f <- function(very_fast, incredibly_dangerous){ data.frame("very_fast" = very_fast, "incredibly_dangerous" = incredibly_dangerous) } df <- f(v = 1, i = 2) df$very * df$inc df$very_broken <- 3 df$very * df$inc
Partial matching of names in #rstats is wild. I knew the function would run, but didn't know it "works" with dataframes as well!
I guess there are some good / historical reasons for this, but it's not an ideal situation and doesn't encourage developers to write tests.
When submitting #Rstats packages to CRAN you can either: spend weeks writing a comprehensive test suite and get threatened with removal when one don't pass under some unusual condition, OR don't write any tests, R CMD check still passes, most users are none the wiser and you won't be threatened.
I can't find the exact versions but this might be useful: ropensci.org/blog/2026/01... R-hub has three versions of mac but not exactly corresponding r-hub.github.io/rhub/ (I don't understand why it's so hard to test on machines that match what CRAN use)
thanks - I will give it a go.
It's surprising that neither R CMD INSTALL nor R CMD check flag having two identically named functions in an #rstats package. It's easy to accidentally duplicate when refactoring code and can lead to much confusion!
There's a version from last year on archive.org which looks similar. "You do not need to be registered with any statistical organisation to come." suggests it isn't RSS any more.
Not sure if that's the same thing?
'<svg><text x="0" y="200" style="font-size: 100px;"> Hello world</text></svg>' |> charToRaw() |> rsvg::rsvg_svg(NULL) |> rawToChar()
Is there a term for figuring something out that LLMs say isn't possible? Grind coding? Organic coding? In this case, rendering <text> elements in an SVG to <path> in #Rstats. (Claude cooked up a whole non-functioning package in an attempt to do it).
That doesn't sound like too much data for leaflet to handle. I noticed your time column is chr rather than a date/time which might be causing problems - maybe some are formatted differently? It's probably worth trying to convert with that column with as.Date() as a first try.
Do you mean www.rdocumentation.org/packages/lea... ? I adapted the example to use 200 data points and that works fine. Can you share an example showing the problem?
(which is what lazydata loads)
Confusingly that package put the data in "cancer.rda" which seems to be what data() references, but that file evidently contains the data in a variable called rotterdam.
Agree that would be cool. There's a CRAN mirror on github but someone needs to make that a single repo and feed it into deepwiki.com
I've spent a lot of time wrestling with base #rstats plots in Shiny this year and thought I'd share the solution I've come up with to ensure they scale correctly and downloads match exactly what is shown in the app: simon-smart88.github.io/forest_plot_...
This is definitely not the case - people have been free to fork wikipedia from the get go as everything is CC licenced. All contributors are made aware of this every time they edit an article.
Also, are you aware of {wallace}? That's what I forked to make {shinyscholar} and seems to have some crossover with what you're doing - I think it can load paleo data as well as current data.
Interesting to see you used {shinymeta} I'd looked at it but wasn't sure how scalable it would be for complex analyses. github.com/simon-smart8... is my attempt at making a generic solution for reproducible apps
* on top!
My vote would be on. I'd also lose the lines between years as they imply there could be data between the points when that's not the case.
.libPaths() gives:
"/home/simon/R/x86_64-pc-linux-gnu-library/4.5"
"/usr/local/lib/R/site-library"
"/usr/lib/R/site-library"
"/usr/lib/R/library"
But I'm not sure how to check where devtools::test() is looking for binaries. I tried installing mirai from github, but test() still couldn't find it
Also, not sure if this is related, but it seems like it could be. Last week mirai wasn't being updated by apt and I may have done something stupid in the process of trying to remove it to force the update. Could remove.packages("mirai") have broken something?
I've run R CMD check and there are no mirai-related problems, but I get `Error: Reason: On CRAN` when it tries to run the tests.
Not that I know of.
I have mirai 2.2.0 in "/usr/lib/R/site-library" for some reason, but remove.packages() won't remove it.
It's definitely in the DESCRIPTION. Tried adding that print statement in various spots e.g. to testthat.R but it doesn't get executed before the error message.
==> devtools::test() ℹ Testing metainsight Error in load_imports(path) : The package "mirai" is required. Calls: <Anonymous> ... load_imports -> deps_check_installed -> check_installed Execution halted Exited with status 1.
#rstats when I run devtools::test() from the Build pane I get an error that mirai isn't installed even though it is. I use r2u and have already uninstalled with apt remove and reinstalled. It runs okay from the terminal but whatever environment shinytest2 runs in can't find mirai either. Any ideas?
