`tibble::view()` fails in Ark (& Positron) with "variable names are limited to 10000 bytes" error

[coatless]

Description

The tibble::view() function (accessed via %>% view) fails when executed in Ark and Positron environments, while working correctly in RStudio and R.app. This prevents users from viewing data frames using the tidyverse pipe syntax.

Reproduction Steps

library(tidyverse)
library(nycflights13)
df1 <- weather %>% view

Expected vs Actual Behavior

✅ Working Environments (RStudio, R.app)	❌ Failing Environments (Ark, Positron)
R.app (macOS)	Positron
RStudio	Ark (Jupyter Console)

Error Details

Error Message:

Error in `exists()`:
! variable names are limited to 10000 bytes

Stack Trace:

Hide Traceback
    ▆
 1. ├─weather %>% view
 2. ├─tibble::view(.)
 3. │ └─rlang::eval_tidy(quo(view_fun(!!x, !!title)))
 4. └─view_fun(`<tibble[,15]>`, ".")
 5.   ├─base::isTRUE(exists(name, envir = env, inherits = FALSE))
 6.   └─base::exists(name, envir = env, inherits = FALSE)

Possible cause

The issue appears to stem from how tibble::view() attempts to retrieve the View function:

view_fun <- get("View", envir = as.environment("package:utils"))

Reference: ark/src/modules/positron/view.R#L51

ark/crates/ark/src/modules/positron/view.R

Line 51 in 74d3f1d

view_function <- function(

The view_fun() function is attempting to retrieve the View function from the utils package, but this lookup mechanism seems to be incompatible with Ark's environment setup, resulting in the "variable names are limited to 10000 bytes" error during the exists() call.

Environment Information

• Tidyverse version: 2.0.0
• R version: 4.5.1
• Affected platforms: Ark 0.1.195, Positron 2025.07.0 (Universal) build 204
• Working platforms: RStudio (Version 2025.05.1+513 (2025.05.1+513)), R.app (R 4.5.1 GUI 1.82 High Sierra build (8536))

Reported initially on Reddit by atthemost7.

[juliasilge]

This works for me with tibble v3.3.0:

Is it possible that this is the same as posit-dev/positron#4702 and posit-dev/positron#5392 and the version of tibble is not up to date?

[coatless]

@juliasilge yes exact issue. Sorry for the noise.

tibble 3.2.1 (installed) vs. tibble 3.3.0 (patched)

[juliasilge]

Great! Glad to hear it 🙌