194 lines
7.6 KiB
R
194 lines
7.6 KiB
R
# =============================================================================
|
|
# Configuraion file
|
|
# =============================================================================
|
|
# This file contains necessary variables that will be used in the whole process
|
|
# and all necessary library imports.
|
|
# See the configuration section for theming, debug mode and so on.
|
|
# =============================================================================
|
|
|
|
# ---- Helper functions -------------------------------------------------------
|
|
|
|
install_if_missing <- function(pkgs) {
|
|
to_install <- pkgs[!vapply(pkgs, requireNamespace, quietly = TRUE, FUN.VALUE = logical(1))]
|
|
if (length(to_install)) {
|
|
message("Installing missing CRAN packages: ", paste(to_install, collapse = ", "))
|
|
install.packages(to_install, repos = "https://cloud.r-project.org/")
|
|
}
|
|
}
|
|
|
|
install_github_if_missing <- function(pkg, repo) {
|
|
if (!requireNamespace(pkg, quietly = TRUE)) {
|
|
if (!requireNamespace("remotes", quietly = TRUE)) {
|
|
install.packages("remotes", repos = "https://cloud.r-project.org/")
|
|
}
|
|
message("Installing GitHub package ", pkg, " from ", repo)
|
|
remotes::install_github(repo)
|
|
}
|
|
}
|
|
|
|
# =============================================================================
|
|
# Install & Load
|
|
# =============================================================================
|
|
|
|
# Define all required packages with comments
|
|
pkgs <- c(
|
|
# Data handling & plotting
|
|
"tidyverse", # data manipulation, ggplot2, etc.
|
|
"patchwork", # combine multiple plots
|
|
"ggthemes",
|
|
"ggpmisc",
|
|
"scales",
|
|
|
|
# Tables & summaries
|
|
"summarytools",
|
|
"gtsummary",
|
|
"gt",
|
|
"knitr",
|
|
"kableExtra",
|
|
"flextable", # docx tables
|
|
|
|
# Misc helpers
|
|
"rlang",
|
|
"rmarkdown",
|
|
"qs2", #
|
|
"presize", # sample size calculations
|
|
"survey"
|
|
)
|
|
|
|
|
|
# Install any that aren't already present
|
|
missing <- pkgs[!pkgs %in% installed.packages()[, "Package"]]
|
|
if (length(missing)) {
|
|
install.packages(missing, repos = "https://cloud.r-project.org/")
|
|
}
|
|
|
|
# Install and load CRAN packages
|
|
install_if_missing(pkgs)
|
|
invisible(lapply(pkgs, library, character.only = TRUE))
|
|
|
|
# ---- GitHub packages ---------------------------------------------------
|
|
|
|
# Serosurvey
|
|
install_github_if_missing("serosurvey", "avallecam/serosurvey")
|
|
library(serosurvey)
|
|
|
|
# ggthemr
|
|
install_github_if_missing("ggthemr", "Mikata-Project/ggthemr")
|
|
library(ggthemr)
|
|
|
|
# =============================================================================
|
|
# Configuration
|
|
# =============================================================================
|
|
|
|
# ---- Output format ----------------------------------------------------------
|
|
# it is important to set the correct pandoc/quarto output format as knitr tables don't work in docx.
|
|
# possible formats:
|
|
# - docx
|
|
# - pdf/tex
|
|
|
|
# Read OUTPUT_FORMAT from env; default to "pdf/tex"
|
|
output_format <- Sys.getenv("OUTPUT_FORMAT", "pdf/tex")
|
|
# Background: some tables needs special treatment for docx rendering.
|
|
|
|
# ---- Debug Mode -------------------------------------------------------------
|
|
debug_mode <- TRUE
|
|
if (isTRUE(debug_mode)) debug_info <- list()
|
|
|
|
# ---- Theme & Colors ---------------------------------------------------------
|
|
# Theme
|
|
ggthemr('fresh')
|
|
|
|
# consitent colors for open science practices among plots
|
|
osp_cols <- c(
|
|
"Preregistration" = "#ee5927",
|
|
"Open Data" = "#321c3d",
|
|
"Open Materials" = "#005c5c",
|
|
"Open Access" = "#bf1869",
|
|
"Statistical Inference" = "#f2a900"
|
|
)
|
|
|
|
name_mapping <- c(
|
|
"Preregistration" = "is_prereg",
|
|
"Open Data" = "is_open_data",
|
|
"Open Materials" = "is_open_materials",
|
|
"Open Access" = "is_open_access",
|
|
"Statistical Inference" = "is_statistical_inference"
|
|
)
|
|
|
|
# map names & colors
|
|
osp_cols2 <- osp_cols
|
|
names(osp_cols2) <- name_mapping[names(osp_cols)]
|
|
|
|
# =============================================================================
|
|
# DIRECTORY PATHS
|
|
# =============================================================================
|
|
|
|
# Base directories
|
|
dir_data <- "data"
|
|
dir_output <- "output"
|
|
dir_logs <- "logs"
|
|
dir_models <- "models"
|
|
dir_code <- "code"
|
|
dir_docs <- "docs"
|
|
dir_renv <- "renv"
|
|
dir_freeze <- "_freeze"
|
|
dir_output_quarto <- "_output" # Quarto's rendered output
|
|
|
|
# Data subdirectories
|
|
dir_data_meta <- file.path(dir_data, "meta")
|
|
dir_data_crossref <- file.path(dir_data_meta, "crossref")
|
|
dir_data_fulltext <- file.path(dir_data, "fulltext")
|
|
dir_data_keywords <- file.path(dir_data, "keyword_dicts")
|
|
|
|
# Output subdirectories
|
|
dir_output_plots <- file.path(dir_output, "plots")
|
|
dir_output_tables <- file.path(dir_output, "tables")
|
|
|
|
# =============================================================================
|
|
# Files
|
|
# =============================================================================
|
|
|
|
# Primary data files
|
|
file_journals <- file.path(dir_data_meta, "journals.qs2")
|
|
file_meta_combined <- file.path(dir_data_meta, "meta_combined.qs2")
|
|
file_meta_final <- file.path(dir_data_meta, "meta_final.qs2")
|
|
file_sample_final <- file.path(dir_data_meta, "sample_final.qs2")
|
|
file_sample_stat_final <- file.path(dir_data_meta, "sample_stat_final.qs2") # statistical inference classification sample
|
|
file_sample_analysis <- file.path(dir_data, "sample_analysis.qs2") # final sample, ready for analysis
|
|
|
|
# Additional data files (wasn't used yet)
|
|
file_dicts <- file.path(dir_data, "dicts.csv")
|
|
|
|
# Statistical inference sample files
|
|
file_train_stat <- file.path(dir_data, "train_stat.csv") # statistical inference classification training sample
|
|
file_train_stat_coded <- file.path(dir_data, "train_stat_coded.csv") # statistical inference classification hand coded training sample
|
|
file_train_stat_downloader_meta <- file.path(dir_data, "train_stat_downloader_meta_export.csv") # statistical inference downloader metadata (containing file_path)
|
|
|
|
file_train_stat_coded_hand <- file.path(dir_data, "train_stat_coded_hand.csv") # statistical inference classification hand coded
|
|
file_train_stat_coded_hand_corrected <- file.path(dir_data, "train_stat_coded_hand_corrected.csv") # statistical inference classification hand coded
|
|
file_train_stat_coded_hand_corrected_gpt <- file.path(dir_data, "train_stat_coded_hand_corrected.csv") # statistical inference classification hand coded
|
|
|
|
# OSP Classification files
|
|
file_sample <- file.path(dir_data, "full-sample.csv") # full sample of all years
|
|
file_sample_statistical <- file.path(dir_data, "sample_statistical.csv") # full sample of statistical inference papers
|
|
file_train <- file.path(dir_data, "train.csv") # training subset of statistical inference papers for osp classification
|
|
file_train_coded_hand <- file.path(dir_data, "train_coded_hand.csv") # training subset (osp), hand-coded
|
|
file_train_coded_hand_corrected <- file.path(dir_data, "train_coded_hand_corrected.csv") # training subset (osp), hand-coded, corrected
|
|
file_train_coded_hand_corrected_gpt <- file.path(dir_data, "train_coded_hand_corrected_gpt.csv") # training subset (osp), corrected, hand-coded + remaining cases coded with gpt
|
|
file_sample_downloader_meta <- file.path(dir_data, "train_downloader_meta_export.csv") # statistical inference classification hand coded
|
|
|
|
# Classified data
|
|
file_classified_stat <- file.path(dir_data, "classified_stat.csv") # statistical inference
|
|
file_classified_oa <- file.path(dir_data, "classified_oa.csv") # open access
|
|
file_classified_osp <- file.path(dir_data, "classified_osp.csv") # other osp's
|
|
file_classified_osp_probs <- file.path(dir_data, "classified_osp_probs.csv") # other osp's
|
|
|
|
file_sample_fully_classified <- file.path(dir_data, "full-sample_classified.csv") # combination of the above
|
|
|
|
# Downloader import CSV files
|
|
file_downloader_full <- file.path(dir_data, "download_full.csv")
|
|
file_downloader_sample <- file.path(dir_data, "download_sample.csv")
|
|
|
|
# Codebook
|
|
file_codebook <- file.path(dir_output, "codebook.docx")
|