Seitenhistorie

# Crosskonkordanzen einlesen
ck <- read_csv('Tc.csv', col_types = 'cccc')

# Wertebereiche gemäß Wiki-Seiten definieren
thesauri <- read_csv('Thesauri.csv', col_types = 'cc')
relationen <- read_csv('Relationen.csv', col_types = 'cc')

# Crosskonkordanzen filtern und zählen
rslt1 <- ck %>%
  filter(Thesaurus %in% thesauri$Code & Relation %in% relationen$Code) %>%
  mutate(
    Thesaurus = factor(Thesaurus, 
                       levels = thesauri$Code, labels = thesauri$Thesaurus),
    Relation  = factor(Relation,
                       levels = relationen$Code, labels = relationen$Relation)
  ) %>% # auch alle leeren Levels jeweils mit Anzahl 0 anzeigen
  group_by(Thesaurus, Relation, .drop = FALSE) %>% 
  # jede Kombination (Thesaurus, Relation) je IDN nur einmal zählen
  summarise(n = n_distinct(IDN)) %>%
  pivot_wider(id_cols = Relation, names_from = Thesaurus, values_from = n)

# Summe
rslt2 <- data.frame(Relation = factor('GESAMT'), 
                    rslt1 %>% select(-Relation) %>% summarise_all(.funs = sum))
colnames(rslt2) <- colnames(rslt1)

# Formatierte Ausgabetabelle
rbind(rslt1, rslt2) %>% 
  mutate_at(.vars = setdiff(colnames(.), 'Relation'), .funs = ~ formatC(
    ., format = 'd', big.mark = '.', decimal.mark = ',')) %>%
  knitr::kable(align = c('l', rep('r', ncol(.)-1)))