Misc

Windows

Shortcut Description

Ctrl + Tab Change application

Ctrl + ~ Change window within an application

Shortcut	Description
Ctrl + Tab	Change application
Ctrl + ~	Change window within an application

Browser

Action	Shortcut
To Address Bar	Ctrl + L
Scroll between focus panes (e.g. address to bar to body)	f6
Open a new window	Ctrl + n
Open a new window in Incognito mode	Ctrl + Shift + n
Open a new tab, and jump to it	Ctrl + t
Reopen previously closed tabs in the order they were closed	Ctrl + Shift + t
Jump to the next open tab	Ctrl + Tab or Ctrl + PgDn
Jump to the previous open tab	Ctrl + Shift + Tab or Ctrl + PgUp
Jump to a specific tab	Ctrl + 1 through Ctrl + 8
Jump to the rightmost tab	Ctrl + 9
Open your home page in the current tab	Alt + Home
Open the previous page from your browsing history in the current tab	Alt + Left arrow
Open the next page from your browsing history in the current tab	Alt + Right arrow
Close the current tab	Ctrl + w or Ctrl + F4
Close the current window	Ctrl + Shift + w or Alt + F4
Minimize the current window	Alt + Space then n
Maximize the current window	Alt + Space then x
Quit Google Chrome	Alt + f then x
Move tabs right or left	Ctrl + Shift + PgUp or Ctrl + Shift + PgDn

R

Search for R packages
- List of CRAN Task Views
- Journal of Statistical Software Search
- R-Universe Search
- {packagefinder} - For searching for packages on CRAN
Changelog for RStudio versions
Detach a package
```
detach("package:vegan", unload = TRUE)
```
Import functions from package
```
# R 4.5.0
use("dplyr", c("filter", "select"))
```
- I think this functionality is also available in R4.0+ with library(pkg, include.only = c("fun1", "fun2")

Reading in zipped files (source)

# unz() lets you read one file from a .zip
one_zipped_csv_among_others <- readr::read_csv(
  unz("lotsa_zipped_csvs.zip", "one_csv.csv")
)

# .zip with a single CSV
one_zipped_csv <- readr::read_csv("big_zipped_file.zip")

R-devel (>= 4.4.0) gained a command-line option to adjust the connections limit (previous limit was 128 parallel workers)

$ R
> parallelly::availableConnections()
[1] 128

$ R --max-connections=512
> parallelly::availableConnections()
[1] 512

Get installed package names, versions, environment, source (article)

> package_names <- installed.packages()[, 1]
> package_data <- lapply(
+     package_names,
+     utils::packageDescription,
+     fields = c("Package", "Version", "Built", "Repository")
+ ) |>
+     lapply(as, Class = "list") |>
+     lapply(setNames, c("Package", "Version", "Environment", "Source")) |>
+     dplyr::bind_rows()

Environment has OS info, R version, and date installed.

Java Dependent Packages
- {rJavaEnv} - Aims to assist users of all {rJava}-dependent packages by providing functions to quickly install the required Java version and set environment variables.
  - It downloads non-installer archives of Java, extracts them to a cache folder, and links them in the current project or working directory. This way, rJavaEnv does not contaminate the user’s machine with unnecessary installations and configurations.
  - When downloading a Java version, it tells you where it’s stored.
```
Java 21 (amazon-corretto-21-x64-windows-jdk.zip) for windows x64 installed at
    C:\Users\ercbk\AppData\Local/R/cache/R/rJavaEnv/installed/windows/x64/21 and symlinked to
    C:/Users/ercbk/Documents/rjavaenv/windows/x64/21
```
  - Code for the path to the installation is then added to your .Rprofile.
  - “if you are on a Linux system that did not have any ‘Java’ version previoulsy installed and you are not using pre-build ‘R’ package binaries (for example from ‘Posit Package Manager’) and instead install all ‘R’ packages from source, after this step you may have to quit R, follow these steps: https://solutions.posit.co/envs-pkgs/using-rjava/#reconfigure-r to set configure ‘Java’ for ‘R’, and only then install ‘rJava’, as otherwise ‘rJava’ cannot be built from source.”

Hackathon Criteria

Update R

Misc
- https://www.r-project.org/
- RStudio NEW.md
- {rig} - r version management system
- update.packages(checkBuilt = TRUE, ask = FALSE) is supposed to search for packages in other R versions and update them in the new R version, but I haven’t tried it, yet.
- Install the newest stable version to check it out
```
rig add next
R-next
```
  - R-<ver> runs a R version without it being the default
- Errors when compiling from source may require installing libraries and they’ll supply code to install via “pacman”
  - Open Start >> scroll down to RTools40 >> RTools Bash
  - Paste pacman code and hit enter to install
- Problem packages in the past
  - {brms} dependency, {igraph}, didn’t have a binary on CRAN and wouldn’t compile from source even with correct libraries installed.
    - Sol’n: install.packages("igraph", repos = 'https://igraph.r-universe.dev')
      - installs dev version from r-universe
  - Some {easystats} packages had gave {pak} some problems. No difficulties using install.packages with default repo or if they had a r-universe repo though.
Steps
1. Copy user installed packages in current R version
  - In R:
```
squirrel <- names(installed.packages(priority = "NA")[, 1]) # user installed packages
readr::write_rds(squirrel, "packages.rds")
```
    - Then, close RStudio
2. Check/Update rig version
  - In powershell: rig --version
  - Check current rig release: link
  - Download and install if your version isn’t current
    - Windows:
      - winget install posit.rig
      - Executable (See Assets section)
3. Install new version of R
  - Close R if not already closed
  - rig add release installs the latest version of R.
  - rig default <new_r_version> sets that version as the default
    - e.g. rig default 4.4.2
4. RTools: Check to see if you have the latest because you’ll need it to compile some of newest versions of packages.
  - Your rtools folder has the version in it’s folder name.
  - rtools website has the latest version and an .exe to download
  - rig add rtools43 will install the RTools for R 4.3 or you can specify any version
  - rig add rtools will install all RTools versions for all R versions that are installed.
5. Add R and RTools to path
  - Right-click Windows >> System >> (right panel) Advanced System Settings >> Environment Variables >> Under User Variables, highlight Path, click Edit >> Click Add
    - R: Add path to directory with all the RScript, R exe, etc. e.g. “C:\Program Files\R\R-4.2.3\bin\x64”
    - RTools: e.g. “C:\rtools43\usr\bin”
6. Open R and confirm new version
  - If RStudio
    - The setting of the new version to the “default” version of R in rig should result in RStudio loading the new version if you have automatic detection enabled.
    - If not, Tools >> Global Options >> General
      - Under “R version”, click “change” button; choose new R version
      - Quit session and restart RStudio
7. Install “high maintenance” packages
  - I’ve had issues with {pak} installing packages that need to be compiled. Maybe be worth trying {pak} first to see if they’ve fixed it.
  - {cmdstanr} doesn’t live on CRAN, so you have to use: install.packages("cmdstanr", repos = c("https://mc-stan.org/r-packages/", getOption("repos")))
    - Check for latest cmdstan version
      - After loading the package, library(cmdstanr) , it should run a check on your cmdstan version and tell you if there’s a newer version.
      - To update, first check toolchain: check_cmdstan_toolchain()
        
        Might tell you to update RTools or that you need some C++ library added
      - Fix C++ toolchain with check_cmdstan_toolchain(fix = TRUE)
      - Update cmdstan: install_cmdstan()
      - May need to install {rstudioapi} and run rstudioapi::restartSession() (programmatically) or just ctrl + shift + f10 so that this package can be used as a dependency for other packages that need to be installed.
  - {rstanarm}: install.packages("rstanarm")
8. Install other packages
```
moose <- readRDS("packages.rds")
moose <- moose[!moose %in% c("cmdstanr", "rstanarm", "ebtools", "translations", "<RStudio add-ins>", "<other pkgs not on CRAN>")]

# Next time, add a try/catch? or maybe purrr::safely, so that it continues through errors. Also, need to log pkgs that do error.
for (i in seq_len(length(moose))) {
  print(moose[i])
  pak::pkg_install(moose[i], ask = FALSE)
}

fs::file_delete("packages.rds")
```
  - {ebtools} is my personal helper package.
  - {translations} is a system package that shouldn’t have been included when I saved the packages from previous version, but was when I recently updated. Might not be necessary to include it in the excuded packages in the future.
  - If there’s an error in downloading a package, continue on to the next package
```
# ex: installing package "mixedup" errored
elt_num <- which(moose == "mixedup")
squirrel <- moose[elt_num:length(moose)]

# carry on
for (i in seq_len(length(squirrel))) {
  print(squirrel[i])
  pak::pkg_install(squirrel[i], ask = FALSE)
}

fs::file_delete("packages.rds")
```
9. Check for updates of RStudio (link)
  - Current version under Help >> About Rstudio
  - Possible to check for updates under Help >> Check for Updates, but that’s failed me before.

Update Python

Linux

Misc
- Each Ubuntu (and probably other distros) version has a python version already installed. To install a newer but not-the-latest stable version, you have to use the PPA: deadsnakes.
  - Don’t set the newer version as default. Just include the version in the command, e.g. python3.12
    - The pre-installed version of Python has important modules installed that doesn’t come with the deadsnakes versions.
    - Even if you do this at the regular user level and not at root, it can still affect scripts (e.g. on PATH) that expect the system-level python version and its modules.
  - Or set the new python version as an alias in your bashrc.
```
# add to ~/.bashrc
alias python='python3.12' alias python3='python3.12'
```
- Always a good idea to hang back from the latest version of Python, because of bugs but also because packages like TensorFlow and PyTorch don’t usually support the latest stable version for awhile.

Issues

GPG (GNU Privacy Guard) Key Issue

Code

sudo apt update 
#> Err:4 https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy InRelease
#>   The following signatures couldn't be verified because the public key is not available: NO_PUBKEY BA6932366A755776
#> Get:5 http://archive.ubuntu.com/ubuntu jammy-backports InRelease [127 kB]
#> Reading package lists... Done
#> W: GPG error: https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY BA6932366A755776
#> E: The repository 'https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy InRelease' is not signed.
#> N: Updating from such a repository can't be done securely, and is therefore disabled by default.
#> N: See apt-secure(8) manpage for repository creation and user configuration details.


############## old way
# Import the missing GPG key
sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys BA6932366A755776

# Alternative method if the above doesn't work:
wget -qO- https://keyserver.ubuntu.com/pks/lookup?op=get&search=0xBA6932366A755776 | sudo apt-key add -

# Now try updating again
sudo apt update

############## modern way
# Remove the old PPA first
sudo add-apt-repository --remove ppa:deadsnakes/ppa

# Add the key to the new keyring location
wget -qO- https://keyserver.ubuntu.com/pks/lookup?op=get&search=0xBA6932366A755776 | sudo gpg --dearmor -o /usr/share/keyrings/deadsnakes.gpg

# Add the repository with the specific keyring
echo "deb [signed-by=/usr/share/keyrings/deadsnakes.gpg] https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy main" | sudo tee /etc/apt/sources.list.d/deadsnakes.list

# Update package list
sudo apt update

I added the deadsnakes PPA (Personal Package Archive) repository to apt and during the add, it got the error, ssl.SSLEOFError: EOF occurred in violation of protocol (_ssl.c:2426)
I thought updating and upgrading might help, but I got this error.
Claude said the key wasn’t imported correctly. The “old way” worked, but it did give me this warning, apt-key is deprecated. Manage keyring files in trusted.gpg.d instead (see apt-key(8)).
- I think this refers to the “modern way” which I also included.

Updating Python
```
sudo apt update
sudo apt upgrade

# Install software-properties-common if not already installed
sudo apt install software-properties-common

sudo add-apt-repository ppa:deadsnakes/ppa

# check that the new repository was added
ls -l /etc/apt/sources.list.d/

sudo apt update
sudo apt install python3.12 -y
```
- software-properties-common is a package that has help functions like add-apt-repository, ppa stuff, gpa key stuff, etc. (It was already came installed with Ubuntu)
- The PPA deadsnakes is a repo that has older-than-the-latest python versions for Ubuntu versions.
- The -y flag just says the answer is “Yes” to any prompts
Additional Modules
- The PPA deadsnakes’ versions seem to be pretty barebones, but the repo also makes available several modules
- Absolutely want venv and Claude suggested a few other system-level packages
```
# Install additional packages you'll likely need
sudo apt install python3.12-venv python3.12-dev python3.12-full
```
  - dev - Contains Python header files needed to compile C extension
  - full - Seems like stuff related to some IDE. Might be one you can skip