在 R 中将 TXT 批量编辑为 CSV

Question

我有一系列 .txt 文件，如下所示：

Button,Intensity,Acc,Intensity,RT,Time
0,30,0,0,0,77987.931
1,30,1,13.5,0,78084.57
1,30,1,15,0,78098.624
1,30,1,6,0,78114.132
1,30,1,15,0,78120.669

它们的文件名类似于 1531_Day49.txt、1531_Day50.txt、1532_Day49.txt、1532_Day50.txt 等

我想将此目录中的所有文件加载到数据框中，附加一列是上面一行中的时间 (tdelta) 之间的差异，并附加前 4 位数字的两列（即 1531 , 1532) 和一列未编码的日期代码，因此列标题将是 PrePost，每一行将是，如果文件名 Day49，则 "Pre"，如果文件名 Day50，则 "Post"。

1531 第 49 天文件的理想输出为：

Button,Intensity,Acc,Intensity,RT,Time,Tdelta,ID,PrePost
0,30,0,0,0,77987.931,0 ,1531,Pre
1,30,1,13.5,0,78084.57,96.693 ,1531,Pre
1,30,1,15,0,78098.624, 14.054,1531,Pre

到目前为止我有：

#call library
library(data.table)

#batch enter .txt files and put them into a data frame 
setwd("~/Documents/PVTPASAT/PVT")
temp = list.files(pattern="*.txt")
list.DFs <- lapply(myfiles,fread)
#view print out to visually check
View(list.DFs)

#add column of time difference 
list.DFs <- lapply(list.DFs, cbind, tDelta = c(0, diff(df$Time)))
#Add empty columns for ID and PrePost
list.DFs <- lapply(list.DFs, cbind, ID = c(""))
list.DFs <- lapply(list.DFs, cbind, PrePost = c(""))

#print one to visually check 
View(list.DFs[3])

Answer 1

我会创建一个函数来进行处理，然后像这样将其应用于您的文件列表：

example <- read.delim(textConnection('
  Button, Intensity, Acc, Intensity, RT, Time
  0,30,0,0,0,77987.931
  1,30,1,13.5,0,78084.57
  1,30,1,15,0,78098.624
  1,30,1,6,0,78114.132
  1,30,1,15,0,78120.669'), 
  header = T, 
  sep = ','
  )
write.table(example, '1531_Day49.txt', row.names = F)

temp <- list.files(pattern="*.txt")

process_txt <- function(x) {
  dat <- data.table::fread(x, header = T)
  dat$tdelta <- c(0, diff(dat$Time))
  dat$ID <- substr(x, 1, 4)
  dat$PrePost <- if (grepl('49\.', x)) {'Pre'} else {'Post'}
  dat
}

out <- lapply(temp, process_txt)

Answer 2

@Heather，主要指导是先正确解决一个文件。然后，将所有工作代码放入一个函数中。

library(dplyr)  ## for lag function
library(stringr) ## for str_detect

# make two test files
dt <- read.csv(text=
'Button,Intensity,Acc,Intensity,RT,Time
0,30,0,0,0,77987.931
1,30,1,13.5,0,78084.57
1,30,1,15,0,78098.624
1,30,1,6,0,78114.132
1,30,1,15,0,78120.669
')
write.csv(dt,"1531_Day49.txt")
write.csv(dt,"1532_Day50.txt")

# function to do the work for one file name - returns a dataframe
doOne <- function (file) {

  # read 
  contents <- fread(file)

  # compute delta
  contents$Tdelta <- contents$Time - lag(contents$Time)

  # prefix up to underscore
  contents$ID <- strsplit(file, c("_"))[[1]][[1]]

  # add the prepost using ifelse and str_detetct
  contents$PrePost <- ifelse(str_detect(file, "Day49"), "Pre", "Post")
  return(contents)
}

#test files
files <- c("1531_Day49.txt", "1532_Day50.txt")

# call the function for each file -- result is
# a list of dataframes
lapply(files, doOne)

# better get them all into a single data frame for analysis
do.call(rbind, lapply(files, doOne))

#    V1 Button Intensity Acc Intensity.1 RT     Time Tdelta   ID PrePost
# 1:  1      0        30   0         0.0  0 77987.93     NA 1531     Pre
# 2:  2      1        30   1        13.5  0 78084.57 96.639 1531     Pre
# 3:  3      1        30   1        15.0  0 78098.62 14.054 1531     Pre
# 4:  4      1        30   1         6.0  0 78114.13 15.508 1531     Pre
# 5:  5      1        30   1        15.0  0 78120.67  6.537 1531     Pre
# 6:  1      0        30   0         0.0  0 77987.93     NA 1532    Post
# 7:  2      1        30   1        13.5  0 78084.57 96.639 1532    Post
# 8:  3      1        30   1        15.0  0 78098.62 14.054 1532    Post
# 9:  4      1        30   1         6.0  0 78114.13 15.508 1532    Post
# 10:  5      1        30   1        15.0  0 78120.67  6.537 1532    Post

在 R 中将 TXT 批量编辑为 CSV

Batch Editing TXT to CSV in R

csv

r

batch-processing

lapply