用 R 中另一列的值替換特定列中的非 NA 值

Question

只要 City.y 列中沒有 NA，我就會嘗試用 City.y 列中的值替換合並表數據框 City.x 列中的值。

換句話說，我想替換 City.x 列中除 NA 之外的所有值。

這是我到目前為止的代碼：

library(tidyverse)
library(dplyr)

# Import food data
food <-
  read_csv(file = 'https://s3.amazonaws.com/notredame.analytics.data/inspections.csv', 
           col_names=c("ID", 
                       "DBAName", 
                       "AKAName", 
                       "License", 
                       "FacilityType",
                       "Risk",
                       "Address",
                       "City",
                       "State",
                       "ZIP",
                       "InspectionDate",
                       "InspectionType",
                       "Results",
                       "Violations",
                       "Latitude",
                       "Longitude",
                       "Location"), 
           col_types = "icccffcfffcffcddc",
           skip = 1)

# Change InspectionDate from character type to datetime type
food$InspectionDate <- strptime(food$InspectionDate, "%m/%d/%Y")

#Import zipcode data
zipcode <- 
  read_csv('https://s3.amazonaws.com/notredame.analytics.data/zipcode.csv', 
           col_names = c("ZIP",
                         "City", 
                         "State",
                         "Latitude",
                         "Longitude"),
           skip = 1)

# Convert ZIP, City, and State from character type to factor type
zipcode$ZIP <- as.factor(zipcode$ZIP)
zipcode$City <- as.factor(zipcode$City)
zipcode$State <- as.factor(zipcode$State)

#Correct zip codes (told these were incorrect)
food <- food %>%
  mutate(food$ZIP = ifelse("60627", "60827", ZIP))

#Create merged table from food and zipcode tables
mergedtable <- merge(x=food,y=zipcode,by="ZIP",all.x=TRUE)

#new_DF <- mergedtable[is.na(mergedtable$ZIP),]

mergedtable <- mergedtable %>%
  mutate(mergedtable$City.x = ifelse(!is.na(mergedtable$City.y), mergedtable$City.y, mergedtable$City.x))

mergedtable$City.x <- ifelse(!is.na(mergedtable$City.y), mergedtable$City.y, mergedtable$City.x)

最后的兩行代碼都沒有做我想要的。 第一個返回錯誤：

Error: unexpected '=' in:  
"mergedtable <- mergedtable %>%  
  mutate(mergedtable$City.x ="

最后一行將 mergetable$City.x 中的值轉換為數字，但我不確定這些數字來自哪里。

Answer 1

coalesce會更容易

library(dplyr)
mergedtable2 <- mergedtable %>%
                     mutate(ZIP = coalesce(City.y, City.x))

---

在 OP 的代碼中，我們只需要不帶引號的名稱（'ZIP'）就可以創建一個新列

mergedtable %>%
         mutate(ZIP = ifelse(!is.na(City.y), City.y, City.x))
                ^^^

相似地

food <- food %>%
              mutate(ZIP = ifelse("60627", "60827", ZIP))
                     ^^^