rename some columns to english

R/repair.R

@@ -29,15 +29,13 @@ repair_speaker <- function(speaker) {
         filter(id != "10000") %>% # invalid id's
         mutate(fraction = Vectorize(repair_fraction)(fraction)) %>% # fix fraction
         group_by(id) %>%
-        summarize(vorname = head(vorname, 1),
-                  nachname = head(nachname, 1),
+        summarize(prename = head(prename, 1),
+                  lastname = head(lastname, 1),
                   fraction = collect_unique(fraction),
-                  titel = longest_titel(titel),
-                  rolle_kurz = collect_unique(str_squish(rolle_kurz)),
-                  rolle_lang = collect_unique(str_squish(rolle_lang))) %>%
+                  title = longest_titel(title),
+                  role_short = collect_unique(str_squish(role_short)),
+                  role_long = collect_unique(str_squish(role_long))) %>%
         ungroup() #%>%
-        # arrange(id) %>%
-        # distinct(vorname, nachname, fraction, titel)
 }
 
 repair_speeches <- function(speeches) {
@@ -68,7 +66,7 @@ repair_talks <- function(talks) {
 lookup_speaker <- function(tb, speaker, name_variable) {
     tobereplaced <- "[-–—‑­­-­­­ ]"
     speaker %>%
-        unite(name, vorname, nachname, sep=".*") %>%
+        unite(name, prename, lastname, sep=".*") %>%
         mutate(name = str_replace_all(name, tobereplaced, ".*")) ->
         rs
     find_match <- function(komm) {
@@ -88,10 +86,10 @@ repair_comments <- function(comments, speaker) {
                "Use repair(, repair_commments = FALSE) to skip this.\n"))
     # try to find a speaker id for each actual comment
     comments %>%
-        filter(!is.na(kommentator)) %>%
-        lookup_speaker(speaker, kommentator) %>%
-        left_join(comments, ., by="kommentator") %>%
-        select(-kommentator)
+        filter(!is.na(commenter)) %>%
+        lookup_speaker(speaker, commenter) %>%
+        left_join(comments, ., by="commenter") %>%
+        select(-commenter)
 }
 
 #' Repair parsed tables