Dan Chaltiel
LightNews LightNews
danchaltiel.bsky.social
Dan Chaltiel
@danchaltiel.bsky.social
110 followers 230 following 89 posts
R nerd, developer of the {crosstable} package.
Biostatistician (PharmD, PhD) in clinical cancer research.
https://github.com/DanChaltiel
Posts Replies Media Videos
danchaltiel.bsky.social
Hey #Rstats!
Tired of manually adding @importFrom all over your 📦 #Rpackage?
Say hello to {autoimport}, now on CRAN! 🚀
It parses your code and adds the necessary tags automatically, at the function or package level.
Less typing, more coding! 🔥
github.com/DanChaltiel/...
February 7, 2025 at 7:47 AM
danchaltiel.bsky.social
The M-opened vs M-closed problem is what keeps me up at night 😱
blog.djnavarro.net/posts/2023-0...
im fine its just that life is meaningless meme
January 7, 2025 at 8:05 PM
danchaltiel.bsky.social
But wouldn't this be more readable using ordered factors?
December 12, 2024 at 8:22 AM
danchaltiel.bsky.social
November 22, 2024 at 7:26 PM
danchaltiel.bsky.social
Hey #rstats!
I made the {autoimport} package 📦, which automatically adds `@importFrom` calls in your #rpackage code.
I would love to get some feedback on it.
Would anyone be interested in trying it out?
Run the code in the image, and let the console guide you 😁
More info: github.com/DanChaltiel/...
pak::pak("DanChaltiel/autoimport") autoimport::autoimport()
November 11, 2024 at 8:11 PM
danchaltiel.bsky.social
I'm interested too! Not sure if all true, but ChatGPT gives some nice insight given that it is not really documented:
ChatGPT answer to the question: In R, the rlang package provides tools for managing errors and debugging. Specifically, last_trace() and last_error() give insights into the stack frames at the time of an error, but they simplify this view by hiding certain frames. The goal is to keep only frames that are relevant to user code, making it easier to understand the source of an error. Here's how rlang decides which frames to hide: Internal and Helper Functions: Frames related to internal functions of rlang, or helper functions that aren't part of the main user logic, are hidden to avoid cluttering the traceback. This includes intermediate steps or utility functions that are not directly related to the error cause. Known S3 and S4 Methods: Some S3 and S4 methods are internal mechanisms to extend behavior in R and can be ignored when they aren’t directly contributing to the error source. Control Flow Wrappers: Frames involving control flow wrappers (like tryCatch(), withCallingHandlers(), etc.) are also usually hidden. These frames are typically there to handle exceptions or modify error propagation behavior, but they’re often irrelevant to the logic that led to the error. Option Customization: The behavior of rlang can sometimes be customized. For instance, using options(rlang__trace_top_env = some_environment) can limit the stack trace to frames in a specific environment (such as a package namespace), making last_trace() and last_error() more focused. Visibility of User Frames: By default, rlang attempts to show the frames that contain user-defined functions and the main execution flow. The hidden frames generally involve functions lower down in the call stack that are part of standard library processes or error handling infrastructure. In practice, this means that last_trace() will only display frames directly tied to the user code that led to the error.
November 4, 2024 at 7:20 PM