将字符串分成多列

Question

我想将下面的数据框分为五列。 列应在每个“-”之后创建。 请注意，某些观测值（26和28）具有一个额外的字段（“ uk”和“ es”）。 因此，在最后一列中，除26和28以外的所有观测值都应包含NA。

26                               paid 21.09 - abs - E X1028 - 61,77 - uk.pdf
27                               paid 21.09 - corefunction - mah - 125,66.PDF
28                               paid 21.09 - mrl - mah - 456,96 - es.PDF
29                               paid 21.09 - mollea - inv - 297,50.pdf
30                               paid 21.09 - saless - inv - 117,81.pdf
31                               paid 23.09 - boc - inv - 59,80.pdf

Answer 1

使用data.table的快速解决方案

library(data.table) # v 1.9.6+
setDT(df)[, tstrsplit(V1, "-")]
#             V1             V2        V3          V4      V5
# 1: paid 21.09            abs   E X1028       61,77   uk.pdf
# 2: paid 21.09   corefunction       mah   125,66.PDF      NA
# 3: paid 21.09            mrl       mah      456,96   es.PDF
# 4: paid 21.09         mollea       inv   297,50.pdf      NA
# 5: paid 21.09         saless       inv   117,81.pdf      NA
# 6: paid 23.09            boc       inv    59,80.pdf      NA

---

数据

df <- structure(list(V1 = structure(c(1L, 2L, 4L, 3L, 5L, 6L), .Label = c("paid 21.09 - abs - E X1028 - 61,77 - uk.pdf", 
"paid 21.09 - corefunction - mah - 125,66.PDF", "paid 21.09 - mollea - inv - 297,50.pdf", 
"paid 21.09 - mrl - mah - 456,96 - es.PDF", "paid 21.09 - saless - inv - 117,81.pdf", 
"paid 23.09 - boc - inv - 59,80.pdf"), class = "factor")), .Names = "V1", class = "data.frame", row.names = c(NA, 
-6L))

Answer 2

或使用splistackshape ：

library(splitstackshape)

cSplit(df, 'V1', sep='-')
#         V1_1         V1_2    V1_3       V1_4   V1_5
#1: paid 21.09          abs E X1028      61,77 uk.pdf
#2: paid 21.09 corefunction     mah 125,66.PDF     NA
#3: paid 21.09          mrl     mah     456,96 es.PDF
#4: paid 21.09       mollea     inv 297,50.pdf     NA
#5: paid 21.09       saless     inv 117,81.pdf     NA
#6: paid 23.09          boc     inv  59,80.pdf     NA

Answer 3

您可以像这样读入数据，如果您从文件中读入数据，它也比先使用read.table然后再使用data.table函数要快。

 library(read)
 df<-read_delim("test.txt", "-", col_names  = F)

Answer 4

我将遍历数据并拆分每一行。 不确定我是否了解NA问题。

for (i in 1:nrow(data)) {
  cols=strsplit(data[i,],"-")
  newdata[i]=cols
}