From ae542cefe44374c025e0ca556b88176e2d8ff8fa Mon Sep 17 00:00:00 2001
From: andreaspacher <ap@ooir.org>
Date: Thu, 15 Sep 2022 09:22:08 +0200
Subject: [PATCH] Drop files from .gitignore

---
 Script/RnR-combine.R | 522 -------------------------------------------
 Script/ranking.R     | 182 ---------------
 2 files changed, 704 deletions(-)
 delete mode 100644 Script/RnR-combine.R
 delete mode 100644 Script/ranking.R

diff --git a/Script/RnR-combine.R b/Script/RnR-combine.R
deleted file mode 100644
index 9efa09f..0000000
--- a/Script/RnR-combine.R
+++ /dev/null
@@ -1,522 +0,0 @@
-library(tidyverse)
-library(data.table)
-
-ALL <- readxl::read_xlsx("./Data/04_publishers_new-15.xlsx") %>%
-  filter(!is.na(results_manual)) %>%
-  select("publisher" = PUBLISHER_NAME, "n" = results_manual) %>%
-  mutate("comment" = "journals were counted manually")
-
-PP <- readxl::read_xlsx("Output/allpublishers-2022-04-15.xlsx")
-P <- PP %>% group_by(publisher) %>% arrange(desc(n))
-
-FINAL <- rbind(ALL, P) %>%
-  arrange(desc(n))
-
-FINAL <- readxl::read_xlsx("Data\\04_publishers.xlsx")
-
-rm(ALL, PP, P)
-
-# ================
-#
-# Comparison with Scopus & Publons
-# & DOAJ & Sherpa Romeo
-#
-# ================
-SCOPUS <- read_csv("Output\\Preliminary_Lists\\03_publishers_Scopus.csv")
-PUBLONS <- read_csv("Output\\Preliminary_Lists\\02_publishers_Publons.csv")
-DOAJ <- read_csv("Output\\Preliminary_Lists\\01_publishers_DOAJ.csv")
-SHERPA <- read_csv("Output\\Preliminary_Lists\\04_publishers_SherpaRomeo.csv")
-
-# cbind(head(SCOPUS,284), FINAL) %>% View()
-# cbind(head(PUBLONS,284), FINAL) %>% View()
-
-harmonia <- read.delim(file = "Data\\03_publishers_harmonization.txt", header = TRUE, sep = "\t")
-colnames(harmonia) = c("Deviation", "Harmonization")
-
-colnames(SCOPUS) = c("publisher", "Freq")
-colnames(SHERPA) = c("publisher", "Freq")
-colnames(DOAJ) = c("publisher", "Freq")
-colnames(PUBLONS) = c("publisher", "Freq", "reviews")
-PUBLONS <- PUBLONS %>% select(-reviews)
-
-for (i in 1:nrow(harmonia)) {
-  
-  SCOPUS$publisher[SCOPUS$publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-  PUBLONS$publisher[PUBLONS$publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-  SHERPA$publisher[SHERPA$publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-  DOAJ$publisher[DOAJ$publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-  
-}
-
-allsources <- list(DOAJ, PUBLONS, SCOPUS, SHERPA)
-
-allsources <- purrr::map(allsources, function(x) {
-  DT <- data.table(x)
-  DT <- DT[, lapply(.SD, sum, na.rm = T), by = publisher]
-  DT <- DT[order(-Freq)]
-  DT <- as.data.frame(DT)
-  DT <- DT[DT$Freq >= 15,]
-  #Encoding(DT$publisher) <- 'latin1'
-  #DT$publisher <- stringi::stri_trans_general(DT$publisher, 'Latin-ASCII')
-  #colnames(DT) <- c("publisher", deparse(substitute(x)))
-})
-
-colnames(allsources[[1]]) <- c("publisher", "doaj")
-colnames(allsources[[2]]) <- c("publisher", "publons")
-colnames(allsources[[3]]) <- c("publisher", "scopus")
-colnames(allsources[[4]]) <- c("publisher", "sherpa")
-
-FINAL <- readxl::read_xlsx("Output\\top-100-publishers.xlsx")
-colnames(FINAL) = c("publisher", "nishikawa", "predatory", "globalsouth")
-
-ALL <- full_join(FINAL, allsources[[1]]) %>%
-  full_join(allsources[[2]]) %>%
-  full_join(allsources[[3]]) %>%
-  full_join(allsources[[4]]) %>%
-  arrange(desc(nishikawa)) %>%
-  select(publisher, nishikawa, scopus, publons, doaj, sherpa)
-
-ALL$really_not_in_scopus <- 0
-ALL$really_not_in_publons <- 0
-ALL$really_not_in_doaj <- 0
-ALL$really_not_in_sherpa <- 0
-
-for (i in 1:nrow(harmonia)) {
-  
-  ALL$publisher[ALL$publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-
-}
-
-ALL <- ALL %>%
-  # mutate(publisher = ifelse(publisher == "Wolters Kluwer",
-  #                           "Lippincott, Williams & Wilkins",
-  #                           publisher)) %>%
-  group_by(publisher) %>%
-  top_n(1, nishikawa) %>%
-  distinct()
-
-ALL <- ALL %>%
-  mutate(publisher = ifelse(publisher == "Universidade de Bras<ed>lia",
-                            "Universidade de Brasilia", publisher)) %>%
-  group_by(publisher) %>%
-  top_n(1, nishikawa) %>%
-  distinct()
-
-
-#write_csv(ALL, "Output\\RnR-examine.csv")
-#writexl::write_xlsx(ALL, "Output\\RnR-final.xlsx")
-
-# Summary Statistics
-Mode <- function(x) {
-  ux <- unique(x)
-  ux[which.max(tabulate(match(x, ux)))]
-}
-ALL %>%
-  head(100) %>%
-  mutate(n = nishikawa) %>%
-  ungroup() %>%
-  #filter(!is.na(n)) %>%
-  summarise(total = sum(n, na.rm = TRUE),
-            count = n(),
-            avg = mean(n, na.rm = TRUE),
-            med = median(n, na.rm = TRUE),
-            mode = Mode(n),
-            sd = sd(n, na.rm = TRUE),
-            min = min(n, na.rm = TRUE),
-            q1 = quantile(n, probs = 0.25, na.rm = TRUE),
-            q3 = quantile(n, probs = 0.75, na.rm = TRUE),
-            max = max(n, na.rm = TRUE))
-
-# =============
-# predatoriness?
-# =============
-ALL$predatory <- 0
-
-ALL <- ALL %>%
-  mutate(predatory = case_when(
-    publisher == "OMICS" ~ 1,
-    publisher == "Science Publishing Group" ~ 1,
-    publisher == "SCIRP" ~ 1,
-    publisher == "Open Access Text (OAT)" ~ 1,
-    publisher == "Bentham" ~ 1, # ???
-    publisher == "Austin Publishing Group" ~ 1,
-    publisher == "Longdom" ~ 1,
-    publisher == "Open Access Pub" ~ 1,
-    publisher == "Gavin Publishers" ~ 1,
-    publisher == "iMedPub" ~ 1,
-    publisher == "Scientific and Academic Publishing" ~ 1,
-    publisher == "JSciMedCentral" ~ 1,
-    #publisher == "Frontiers" ~ 1, # ???
-    publisher == "Hans Publishers" ~ 1, # ???
-    publisher == "Advanced Research Publications" ~ 1,
-    publisher == "Hilaris" ~ 1,
-    publisher == "Academic Journals" ~ 1,
-    publisher == "Science and Education Publishing" ~ 1,
-    #publisher == "Conscientia Beam" ~ 1, # not in Beall's List
-    # Conscientia Beam:
-    # promises quick publication (1st decision within 20-30 days,
-    # acceptance in 50-60 days); strangely, the field phone nr
-    # is required in manuscript submission system; one can also
-    # add WhatsApp nr; many journals actually do not require
-    # APCs, and four journals seem to be in Scopus
-    # but not in DOAJ
-    publisher == "Sciencedomain International" ~ 1,
-    publisher == "Peertechz Publications" ~ 1,
-    publisher == "Medcrave" ~ 1,
-    publisher == "SciTechnol" ~ 1,
-    publisher == "IOS Press" ~ 1, # ???
-    publisher == "Internet Scientific Publications" ~ 1,
-    publisher == "International Scholars Journals" ~ 1,
-    publisher == "Annex Publishers" ~ 1, # not in Beall's List
-    # Annex Publishers: refers to "CiteFactor" as a reference,
-    # promises rapid peer review (21 days), visibility on
-    # Google Scholar (inter alia), and a publication within
-    # 24h after acceptance; not in DOAJ; the APCs are quite
-    # high (all between $1200 to $3600)!
-    publisher == "Open Access Journals" ~ 1,
-    publisher == "Herbert Publications" ~ 1,
-    publisher == "Medwin Publishers LLC" ~ 1,
-    publisher == "Premier Publishers" ~ 1,
-    publisher == "Pulsus Group" ~ 1,
-    publisher == "Scholarena" ~ 1,
-    TRUE ~ 0
-  ))
-
-head(ALL, 100) %>%
-  group_by(predatory) %>%
-  summarise(n = n(),
-            share = paste0(round(n/100*100, 1), "%"))
-
-#writexl::write_xlsx(ALL, "Output\\RnR-final.xlsx")
-
-# =============
-# Global Southiness?
-# =============
-ALL$globalsouth <- 0
-
-ALL <- ALL %>%
-  mutate(globalsouth = case_when(
-    publisher == "Universitas Pendidikan Indonesia" ~ 1,
-    publisher == "Universidad de Buenos Aires" ~ 1,
-    publisher == "Universidad Nacional Autonoma de Mexico" ~ 1,
-    publisher == "Universitas Gadjah Mada" ~ 1,
-    publisher == "Universitas Negeri Semarang" ~ 1,
-    publisher == "University of Tehran" ~ 1,
-    publisher == "Universidad Nacional de La Plata" ~ 1,
-    publisher == "Universitas Airlangga" ~ 1,
-    publisher == "University of Malaya" ~ 1,
-    publisher == "Universitas Negeri Yogyakarta" ~ 1,
-    publisher == "Universidade Federal do Espirito Santo" ~ 1,
-    publisher == "Universidad Nacional de Cordoba" ~ 1,
-    publisher == "Universitas Negeri Surabaya" ~ 1,
-    publisher == "Universidade Federal do Rio Grande do Sul" ~ 1,
-    publisher == "Universitas Diponegoro" ~ 1,
-    publisher == "Universidade de Brasilia" ~ 1,
-    publisher == "Pontificia Universidad Javeriana, Bogota" ~ 1,
-    publisher == "Universidade de Bras<ed>lia" ~ 1,
-    TRUE ~ 0
-  ))
-
-head(ALL, 100) %>%
-  group_by(globalsouth) %>%
-  summarise(n = n(),
-            share = paste0(round(n/100*100, 1), "%"))
-# =============
-# manual examinations
-# =============
-
-SCOPUS %>%
-  filter(grepl("Austin", publisher))
-SCOPUS %>%
-  filter(grepl("Open Access Pub", publisher))
-SCOPUS %>%
-  filter(grepl("Longdom", publisher))
-SCOPUS %>%
-  filter(grepl("Gavin", publisher))
-SCOPUS %>%
-  filter(grepl("Scientific and Ac", publisher))
-SCOPUS %>%
-  filter(grepl("Advanced Res", publisher))
-
-PUBLONS %>%
-  filter(grepl("Open Access", publisher))
-PUBLONS %>%
-  filter(grepl("Negeri", publisher)) %>% View()
-PUBLONS %>%
-  filter(grepl("Pleiad", publisher))
-PUBLONS %>%
-  filter(grepl("B.g.ta|P.ntific", publisher))
-PUBLONS %>%
-  filter(grepl("Mulino", publisher))
-
-DOAJ %>%
-  filter(grepl("OMICS", publisher))
-DOAJ %>%
-  filter(grepl("Open Access Te", publisher))
-DOAJ %>%
-  filter(grepl("Longdom", publisher))
-DOAJ %>%
-  filter(grepl("Gavin", publisher))
-DOAJ %>%
-  filter(grepl("Scientific (and|&) Ac", publisher))
-# eScholarship, KeAi, Sao Paolo, Intellect
-SHERPA %>%
-  filter(grepl("OMICS", publisher))
-SHERPA %>%
-  filter(grepl("INDER|Inder", publisher))
-SHERPA %>%
-  filter(grepl("Science Pub", publisher))
-SHERPA %>%
-  filter(grepl("Philosophy D", publisher))
-SHERPA %>%
-  filter(grepl("SCIRP", publisher))
-SHERPA %>%
-  filter(grepl("IRMA|Information Res", publisher))
-SHERPA %>%
-  filter(grepl("IGI|I G I", publisher))
-SHERPA %>%
-  filter(grepl("Open Access T|OAT", publisher))
-SHERPA %>%
-  filter(grepl("Austin", publisher))
-SHERPA %>%
-  filter(grepl("Bentham", publisher))
-SHERPA %>%
-  filter(grepl("Open Access Pu", publisher))
-SHERPA %>%
-  filter(grepl("Longdom", publisher))
-SHERPA %>%
-  filter(grepl("Gavin", publisher))
-SHERPA %>%
-  filter(grepl("Schweizerb", publisher))
-SHERPA %>%
-  filter(grepl("Fabriz", publisher))
-
-# ================
-# calculate probability
-# ================
-TOP <- head(ALL, 100)
-
-TOP <- readxl::read_xlsx("Output\\top-100-publishers.xlsx") %>%
-  head(100) %>%
-  mutate(publisher = case_when(
-    publisher == "John Hopkins University Press" ~ "Johns Hopkins University Press",
-    TRUE ~ publisher
-  ))
-
-TOP %>% summarise(sum = sum(journals))
-
-for (i in 1:nrow(harmonia)) {
-  
-  TOP$publisher[TOP$publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-  
-}
-
-
-DOAJ2 <- DOAJ %>%
-  mutate(is_in_top_list = ifelse(publisher %in% TOP$publisher, 1, 0)) %>%
-  group_by(publisher) %>%
-  top_n(1, Freq) %>%
-  distinct() %>%
-  ungroup()
-PUBLONS2 <- PUBLONS %>%
-  mutate(is_in_top_list = ifelse(publisher %in% TOP$publisher, 1, 0)) %>%
-  group_by(publisher) %>%
-  top_n(1, Freq) %>%
-  distinct() %>%
-  ungroup()
-SCOPUS2 <- SCOPUS %>%
-  mutate(is_in_top_list = ifelse(publisher %in% TOP$publisher, 1, 0)) %>%
-  group_by(publisher) %>%
-  top_n(1, Freq) %>%
-  distinct() %>%
-  ungroup()
-SHERPA2 <- SHERPA %>%
-  mutate(is_in_top_list = ifelse(publisher %in% TOP$publisher, 1, 0)) %>%
-  group_by(publisher) %>%
-  top_n(1, Freq) %>%
-  distinct() %>%
-  ungroup()
-
-DOAJ3 <- DOAJ2 %>%
-  filter(is_in_top_list == 1) %>%
-  select(publisher, "doaj" = Freq)
-PUBLONS3 <- PUBLONS2 %>%
-  filter(is_in_top_list == 1) %>%
-  select(publisher, "publons" = Freq)
-SCOPUS3 <- SCOPUS2 %>%
-  filter(is_in_top_list == 1) %>%
-  select(publisher, "scopus" = Freq)
-SHERPA3 <- SHERPA2 %>%
-  filter(is_in_top_list == 1) %>%
-  select(publisher, "sherpa" = Freq)
-
-MADE.IT <- full_join(DOAJ3, PUBLONS3) %>%
-  full_join(SCOPUS3) %>%
-  full_join(SHERPA3) %>%
-  add_row(publisher = "Universidade Federal do Espirito Santo",
-          doaj = NA,
-          publons = NA,
-          scopus = NA,
-          sherpa = 18) %>%
-  add_row(publisher = "Universidade de Brasilia",
-          doaj = 26,
-          publons = NA,
-          scopus = NA,
-          sherpa = 31) %>%
-  add_row(publisher = "Intellect Books",
-          doaj = NA,
-          publons = 17,
-          scopus = 38,
-          sherpa = 111) %>%
-  add_row(publisher = "Universidad Nacional de Cordoba",
-          doaj = 19,
-          publons = NA,
-          scopus = NA,
-          sherpa = 54) %>% 
-  add_row(publisher = "International Scholars Journals",
-          doaj = NA,
-          publons = 15,
-          scopus = NA,
-          sherpa = NA) %>%
-  add_row(publisher = "Longdom",
-          doaj = NA,
-          publons = 15,
-          scopus = NA,
-          sherpa = NA) %>%
-  rowwise() %>% 
-  mutate(max = max(doaj, publons, scopus, sherpa, na.rm = T))
-  
-setdiff(MADE.IT$publisher, TOP$publisher)
-setdiff(TOP$publisher, MADE.IT$publisher)
-MADE.IT <- MADE.IT[-c(46, 86, 88, 89, 90),]
-
-MIX <- rbind(DOAJ2, PUBLONS2, SCOPUS2, SHERPA2) #%>%
-  #select(Freq, is_in_top_list)
-# prop.table(table(MIX$is_in_top_list, MIX$Freq)) %>%
-#   as.data.table() 
-
-MIX <- MIX %>%
-  group_by(publisher, is_in_top_list) %>%
-  summarise(Freq = max(Freq)) %>%
-  arrange(desc(Freq)) %>%
-  filter(Freq >= 15)
-MIX$formerrank <- seq.int(nrow(MIX))
-
-QUANTILES <- MIX %>%
-  #filter(Freq >= 15) %>%
-  ungroup() %>%
-  mutate(quantile = ntile(-formerrank, 10)) %>% 
-  #mutate(quantile = Hmisc::cut2(Freq, 10)) %>%
-  group_by(quantile) %>%
-  summarise(total = n(),
-            #in_final = sum(is_in_top_list),
-            min = min(formerrank),
-            max = max(formerrank)#,
-            #share = round(in_final / total * 100, 1)
-            )
-
-MADE.IT <- MADE.IT %>%
-  left_join(MIX %>% select(publisher, formerrank)) %>%
-  mutate(quantile = case_when(
-    formerrank >= 1 & formerrank <= 41 ~ 10,
-    formerrank >= 42 & formerrank <= 82 ~ 9,
-    formerrank >= 83 & formerrank <= 123 ~ 8,
-    formerrank >= 124 & formerrank <= 164 ~ 7,
-    formerrank >= 165 & formerrank <= 205 ~ 6,
-    formerrank >= 206 & formerrank <= 246 ~ 5,
-    formerrank >= 247 & formerrank <= 288 ~ 4,
-    formerrank >= 289 & formerrank <= 330 ~ 3,
-    formerrank >= 331 & formerrank <= 372 ~ 2,
-    formerrank >= 373 & formerrank <= 414 ~ 1,
-    TRUE ~ 0
-  ))
-QUANTILES <- QUANTILES %>%
-  mutate(max = MIX$Freq[QUANTILES$min],
-         min = MIX$Freq[QUANTILES$max],
-  )
-
-FF <- MADE.IT %>%
-  group_by(quantile) %>%
-  summarise(in_final = n())
-left_join(QUANTILES, FF) %>%
-  mutate(share = round(in_final / total * 100, 1)) %>%
-  mutate(diff = share-lag(share))
-
-
-MIX %>%
-  filter(Freq >= 15) %>%
-  ungroup() %>%
-  mutate(quantile = ntile(-formerrank, 10)) %>%
-  group_by(quantile) %>%
-  summarise(total = n(),
-            in_final = sum(is_in_top_list),
-            min = min(formerrank),
-            max = max(formerrank),
-            share = round(in_final / total * 100, 1)) %>%
-  mutate(diff = share-lag(share)) %>%
-  summarise(median(diff, na.rm = T))
-aa <- MIX %>%
-  filter(Freq >= 15) %>%
-  group_by(Freq) %>%
-  summarise(total = n(),
-            in_final = sum(is_in_top_list),
-            min = min(Freq),
-            max = max(Freq),
-            share = round(in_final / total * 100, 1)) %>%
-  ungroup() %>%
-  arrange(desc(Freq))
-cor.test(aa$Freq, aa$share, method = "spearman")  
-
-MIX <- MIX %>%
-  group_by(Freq) %>%
-  summarise(total = n(),
-            in_final = sum(is_in_top_list),
-            share = round(in_final / total * 100, 1)) %>%
-  arrange(desc(Freq)) %>%
-  mutate(diff = share-lag(share))
-MIX %>%
-  filter(Freq >= 15) %>%
-  summarise(mean(diff, na.rm = T))
-
-quantile(MIX$Freq)
-cor.test(MIX$Freq, MIX$share, method = "kendall")
-plot(MIX$share, MIX$Freq)
-library(ggplot2)
-ggplot(data = MIX %>% filter(Freq <= 150 & Freq >= 20),
-       aes(x = Freq, y = share)) +
-  geom_point()
-
-head(ALL, 100) %>%
-  mutate(rank = seq(1:100)) %>%
-  select(rank, everything()) %>%
-  writexl::write_xlsx("Output\\finallist-temp.xlsx")
-
-FINFIN <- head(ALL, 100)
-FINFIN <- FINFIN %>%
-  select(publisher, "journals" = nishikawa, predatory, globalsouth)
-
-writexl::write_xlsx()
-
-CSS <- readxl::read_xlsx("Data\\04_publishers_new-15.xlsx") %>%
-  select("publisher" = PUBLISHER_NAME,
-         "webscrape_url" = PUBLISHER_URL,
-         "webscrape_css_selector" = JOURNALNAME_CSS,
-         
-         "webscrape_url2" = PUBLISHER_PAGE_URL_BEFORE)
-
-harmonia <- read.delim(file = "Data\\03_publishers_harmonization.txt", header = TRUE, sep = "\t")
-colnames(harmonia) = c("Deviation", "Harmonization")
-
-for (i in 1:nrow(harmonia)) {
-  
-  CSS$publisher[CSS$publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-
-}
-
-rm(harmonia)
-
-FINFIN %>%
-  left_join(CSS) %>%
-  filter(!(publisher == "Springer" & is.na(webscrape_url))) %>%
-  filter(is.na(webscrape_url) & is.na(webscrape_url2))
-         
\ No newline at end of file
diff --git a/Script/ranking.R b/Script/ranking.R
deleted file mode 100644
index 3df8abe..0000000
--- a/Script/ranking.R
+++ /dev/null
@@ -1,182 +0,0 @@
-library(tidyverse)
-library(ggplot2)
-
-PP <- read_csv("Output/allpublishers-PRELIMINARY-2021-12-09.csv")
-P <- PP %>% group_by(publisher) %>% count() %>% arrange(desc(n))
-
-
-OR <- read_tsv("Data\\04_publishers.tsv") %>%
-  select(PUBLISHER_NAME)
-names(OR)[names(OR) == 'PUBLISHER_NAME'] <- 'publisher'
-
-DF <- left_join(OR, P) %>%
-  arrange(desc(n)) %>%
-  mutate(comment = ifelse(is.na(n), "journals were counted manually", NA))
-#writexl::write_xlsx(DF, "Output\\allpublishers-FINAL-2022-04-14.xlsx")
-
-DF <- DF %>%
-  distinct()
-
-rm(DF, OR, P, PP)
-DF <- readxl::read_xlsx("Output\\allpublishers-FINAL-2022-04-14.xlsx")
-
-# histogram
-ggplot(DF, aes(x = n)) + 
-  geom_histogram(color="black", fill="white", bins = 140)+
-  geom_vline(aes(xintercept = mean(n)),
-               color="red", linetype="dashed", size=1) +
-  theme_classic()
-
-# SUMMARY STATISTICS
-Mode <- function(x) {
-  ux <- unique(x)
-  ux[which.max(tabulate(match(x, ux)))]
-}
-DF %>%
-  filter(!is.na(n)) %>%
-  summarise(total = sum(n, na.rm = TRUE),
-            count = n(),
-            avg = mean(n, na.rm = TRUE),
-            med = median(n, na.rm = TRUE),
-            mode = Mode(n),
-            sd = sd(n, na.rm = TRUE),
-            min = min(n, na.rm = TRUE),
-            q1 = quantile(n, probs = 0.25, na.rm = TRUE),
-            q3 = quantile(n, probs = 0.75, na.rm = TRUE),
-            max = max(n, na.rm = TRUE))
-
-# =============
-#
-# save for table
-#
-# =============
-DF %>%
-  select(publisher, n) %>%
-  arrange(desc(n)) %>%
-  top_n(50) %>%
-  write_tsv("prelim_top50_for_table.tsv")
-
-# =================
-# compare rankings
-# =================
-#
-# IMPORT DATA
-#
-library(readr)
-csv_doaj <- read_csv(file = "Output\\01_publishers_DOAJ.csv")
-csv_publons <- read_csv(file = "Output\\02_publishers_Publons.csv")
-csv_scopus <- read_csv(file = "Output\\03_publishers_Scopus.csv")
-csv_romeo <- read_csv(file = "Output\\04_publishers_SherpaRomeo.csv")
-
-names(csv_doaj) <- c("Publisher", "Journals")
-names(csv_publons) <- c("Publisher", "Journals", "Reviews")
-names(csv_scopus) <- c("Publisher", "Journals")
-names(csv_romeo) <- c("Publisher", "Journals")
-
-# ============================
-# Preliminary Results
-# ============================
-#
-# DOAJ       7217 Publishers
-# Publons     430 Publishers
-# Romeo      4611 Publishers
-# Scopus    11881 Publishers
-#
-
-csv_publons$Reviews <- NULL
-
-alljournals <- list(
-  DOAJ = csv_doaj,
-  Publons = csv_publons,
-  Romeo = csv_romeo,
-  Scopus = csv_scopus
-)
-
-# How many journals should a publisher have in order to be listed?
-# This example shows a threshold of 30 journals per publisher
-threshold <- 30
-
-alljournals <- lapply(alljournals, function(x) subset(x, Journals >= threshold))
-
-csv_doaj <- as.data.frame(alljournals[1])
-csv_publons <- as.data.frame(alljournals[2])
-csv_romeo <- as.data.frame(alljournals[3])
-csv_scopus <- as.data.frame(alljournals[4])
-
-colnames(csv_doaj)[1] <- "Publisher"
-colnames(csv_publons)[1] <- "Publisher"
-colnames(csv_romeo)[1] <- "Publisher"
-colnames(csv_scopus)[1] <- "Publisher"
-
-harmonia <- read.delim(file = "Data\\03_publishers_harmonization.txt", header = TRUE, sep = "\t")
-for (i in 1:nrow(harmonia)) {
-  csv_doaj$Publisher[csv_doaj$Publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-  csv_publons$Publisher[csv_publons$Publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-  csv_romeo$Publisher[csv_romeo$Publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-  csv_scopus$Publisher[csv_scopus$Publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-}
-#add rank
-csv_doaj$rank_doaj <- seq.int(nrow(csv_doaj))
-csv_publons$rank_publons <- seq.int(nrow(csv_publons))
-csv_romeo$rank_romeo <- seq.int(nrow(csv_romeo))
-csv_scopus$rank_scopus <- seq.int(nrow(csv_scopus))
-
-names(csv_doaj) <- c("Publisher", "DOAJ_Journals", "rank_doaj")
-names(csv_publons) <- c("Publisher", "Publons_Journals", "rank_publons")
-names(csv_romeo) <- c("Publisher", "Romeo_Journals", "rank_romeo")
-names(csv_scopus) <- c("Publisher", "Scopus_Journals", "rank_scopus")
-
-
-doaj_publons <- merge(csv_doaj, csv_publons, by = "Publisher", all = T)
-romeo_scopus <- merge(csv_romeo, csv_scopus, by = "Publisher", all = T)
-alljournaldata <- merge(doaj_publons, romeo_scopus, by = "Publisher", all = T)
-
-# harmonize publisher names
-# you can add additional names to harmonize into "03_publishers_harmonization.txt"
-# separated by tabs (\t) (first, the deviation, then, second, after the tab, the harmonization)
-harmonia <- read.delim(file = "Data\\03_publishers_harmonization.txt", header = TRUE, sep = "\t")
-
-for (i in 1:nrow(harmonia)) {
-  alljournaldata$Publisher[alljournaldata$Publisher == harmonia$Deviation[i]] <- harmonia$Harmonization[i]
-}
-
-alljournaldata$DOAJ_Journals <- as.numeric(alljournaldata$DOAJ_Journals)
-alljournaldata$Publons_Journals <- as.numeric(alljournaldata$Publons_Journals)
-alljournaldata$Romeo_Journals <- as.numeric(alljournaldata$Romeo_Journals)
-alljournaldata$Scopus_Journals <- as.numeric(alljournaldata$Scopus_Journals)
-
-
-library(data.table)
-DT <- data.table(alljournaldata)
-DT <- DT[, lapply(.SD, sum, na.rm = T), by = Publisher]
-DT <- DT[order(Publisher)]
-
-DT <- dplyr::mutate(DT, maxjournals = pmax(DOAJ_Journals, Publons_Journals, Romeo_Journals, Scopus_Journals))
-DT <- DT[order(-maxjournals)]
-
-names(DT) <- c("publisher", "j_doaj", "rank_doaj", "j_publons", "rank_publons",
-               "j_romeo", "rank_romeo", "j_scopus", "rank_scopus", "maxjournals")
-DT <- DT %>%
-  select("publisher", starts_with("j_"), starts_with("rank_"))
-
-TEST <- merge(DF, DT, all = TRUE) %>%
-  arrange(desc(n))
-
-TEST$rank_final <- seq.int(nrow(TEST))
-
-RANK <- TEST %>%
-  select(publisher, n, starts_with("rank"))
-RANK$rank_scopus[RANK$publisher == "Springer"] <- 2
-head(RANK)
-
-TEST %>% 
-  select(publisher, n, starts_with("j_")) %>%
-  View()
-# rank overlap
-if (!require("BiocManager", quietly = TRUE))
-  install.packages("BiocManager")
-BiocManager::install("gespeR")
-library(gespeR)
-gespeR::rbo(TEST$rank_final, TEST$rank_doaj)
-
-#library(TopKLists)
\ No newline at end of file