Joel Nitta joelnitta

## compare_project_files.R
# Compare hashes of identically named files across two project versions,
# e.g. if the project was run again with different versions of dependencies.
# >>> Assumes files are uniquely named within a project <<<
# (replace vector of paths appropriately)

library(digest)
library(tidyverse)

files_to_check_01 <- list.files(
  # Check all of the files in these directories

## make_ref_key_list.R
# Extract a list of citation keys from an Rmd file,
# paste them together to search in Mendeley to make
# a list of references in that Rmd file

rmd_file <- "path_to_Rmd_file"

citations <-
  readr::read_lines(rmd_file) %>%
  stringr::str_split(" |;") %>%
  unlist %>%

## gni_kubernetes.yaml
---
apiVersion: v1
kind: Service
metadata:
  name: resolver
  namespace: gn
  labels:
    app: resolver
    tier: frontend
spec:

## check_doi_in_bib.R
library(bib2df)
library(rcrossref)
library(tidyverse)

# Double check to make sure DOIs in bibliography match the right papers

# Load clean bibliography, select just relevant columns
path_to_bib <- "ms/references.bib"

bib <- bib2df(path_to_bib) %>%

## extract_latex_pkg_names_from_log.R
library(tidyverse)

# Find the names of latex packages installed by tlmgr
# (https://www.tug.org/texlive/tlmgr.html)
# The main reason to do this is so they can be pre-installed to a docker image.
#
# The latex package install output from tlmgr tends to look like this:
#
# ```
# tlmgr search --file --global '/multirow.sty'

## extract_raw_data_file_names_from_plan.R
library(tidyverse)

# Extract names of files in a "data_raw" folder
# and append "put" so they can easily be uploaded to
# a server using sftp

plan_lines <- read_lines("code/world_biogeo_plan.R")

plan_lines %>%
  magrittr::extract(str_detect(., "data_raw")) %>%

## write_fasta_files_chunked.R
#' Write out a list of DNA sequences in chunks
#'
#' The list will be split into a list of lists, to
#' save memory when writing out sequences.
#'
#' @param fasta_list The list of DNA sequences
#' @param chunk_size Size of chunks to split up list into.
#' @param out_dir Directory to write to
#'
#' @return List of hashes; digest of each chunk. Externally, the

## i18n_remote_error.log
joelnitta@Joels-iMac repos % git clone https://github.com/joelnitta/i18n
Cloning into 'i18n'...
remote: Enumerating objects: 1672, done.
remote: Total 1672 (delta 0), reused 0 (delta 0), pack-reused 1672
Receiving objects: 100% (1672/1672), 2.73 MiB | 2.43 MiB/s, done.
Resolving deltas: 100% (1167/1167), done.
joelnitta@Joels-iMac repos % cd i18n
joelnitta@Joels-iMac i18n % git submodule init
Submodule 'git-novice' (https://github.com/swcarpentry-i18n/git-novice.git) registered for path 'git-novice'
Submodule 'po4gitbook' (https://github.com/swcarpentry-i18n/po4gitbook.git) registered for path 'po4gitbook'

## combine_huc.R
# Set working directory
setwd(here::here())

# Load packages and functions
source("r-code/01_start_up.R")

# Load plan
source("r-code/plan.R")

# Load cache

## start-dockerd.sh
#!/usr/bin/bash

# Start docker daemon in background
nohup dockerd-rootless.sh --experimental --storage-driver vfs >| dockerd.log 2>&1 &
# Save the process ID so we can kill it later
echo $! > dockerd.pid
	# Compare hashes of identically named files across two project versions,
	# e.g. if the project was run again with different versions of dependencies.
	# >>> Assumes files are uniquely named within a project <<<
	# (replace vector of paths appropriately)

	library(digest)
	library(tidyverse)

	files_to_check_01 <- list.files(
	# Check all of the files in these directories
	# Extract a list of citation keys from an Rmd file,
	# paste them together to search in Mendeley to make
	# a list of references in that Rmd file

	rmd_file <- "path_to_Rmd_file"

	citations <-
	readr::read_lines(rmd_file) %>%
	stringr::str_split(" \|;") %>%
	unlist %>%
	---
	apiVersion: v1
	kind: Service
	metadata:
	name: resolver
	namespace: gn
	labels:
	app: resolver
	tier: frontend
	spec:
	library(bib2df)
	library(rcrossref)
	library(tidyverse)

	# Double check to make sure DOIs in bibliography match the right papers

	# Load clean bibliography, select just relevant columns
	path_to_bib <- "ms/references.bib"

	bib <- bib2df(path_to_bib) %>%
	library(tidyverse)

	# Find the names of latex packages installed by tlmgr
	# (https://www.tug.org/texlive/tlmgr.html)
	# The main reason to do this is so they can be pre-installed to a docker image.
	#
	# The latex package install output from tlmgr tends to look like this:
	#
	# ```
	# tlmgr search --file --global '/multirow.sty'
	library(tidyverse)

	# Extract names of files in a "data_raw" folder
	# and append "put" so they can easily be uploaded to
	# a server using sftp

	plan_lines <- read_lines("code/world_biogeo_plan.R")

	plan_lines %>%
	magrittr::extract(str_detect(., "data_raw")) %>%
	#' Write out a list of DNA sequences in chunks
	#'
	#' The list will be split into a list of lists, to
	#' save memory when writing out sequences.
	#'
	#' @param fasta_list The list of DNA sequences
	#' @param chunk_size Size of chunks to split up list into.
	#' @param out_dir Directory to write to
	#'
	#' @return List of hashes; digest of each chunk. Externally, the
	joelnitta@Joels-iMac repos % git clone https://github.com/joelnitta/i18n
	Cloning into 'i18n'...
	remote: Enumerating objects: 1672, done.
	remote: Total 1672 (delta 0), reused 0 (delta 0), pack-reused 1672
	Receiving objects: 100% (1672/1672), 2.73 MiB \| 2.43 MiB/s, done.
	Resolving deltas: 100% (1167/1167), done.
	joelnitta@Joels-iMac repos % cd i18n
	joelnitta@Joels-iMac i18n % git submodule init
	Submodule 'git-novice' (https://github.com/swcarpentry-i18n/git-novice.git) registered for path 'git-novice'
	Submodule 'po4gitbook' (https://github.com/swcarpentry-i18n/po4gitbook.git) registered for path 'po4gitbook'
	# Set working directory
	setwd(here::here())

	# Load packages and functions
	source("r-code/01_start_up.R")

	# Load plan
	source("r-code/plan.R")

	# Load cache
	#!/usr/bin/bash

	# Start docker daemon in background
	nohup dockerd-rootless.sh --experimental --storage-driver vfs >\| dockerd.log 2>&1 &
	# Save the process ID so we can kill it later
	echo $! > dockerd.pid