fetch_openaire.r

fetch_openaire.r

require(ropenaire)
require(dplyr)

# fetch ugoe projects
ugoe <- roa_projects(org = "UGOE")
# fetch publications
tt <- plyr::ldply(as.character(ugoe$grantID), roa_pubs)
# tidy up: split collapsed ec projects
tmp <-
  tidyr::separate(
    tt, Project.Name..GA.Number., into = c("grant_1", "grant_2", "grant_3", "grant_4"),
    sep = ", ", extra = "merge"
  )
test <- reshape2::melt(tmp, id.vars = c("Access"),
                       c("grant_1","grant_2","grant_3","grant_4"))
# grep ugoe projects
testa <-
  lapply(ugoe$acronym, function(x)
    test[grep(x, test$value),]) %>% rbind_all
# prepare merging
testa$grant_id <- stringr::str_extract(testa$value, "[:digit:]+")
# tabulate publications for each project and access level
ugoe_table <- data.frame(table(testa$grant_id, testa$Access))
# merge with project list to receive more infos
my_df <-
  merge(ugoe, ugoe_table, by.x = "grantID", by.y = "Var1", all.x = T)

# table
tt_table <-
  reshape::cast(my_df, grantID + acronym + title + startdate + enddate + ecsc39 ~ Var2, value = "Freq")
tt_table[is.na(tt_table)] <- 0
tt_table <- tt_table[,c(1:ncol(tt_table) - 1)]
write.csv(tt_table, "zzz/table_publications.csv", row.names = FALSE)

# plot bar chart
require(ggplot2)
tt_plot <-
  ggplot(my_df, aes(acronym, Freq, fill = Var2)) + geom_bar(stat = "identity") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1))
ggsave("zzz/chart_publications.pdf")