In friedav/fdhelpers: Collection of FD's personal helper functions

Assumptions

required columns:
Author name
affiliation 1, affiliation 2, ...
rows in input tsv are sorted by desired order of authors on manuscript
in output, affiliation indices per author are sorted numerically

Processing steps

remove all superfluous whitespaces in author names and affiliations
exclude affiliation "C" (-> error in input?)
format any listed name initials as X.
consistently use USA (not US) as country code in affiliations
consistently use UK (not GB) as country code in affiliations

knitr::opts_chunk$set(echo = F)

library(tidyverse)
library(here)
library(data.table)

dat <- fread(params$input, encoding = 'UTF-8') %>% as_tibble() %>%
  set_names(make.names(colnames(.))) %>%
  select(author = Author.name, contains("affiliation.")) %>%
  mutate_all(~ str_squish(.x)) %>%
  mutate(author = str_replace_all(author, "\\b([[:alpha:]])\\b(?!\\.)", "\\1\\.")) %>%
  mutate(author_rank = 1:nrow(.)) %>%
  pivot_longer(cols = contains("affiliation."), 
               names_to = "affil_rank_author",
               values_to = "affil") %>%
  mutate(affil_rank_author = as.integer(str_extract(affil_rank_author, "\\d+"))) %>%
  filter(affil != "") %>%
  filter(affil != "C") %>% 
  mutate(affil = str_replace(affil, "\\bUSA?$", "USA"),
         affil = str_replace(affil, "\\bGB$", "UK")) %>% 
  mutate(affil_rank = as.numeric(factor(affil, levels = unique(affil))))

List of similar affiliations (to look for errors)

# unique affiliations in data
uaffil <- unique(dat$affil)

# fuzzy matching to other listed affiliations (only return imperfect matches)
affilmatch <- lapply(seq_along(uaffil), function(i) {
  agrep(uaffil[i], uaffil[-(1:i)], ignore.case = T, value = T)
}) %>% 
  set_names(uaffil) %>% 
  .[sapply(., function(x) length(x) > 0)]

r sapply(seq_along(affilmatch), function(i) { paste0("- ", names(affilmatch)[i], "\n", paste0(" - ", affilmatch[[i]], "\n", collapse = "")) }) %>% paste0(collapse = "")

Authors

# compose formatted string of authors
authors <- dat %>%
  group_by(author_rank, author) %>%
  arrange(affil_rank) %>% 
  summarize(aff_ranks = paste0(affil_rank, collapse = ","),
            string = paste0(unique(author), "^", aff_ranks, "^")) %>% 
  ungroup() %>%
  summarize(out = paste0(string, collapse = ", ")) %>% .$out

r authors

Affiliations

# compose formatted string of affiliations
affiliations <- dat %>% 
  select(affil, affil_rank) %>% 
  filter(!duplicated(affil)) %>% 
  summarize(out = paste0("^", affil_rank, "^", affil, ".", 
                         collapse = " ")) %>% 
  .$out