I would be surprised if there were any CRAN package that you couldn't get running on Linux.
In general Unix users tend to install packages from source: CRAN doesn't provide binaries, but source installation is usually painless. The package binaries that are available (the CRAN Linux page has links for Debian, Ubuntu, SUSE, and Red Hat) tend to focus on packages that have extra system-level dependencies (e.g. FFT libraries, or spatial data analysis libraries) where it's more of a nuisance to assemble the needed dependencies for a particular system.
From the CRAN repository policy:
Package authors should make all reasonable efforts to provide cross-platform portable code. Packages will not normally be accepted that do not run on at least two of the major R platforms [i.e. Windows, MacOS, Linux]. Cases for Windows-only packages will be considered, but CRAN may not be the most appropriate place to host them.
When a package fails to run on of one of the three platforms, it's usually Windows. The only package I've ever had real trouble installing on Linux is
R2OpenBUGS on 64-bit systems, because it requires installing a 32-bit toolchain.
I want to use some packages, but they has been deleted in CRAN, though their formerly available versions can be obtained from the archive. Those packages are not in r-forge, too.
For instance, dynamo and gafit.
Though those packages have been removed, I find them still useful and can help me.
Is it possible to and how to install them easily? like one_line_install('http://sss.tar.gz').
By the way, I may use them in Windows and/or in OS X. So the code should be designed as "multi-platform".
It is easy using the devtools package as there is no need to download the package yourself. For example:
library(devtools)
install_url('http://cran.r-project.org/src/contrib/Archive/dynamo/dynamo_0.1.3.tar.gz')
install_url('http://cran.r-project.org/src/contrib/Archive/gafit/gafit_0.4.tar.gz')
I have no doubt this will be platform-independent.
The package has other related and useful functions such as install_version, install_local, install_github, etc.
Depending on how far back you want to go, you can also use the MRAN snapshot archive. This is a sequence of daily snapshots of CRAN, going back to September 2014.
Eg to install a package from CRAN as it was on 30 June 2015:
install.package("my_package",
repos="https://mran.microsoft.com/snapshot/2015-06-30")
Try
install.packages('/path/dynamo_0.1.3.tar.gz', type = 'source')
where path is the path to the file you downloaded. That is it on a Mac! On Windows you have to play a bit with the slashes :-)
For what it's worth, package gafit is available again without resorting to tricks.
As suggested above it was broken for some years as the core R system changed some packaging conventions. Hopefully fully fixed now.