alanocallaghan

A catalogue of the bugs that have caused me not insignificant amounts of pain

R

• Dropping attributes

  matrix[, 1, drop=FALSE]

• Autocompletion for thee but not for me

moodymudskipper

`for` <- function(it, seq, expr) {
  if(!identical(substitute(it), quote(.))) {
    tryCatch(
      eval.parent(substitute(base::`for`(it, seq, expr))),
      error = function(e) {stop(c("In `for`: ", e$message), call. = FALSE)})
    return(invisible())
  }
  l <- length(seq[[1]])
  nms1 <- names(seq)
  nms2 <- paste0("*", names(seq), "*")

jimhester

Q1. Are all the latest release versions of tidyverse and tidymodels packages already using cpp11?

All tidyverse packages that used Rcpp have been converted except for readxl and lubridate. We plan to finish converting the remainder of the packages in the next 6 months.

Q2. C++20 brings a lot of exciting features to the language, do you plan to also add a {cpp20} package too?

I agree! It is possible, but it would not be for at least 5 to 7 years from now, if ever. Compiler support and availablity often lags the language specifications significantly. However you can definitely use C++20 features with cpp11 in your package today if you like!

Q3. So writable:: is like not using const& in Rcpp?

moodymudskipper

options(typed.typed = TRUE)

`%typed%` <- function(e1, e2) {
  if(getOption("typed.typed")) {
    assert_call <- substitute(e1)
    assert_call[[1]] <- quote(assertthat::assert_that)
    eval.parent(assert_call)
  }
  e2
}

thoolihan

data <- -10:10
lims <- range(data)
x <- sample(data, 50, replace = TRUE)

par(mfrow=c(1, 5))

plot(x, ylim=lims, main="initial data")

# only positive
plot(pmax(x, 0), ylim=lims, main="clip bottom")

mfansler

#!/usr/bin/env awk -f
#' Author: Mervin Fansler
#' GitHub: @mfansler
#' License: MIT
#' 
#' Basic usage
#' $ conda list --export | awk -f list_export_to_yaml.awk
#' 
#' Omitting builds with 'no_builds'
#' $ conda list --export | awk -v no_builds=1 -f list_export_to_yaml.awk

moodymudskipper

create_lazy_bindings <- function(x, ...){
  environment()
}

# object df is in the environment but has not been evaluated
e <- create_lazy_bindings(y)
ls(e)
#> [1] "x"

# thus this won't work

dmi3kno

is.unset<-function(x)
  x==as.list(alist(a=))[[1]]

is.nameempty <- function(x){
  nx <- names(x)
  if(is.null(nx))
    return(rep(TRUE, length(x)))
  is.na(nx) | nx==""
}

akhanf

#!/bin/bash

hours=6
mem=128000
gpus=1
cpus=32

if [ "$#" -lt 2 ]
then
  echo "Usage $0 RULE_NAME NUM_BATCHES <optional snakemake args>"

ha0ye

calc3 <- function(sets)
{
  sets <- check_sets(sets)
  set_lengths <- vapply(sets, length, 0)
  set_order <- order(set_lengths)
  sets <- sets[set_order]
  set_lengths <- set_lengths[set_order]
  
  n_sets <- length(sets)
  set_names <- names(sets)