[R-pkg-devel] Will use split option
Alex Hallam
@|exh@||@m6@28 @end|ng |rom gm@||@com
Wed Jul 10 15:26:58 CEST 2019
On Wed, Jul 10, 2019 at 6:14 AM Dirk Eddelbuettel <edd using debian.org> wrote:
>
> On 10 July 2019 at 05:53, Alex Hallam wrote:
> | I am working on a data package. Due to the size of the package I opted
> | to store the package as a drat repo.
> |
> | I am able to install my package locally, but I can't seem to install the
> | package
> | from the drat repo. The error I get is:
> |
> | ```
> | Installing package into ‘/usr/local/lib/R/site-library’
> | (as ‘lib’ is unspecified)
> | trying URL '
> | https://alexhallam.github.io/drat/src/contrib/cfsalesdata_0.0.1.1.tar.gz
> '
> | Content type 'application/gzip' length 134 bytes
> | ==================================================
> | downloaded 134 bytes
> |
> | Error in untar2(tarfile, files, list, exdir, restore_times) :
> | incomplete block on file
> | Warning in install.packages :
> | installation of package ‘cfsalesdata’ had non-zero exit status
> | ```
> |
> | These are the commands I used to install my data package. The local
> | install works, but my two attempts at installing from a remote repo
> failed.
> |
> | # This works on my local computer
> | install.packages(file.choose(),repos = NULL, type = "source")
> |
> | # Install from remote (fails)
> | install.packages("cfsalesdata",repos = "
> https://alexhallam.github.io/drat/",
> | type = "source")
> |
> | # Install from remote with drat (fails)
> | library(drat)
> | addRepo("alexhallam")
> | install.packages("cfsalesdata")
> |
> | My hunch is that this has to do with the tar ball being tracked with "git
> | large file storage". This is because the file size of my tar ball is
> 154.9
> | MB which exceeds github's 100 MB file size limit. What options can I try
> in
> | this situation?
>
> Shucks. Should have realized the >100gb issue when I reocmmended drat. We
> have GH LFS support as a feature request issue, but I cannot / will not
> work
> on it as I do not have such files or repos ...
>
> Easiest hack: split your data package into chunks < 100gb and use standard
> more.
>
> Sorry about that, Dirk
>
> --
> http://dirk.eddelbuettel.com | @eddelbuettel | edd using debian.org
>
Thanks everyone. I will go ahead and split the repo into a couple pieces.
--
Thanks!
-Alex
[[alternative HTML version deleted]]
More information about the R-package-devel
mailing list