R data.table fread command : how to read large files with irregular separators?

Question

I have to work with a collection of 120 files of ~2 GB (525600 lines x 302 columns). The goal is to make some statistics and put the results in a clean SQLite database.

Answer 1

If peak memory is not an issue or you can stream it in chunks that are manageable, the following gsub()/fread() hybrid should work, converting all continuous space characters to a single delimiter of your choosing (e.g. "\t"), prior to parsing by fread():

fread_blank = function(inputFile, spaceReplace = "\t", n = -1, ...){
  fread(
    input = paste0(
      gsub(pattern = "[[:space:]]+",
           replacement = spaceReplace,
           x = readLines(inputFile, n = n)),
      collapse = "\n"),
    ...)
}

I must agree with others that space-delimited files is not ideal choice, but I come across them pretty often whether I like it or not.

Answer 2

With the answers of NeronLeVelu and Clayton Stanlay, I completed the answer with a custom function, example data and some system.time() for comparison purpose. Those test were made on Mac os 10.9 and R 3.0.2. However, I've made the same test on a linux machine and the sed command was really slow to perform, compared with read.table() with nrows and colClasses pre-calculated. The fread part was really fast, about 5 seconds for 5e6 rows on both system.

library(data.table)


# create path to new temporary file
origData <- tempfile(pattern="origData",fileext=".txt")
# write table with irregular blank spaces separators.
write(paste0(" YYYY MM DD HH mm             19490             40790","\n",
                 paste(rep(" 1991 10  1  1  0      1.046465E+00      1.568405E+00", 5e6), 
                       collapse="\n"),"\n"),
      file=origData
)

# define column classes for read.table() optimization
colClasses <- c(rep('integer',5),rep('numeric',2))

# Function to count rows with command wc for read.table() optimization.
fileRowsCount <- function(file){
    if(file.exists(file)){
            sysCmd <- paste("wc -l", file)
            rowCount <- system(sysCmd, intern=T)
            rowCount <- sub('^\\s', '', rowCount)
        as.numeric(
                       strsplit(rowCount, '\\s')[[1]][1]
                      )
    }
}

# Function to sed data into temp file before importing with sed
sedFread<-function(file, sedCmd=NULL, ...){
    require(data.table)
    if(is.null(sedCmd)){
        #default : sed for convert blank separated table to csv. Thanks NeronLevelu !
        sedCmd <- "'s/^[[:blank:]]*//;s/[[:blank:]]\\{1,\\}/,/g'"
    }
    #sed into temp file
    tmpPath<-tempfile(pattern='tmp',fileext='.txt')
    sysCmd<-paste('sed',sedCmd, file, '>',tmpPath)
    try(system(sysCmd))
    DT<-fread(tmpPath,...)
    try(system(paste('rm',tmpPath)))
    return(DT)
}

Mac OS results :

# First sed into temp file and then fread.
system.time(
DT<-sedFread(origData, header=TRUE)
)
> user  system elapsed
> 23.847   0.628  24.514

# Sed directly in fread command :
system.time(
DT <- fread(paste("sed 's/^[[:blank:]]*//;s/[[:blank:]]\\{1,\\}/,/g'", origData),
            header=T)
)
> user  system elapsed
> 23.606   0.515  24.219


# read.table without nrows and colclasses
system.time(
DF<-read.table(origData, header=TRUE)
)
> user  system elapsed
> 38.053   0.512  38.565

# read.table with nrows an colclasses
system.time(
DF<-read.table(origData, header=TRUE, nrows=fileRowsCount(origData), colClasses=colClasses)
)
> user  system elapsed
> 33.813   0.309  34.125

Linux results :

# First sed into temp file and then fread.
system.time(
  DT<-sedFread(origData, header=TRUE)
)
> Read 5000000 rows and 7 (of 7) columns from 0.186 GB file in 00:00:05
> user  system elapsed 
> 47.055   0.724  47.789 

# Sed directly in fread command :
system.time(
DT <- fread(paste("sed 's/^[[:blank:]]*//;s/[[:blank:]]\\{1,\\}/,/g'", origData),
            header=T)
)
> Read 5000000 rows and 7 (of 7) columns from 0.186 GB file in 00:00:05
> user  system elapsed 
> 46.088   0.532  46.623 

# read.table without nrows and colclasses
system.time(
DF<-read.table(origData, header=TRUE)
)
> user  system elapsed 
> 32.478   0.436  32.912 

# read.table with nrows an colclasses
system.time(
DF<-read.table(origData,
               header=TRUE, 
               nrows=fileRowsCount(origData),
               colClasses=colClasses)
 )
> user  system elapsed 
> 21.665   0.524  22.192 

# Control if DT and DF are identical : 
setnames(DT, old=names(DT), new=names(DF))
identical(as.data.frame(DT), DF)                                                              
>[1] TRUE

In fine : the method I used in first place was the most efficient, in this case.

Thanks to NeronLeVelu, Matt Dowle and Clayton Stanley!

Answer 3

I've found another way to do it, much faster, with awk instead of sed. Here is another example :

library(data.table)

# create path to new temporary file
origData <- tempfile(pattern="origData",fileext=".txt")

# write table with irregular blank spaces separators.
write(paste0(" YYYY MM DD HH mm             19490             40790","\n",
            paste(rep(" 1991 10  1  1  0      1.046465E+00      1.568405E+00", 5e6),
            collapse="\n"),"\n"),
            file=origData
  )


# function awkFread : first awk, then fread. Argument : colNums = selection of columns. 
awkFread<-function(file, colNums, ...){
        require(data.table)
        if(is.vector(colNums)){
            tmpPath<-tempfile(pattern='tmp',fileext='.txt')
            colGen<-paste0("$",colNums,"\",\"", collapse=",")
            colGen<-substr(colGen,1,nchar(colGen)-3)
            cmdAwk<-paste("awk '{print",colGen,"}'", file, '>', tmpPath)
            try(system(cmdAwk))
            DT<-fread(tmpPath,...)
            try(system(paste('rm', tmpPath)))
            return(DT)
        }
}

# check read time :
system.time(
            DT3 <- awkFread(origData,c(1:5),header=T)
            )

> user  system elapsed 
> 6.230   0.408   6.644

Answer 4

Just committed to devel, v1.9.5. fread() gains strip.white argument with default TRUE (as opposed to base::read.table(), because it's more desirable). The example data is now added to tests.

With this recent commit:

require(data.table) # v1.9.5, commit 0e7a835 or more recent
ans <- fread(" YYYY MM DD HH mm             19490             40790\n   1991 10  1  1  0      1.046465E+00      1.568405E+00")
#      V1 V2 V3 V4 V5           V6           V7
# 1: YYYY MM DD HH mm 19490.000000 40790.000000
# 2: 1991 10  1  1  0     1.046465     1.568405
sapply(ans, class)
#          V1          V2          V3          V4          V5          V6          V7 
# "character" "character" "character" "character" "character"   "numeric"   "numeric" 

Answer 5

sed 's/^[[:blank:]]*//;s/[[:blank:]]\{1,\}/,/g' 

for you sed

it's not possible to collect all result of fread into 1 (temporary) file (adding the source reference) and treat this file with sed (or other tool) to avoid a fork of the tools at every iteration ?