-
Notifications
You must be signed in to change notification settings - Fork 990
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
trim suggested dependencies #5745
Comments
I tinkered about removing |
rmarkdown (many, and some heavy, recursive deps), nanotime (heavy recursive deps) and curl (OS dep) are perfect candidates for removal.
|
I too wish I went the hard route in a few of my packages and discarded rmarkdown/knitr in favour of lighter alternative but I understand that that is not everybody's coup of tea. And I concur that we no longer need to cater to users of R 3.1.0. |
@eddelbuettel support of nanotime in data.table would not change at all, we don't need to access nanotime namespace for that. Only tests require loading nanotime that AFAIK, and those tests could be migrated to nanotime package. |
Doh, I missed the gist of that. No problem, happy to farm some tests out to my side and have them tickle up in reverse-dependency checks. [ Also, I am in your crosshairs for the coordinated change of renaming the C API access point I once contributed. (I never quite understood why it needed renaming, but I defer to team data.table here.) So if you need me to change that in the 'consuming' package that accesses it I can of course help with a coordinated upload. Either 1.14.10 or 1.15.0. Happy to assist. ] |
@ben-schwen in case you abandoned your PR branch let us know, curl is now last missing piece to close this issue |
resolved |
Currently our most heavy dep is knitr. Fortunately it is not that heavy one. Anyway deps could be further improved if/when rstudio/markdown#109 (knitting vignettes directly from markdown pkg without knitr) will be provided |
Complete installation of data.table + suggested dependencies results in following packages being installed in a library. While data.table doesn't have any hard dependencies, the soft ones could be trimmed to keep that to the minimum. Not sure how easy, and at what cost, would be possible.
including recursive deps, total 38 packages
and few of them are quite heavy.
curl
package also imposes extra OS dependency , devel version oflibcurl
.Broken down by each Suggested dep:
The text was updated successfully, but these errors were encountered: