jennybc
commented
2 months ago I am literally laughing out right now because I see this all the time and none of my colleagues do! Thanks for reporting and for the investigation you've done so far. 😅
Open mattkerlogue opened 2 months ago
jennybc
commented
2 months ago I am literally laughing out right now because I see this all the time and none of my colleagues do! Thanks for reporting and for the investigation you've done so far. 😅
mattkerlogue
commented
2 months ago I am literally laughing out right now because I see this all the time and none of my colleagues do! Thanks for reporting and for the investigation you've done so far. 😅
When it first started happening I thought I was going mad and couldn't work out what was going on. It seemed to be random as I couldn't work out how to replicate it.
At some point I googled the author names and they seemed to be connected to {ade4} but wasn't the current set of authors in the DESCRIPTION or CITATION file. And as I didn't have that package installed I was utterly baffled.
Eventually I clocked that it seemed to be happening when I was using {skimr}, and subsequently seemed to be only when the package was being loaded for the first time.
This lunchtime I finally got sufficiently fed up that I thought I'd try and work out was going on because I wasn't sure it was an issue with {skimr}. Some my afternoon included manually typing into the console every single one of my installed packages 🙃🫠.
jennybc
commented
2 months ago At some point I googled the author names and they seemed to be connected to {ade4} but wasn't the current set of authors in the DESCRIPTION or CITATION file. And as I didn't have that package installed I was utterly baffled.
Yeah I (and other folks I have dragged into this in slack) have been through all of the same motions. For me it's usually (always?) devtools::document() that triggers the warning.
cathblatter
commented
1 month ago TYSM for reporting this!! I have the same and thought I'm going crazy 😅
System details:
Positron and OS details:
Positron Version: 2024.08.0 (Universal) build 48 Code - OSS Version: 1.91.0 Commit: ed616b36655fb24d116108bdd833f1321704315b Date: 2024-08-19T04:26:51.868Z Electron: 29.4.0 Chromium: 122.0.6261.156 Node.js: 20.9.0 V8: 12.2.281.27-electron.0 OS: Darwin arm64 23.3.0
Interpreter details:
R 4.4.1
Describe the issue:
I have a very weird issue whereby I get a very strange and unexpected warning message appear when I'm typing in the console. I've identified it occurring when I start to call functions from certain unloaded packages via their namespace, it also occurs immediately after I attach these package via a
library()call.This issue only happens with 6 of the 307 packages that are installed across my user and system library, namely: {skimr}, {formattable}, {knitr}, {quarto}, {rmarkdown}, {roxygen2}. All of these are at their latest version. In the case of {knitr}, {quarto} and {rmarkdown} I can confirm the issue also happened with the previous release (I noticed these three packages were behind the latest while exploring this issue and updated them).
What is more puzzling is the nature of the error message which is the same for all six packages and seems to be unrelated to either the packages themselves or the process of loading/attaching them.
Screen capture:
An initial Google of the author names implied it might be related to the {ade4} package, but this isn't a package I've ever used or installed, and the current set of authors in the {ade4} DESCRIPTION and CITATION files is different
A further Google led me to discover that it seems to be related to the
tweak_bib.csvfile in the {knitr} package.Of the 6 packages this issue affects, 4 import {knitr} and {quarto} imports {knitr} indirectly via its import of {rmarkdown}. However, I have are two installed packages that do import {knitr} but where this issue does not occur: {htmlwidgets} and {reprex}.
The issue only occurs the first time I try to load one of these packages, i.e. within the same session a call to a function in a different package does not cause a repeat of the message, but the message does reoccur when I restart the R session within the same Posit window/instance.
This issue has does not happen when using RStudio or the R GUI for Mac OS X, which suggests its maybe something related to ark but I can't raise an issue in that repo.
Steps to reproduce the issue:
knitr::knit()in the console, message should appear shortly after the second colon is typed. Alternatively message will appear afterlibrary(knitr).knitr::knit()orlibrary(knitr)can be supplemented for any of the packages mentioned.Expected or desired behavior:
No error message.
Were there any error messages in the UI, Output panel, or Developer Tools console?
Dumps from the Developer Tools and Output Channel logs from a fresh window where I typed
knitr::knit()but I did not execute the command - message showed after the second colon while I was typing "knit" as per the screencap above.Dump of the Developer Tools console
Dump of the Output Channel log for the Console