Fixed last dataframe joins + minor cleanup.

scraping/geocoding-viviendas-turisticas-comunitat-valenciana.R

@@ -25,8 +25,6 @@ df = read.csv("data/original/190302_viviendas-turisticas-comunidad-valenciana_va
   separate(num, c("num", "puerta"), extra = "merge", sep = ", ", remove = FALSE) %>%
   mutate(nombre_es_raw = as.factor(toupper(nombre_es_raw)))
 
-
-
 # Basic Record linkage.
 record.linkage.names = function(names,
                                 officialnames = calles_valencia$nombre_es) {
@@ -43,15 +41,13 @@ record.linkage.names = function(names,
 }
 
 
-# df$nombre_es = record.linkage.names(df$nombre_es_raw)
-
 df = df %>%
   mutate(nombre_es = record.linkage.names(nombre_es_raw)) %>%
   left_join(calles_valencia)
 
 # Geocoding with Photon ---------------------------------------------------
 
-
+# Build a new dataframe with desired information.
 df2 = df %>%
   select(Signatura, Municipio, tipovia_ca, nomoficial, num) %>%
   mutate(full_address_ca = paste(tipovia_ca, nomoficial, num, Municipio,
@@ -64,8 +60,7 @@ geocoded.df = photon::geocode(head(df2$full_address_ca), limit = 1,
 
 
 # Combine geocoded dataframe with original one.
-df.combined = geocoded.df %>%
-  select(location, lon, lat) %>%
-  right_join(df, by = c("location" = "Address2"))
+df.combined = df2 %>%
+  left_join(geocoded.df, by = c("full_address_ca" = "location"))
 
 write.csv(df.combined, file = "data/output/filename.csv")