我有一个复杂的编码问题，希望您能提供帮助：
每个月我都必须使用更改表对查找表进行更改，该更改表概述了（1）要添加的新产品和（2）产品名称的更改。我正在尝试使此过程自动化，因为我正在处理40,000行以上的数据。这里有太多敏感数据无法显示，但是我在下面给出了一个有关数据和过程的小例子：
查询表
dput(LookupFile)
structure(list(ProductTag = structure(c(1L, 6L, 3L, 5L, 2L, 7L,
4L), .Label = c("Confect", "Orange", "Pastries", "Root", "Sauces",
"Spread", "White"), class = "factor"), Brand = structure(c(1L,
1L, 1L, 2L, 2L, 3L, 4L), .Label = c("A", "B", "C", "D"), class = "factor"),
BrandID = c(340, 340, 340, 400, 400, 402, 403), Name = structure(c(3L,
4L, 1L, 7L, 2L, 5L, 6L), .Label = c("Cakes", "Carrots", "Choc",
"Honey", "Onions", "Potatoes", "Syrups"), class = "factor"),
NameID = c(200, 700, 100, 300, 1000, 1200, 1700)), class = "data.frame", row.names = c(NA,
-7L))
变更表
dput(ChangesFiles)
structure(list(ProductTag = structure(c(1L, 3L, 2L), .Label = c("Breads",
"Confect", "Flours"), class = "factor"), Brand = structure(c(1L,
2L, 1L), .Label = c("A", "D"), class = "factor"), BrandID = c(340,
403, 340), Name = structure(c(2L, 3L, 1L), .Label = c("Chocolate",
"Gluten", "Staple"), class = "factor"), NameID = c(4000, 3400,
200)), class = "data.frame", row.names = c(NA, -3L))
我知道rbind允许我将新数据附加到查找表上：
rbind(LookupFile,ChangesFiles)
但是，这又增加了一层必须修改现有行（即产品名称）的内容。在我给出的示例中，我想使用ChangesFiles记录更改以下记录，该记录将Name ='Choc'替换为'Chocolate'：
<ProductTag = 'Confect', Brand = 'A', BrandID = '340', Name = 'Choc', NameID = '200'
这是一个小数据样本，因此我可以轻松地手动进行此更改。但是，我将如何进行大规模更改？
我希望它看起来像这样：
ProductTag Brand BrandID Name NameID
1 Confect A 340 Chocolate 200
2 Spread A 340 Honey 700
3 Pastries A 340 Cakes 100
4 Sauces B 400 Syrups 300
5 Orange B 400 Carrots 1000
6 White C 402 Onions 1200
7 Root D 403 Potatoes 1700
8 Breads A 340 Gluten 4000
9 Flours D 403 Staple 3400
答案 0 :(得分：0)
这是一个Vlookup选项
> library(qdapTools)
> LookupFile$Name <- as.character(LookupFile$Name) # just for not dealing with factors
> replace <- as.character(LookupFile$ProductTag %l% ChangesFiles[,c(1,4)]) # here's a Vlookup
> ind <- !is.na(replace) # identifying value to replace
> LookupFile$Name[ind] <- replace[ind] # replacing
> rbind(LookupFile,ChangesFiles) # delivering your desired output
ProductTag Brand BrandID Name NameID
1 Confect A 340 Chocolate 200
2 Spread A 340 Honey 700
3 Pastries A 340 Cakes 100
4 Sauces B 400 Syrups 300
5 Orange B 400 Carrots 1000
6 White C 402 Onions 1200
7 Root D 403 Potatoes 1700
8 Breads A 340 Gluten 4000
9 Flours D 403 Staple 3400
10 Confect A 340 Chocolate 200
答案 1 :(得分：0)
在Changes中处理数据（并将其重命名为Lookup和tidyverse）：
library(tidyverse)
bind_rows(
anti_join(Changes, Lookup, by = colnames(Changes)), # data without changes
anti_join(Lookup, Changes, by = colnames(Lookup)[-4]) # changed names & added
) %>%
arrange(BrandID, NameID)
# ProductTag Brand BrandID Name NameID
# 1 Pastries A 340 Cakes 100
# 2 Confect A 340 Chocolate 200
# 3 Breads A 340 Gluten 4000
# 4 Spread A 340 Honey 700
# 5 Orange B 400 Carrots 1000
# 6 Sauces B 400 Syrups 300
# 7 White C 402 Onions 1200
# 8 Root D 403 Potatoes 1700
# 9 Flours D 403 Staple 3400
答案 2 :(得分：0)
仅从LookupFile中获取ChangesFiles中未定义的行，而从ChangesFiles中获取所有内容。
key <- c("BrandID", "NameID") #Assuming that BrandID and NameID need to be unique
rbind(LookupFile[!interaction(LookupFile[key]) %in% interaction(ChangesFiles[key]),], ChangesFiles)
ProductTag Brand BrandID Name NameID
2 Spread A 340 Honey 700
3 Pastries A 340 Cakes 100
4 Sauces B 400 Syrups 300
5 Orange B 400 Carrots 1000
6 White C 402 Onions 1200
7 Root D 403 Potatoes 1700
1 Breads A 340 Gluten 4000
21 Flours D 403 Staple 3400
31 Confect A 340 Chocolate 200
如果还有要删除的内容，并且在ChangesFiles中用NA标记，则可以使用：
rbind(LookupFile[!interaction(LookupFile[key]) %in% interaction(ChangesFiles[key]),], ChangesFiles[complete.cases(ChangesFiles),])
答案 3 :(得分：0)
data.table来营救！这是单线的。.
使用更新联接仅更新名称列。
library( data.table )
setDT(LookupFile)[ setDT(ChangesFiles), Name := i.Name, on = .(ProductTag, Brand, BrandID)][]
# ProductTag Brand BrandID Name NameID
# 1: Confect A 340 Chocolate 200
# 2: Spread A 340 Honey 700
# 3: Pastries A 340 Cakes 100
# 4: Sauces B 400 Syrups 300
# 5: Orange B 400 Carrots 1000
# 6: White C 402 Onions 1200
# 7: Root D 403 Potatoes 1700
setDT()用于将data.frames转换为data.table-class。如果您的数据已经在data.table-类中，则
LookupFile [ChangesFiles，Name：= i.Name，on =。（ProductTag，Brand，BrandID）] []
也可以。