Only read selected columns
RImportR FaqR Problem Overview
Can anyone please tell me how to read only the first 6 months (7 columns) for each year of the data below, for example by using read.table()?
Year Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
2009 -41 -27 -25 -31 -31 -39 -25 -15 -30 -27 -21 -25
2010 -41 -27 -25 -31 -31 -39 -25 -15 -30 -27 -21 -25
2011 -21 -27 -2 -6 -10 -32 -13 -12 -27 -30 -38 -29
R Solutions
Solution 1 - R
Say the data are in file data.txt, you can use the colClasses argument of read.table() to skip columns. Here the data in the first 7 columns are "integer" and we set the remaining 6 columns to "NULL" indicating they should be skipped
> read.table("data.txt", colClasses = c(rep("integer", 7), rep("NULL", 6)),
+ header = TRUE)
Year Jan Feb Mar Apr May Jun
1 2009 -41 -27 -25 -31 -31 -39
2 2010 -41 -27 -25 -31 -31 -39
3 2011 -21 -27 -2 -6 -10 -32
Change "integer" to one of the accepted types as detailed in ?read.table depending on the real type of data.
data.txt looks like this:
$ cat data.txt
"Year" "Jan" "Feb" "Mar" "Apr" "May" "Jun" "Jul" "Aug" "Sep" "Oct" "Nov" "Dec"
2009 -41 -27 -25 -31 -31 -39 -25 -15 -30 -27 -21 -25
2010 -41 -27 -25 -31 -31 -39 -25 -15 -30 -27 -21 -25
2011 -21 -27 -2 -6 -10 -32 -13 -12 -27 -30 -38 -29
and was created by using
write.table(dat, file = "data.txt", row.names = FALSE)
where dat is
dat <- structure(list(Year = 2009:2011, Jan = c(-41L, -41L, -21L), Feb = c(-27L,
-27L, -27L), Mar = c(-25L, -25L, -2L), Apr = c(-31L, -31L, -6L
), May = c(-31L, -31L, -10L), Jun = c(-39L, -39L, -32L), Jul = c(-25L,
-25L, -13L), Aug = c(-15L, -15L, -12L), Sep = c(-30L, -30L, -27L
), Oct = c(-27L, -27L, -30L), Nov = c(-21L, -21L, -38L), Dec = c(-25L,
-25L, -29L)), .Names = c("Year", "Jan", "Feb", "Mar", "Apr",
"May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"), class = "data.frame",
row.names = c(NA, -3L))
If the number of columns is not known beforehand, the utility function count.fields will read through the file and count the number of fields in each line.
## returns a vector equal to the number of lines in the file
count.fields("data.txt", sep = "\t")
## returns the maximum to set colClasses
max(count.fields("data.txt", sep = "\t"))
Solution 2 - R
To read a specific set of columns from a dataset you, there are several other options:
1) With freadfrom the data.table-package:
You can specify the desired columns with the select parameter from fread from the data.table package. You can specify the columns with a vector of column names or column numbers.
For the example dataset:
library(data.table)
dat <- fread("data.txt", select = c("Year","Jan","Feb","Mar","Apr","May","Jun"))
dat <- fread("data.txt", select = c(1:7))
Alternatively, you can use the drop parameter to indicate which columns should not be read:
dat <- fread("data.txt", drop = c("Jul","Aug","Sep","Oct","Nov","Dec"))
dat <- fread("data.txt", drop = c(8:13))
All result in:
> data
Year Jan Feb Mar Apr May Jun
1 2009 -41 -27 -25 -31 -31 -39
2 2010 -41 -27 -25 -31 -31 -39
3 2011 -21 -27 -2 -6 -10 -32
UPDATE: When you don't want fread to return a data.table, use the data.table = FALSE-parameter, e.g.: fread("data.txt", select = c(1:7), data.table = FALSE)
2) With read.csv.sql from the sqldf-package:
Another alternative is the read.csv.sql function from the sqldf package:
library(sqldf)
dat <- read.csv.sql("data.txt",
sql = "select Year,Jan,Feb,Mar,Apr,May,Jun from file",
sep = "\t")
3) With the read_*-functions from the readr-package:
library(readr)
dat <- read_table("data.txt",
col_types = cols_only(Year = 'i', Jan = 'i', Feb = 'i', Mar = 'i',
Apr = 'i', May = 'i', Jun = 'i'))
dat <- read_table("data.txt",
col_types = list(Jul = col_skip(), Aug = col_skip(), Sep = col_skip(),
Oct = col_skip(), Nov = col_skip(), Dec = col_skip()))
dat <- read_table("data.txt", col_types = 'iiiiiii______')
From the documentation an explanation for the used characters with col_types:
> each character represents one column: c = character, i = integer, n = number, d = double, l = logical, D = date, T = date time, t = time, ? = guess, or _/- to skip the column
Solution 3 - R
You could also use JDBC to achieve this. Let's create a sample csv file.
write.table(x=mtcars, file="mtcars.csv", sep=",", row.names=F, col.names=T) # create example csv file
Download and save the the CSV JDBC driver from this link: http://sourceforge.net/projects/csvjdbc/files/latest/download
> library(RJDBC)
> path.to.jdbc.driver <- "jdbc//csvjdbc-1.0-18.jar"
> drv <- JDBC("org.relique.jdbc.csv.CsvDriver", path.to.jdbc.driver)
> conn <- dbConnect(drv, sprintf("jdbc:relique:csv:%s", getwd()))
> head(dbGetQuery(conn, "select * from mtcars"), 3)
mpg cyl disp hp drat wt qsec vs am gear carb
1 21 6 160 110 3.9 2.62 16.46 0 1 4 4
2 21 6 160 110 3.9 2.875 17.02 0 1 4 4
3 22.8 4 108 93 3.85 2.32 18.61 1 1 4 1
> head(dbGetQuery(conn, "select mpg, gear from mtcars"), 3)
MPG GEAR
1 21 4
2 21 4
3 22.8 4
Solution 4 - R
The vroom package provides a 'tidy' method of selecting / dropping columns by name during import. Docs: https://www.tidyverse.org/blog/2019/05/vroom-1-0-0/#column-selection
Column selection (col_select)
The vroom argument 'col_select' makes selecting columns to keep (or omit) more straightforward. The interface for col_select is the same as dplyr::select().
Select columns by name
data <- vroom("flights.tsv", col_select = c(year, flight, tailnum))
#> Observations: 336,776
#> Variables: 3
#> chr [1]: tailnum
#> dbl [2]: year, flight
#>
#> Call `spec()` for a copy-pastable column specification
#> Specify the column types with `col_types` to quiet this message
Drop columns by name
data <- vroom("flights.tsv", col_select = c(-dep_time, -air_time:-time_hour))
#> Observations: 336,776
#> Variables: 13
#> chr [4]: carrier, tailnum, origin, dest
#> dbl [9]: year, month, day, sched_dep_time, dep_delay, arr_time, sched_arr_time, arr...
#>
#> Call `spec()` for a copy-pastable column specification
#> Specify the column types with `col_types` to quiet this message
Use the selection helpers
data <- vroom("flights.tsv", col_select = ends_with("time"))
#> Observations: 336,776
#> Variables: 5
#> dbl [5]: dep_time, sched_dep_time, arr_time, sched_arr_time, air_time
#>
#> Call `spec()` for a copy-pastable column specification
#> Specify the column types with `col_types` to quiet this message
Or rename columns by name
data <- vroom("flights.tsv", col_select = list(plane = tailnum, everything()))
#> Observations: 336,776
#> Variables: 19
#> chr [ 4]: carrier, tailnum, origin, dest
#> dbl [14]: year, month, day, dep_time, sched_dep_time, dep_delay, arr_time, sched_arr...
#> dttm [ 1]: time_hour
#>
#> Call `spec()` for a copy-pastable column specification
#> Specify the column types with `col_types` to quiet this message
data
#> # A tibble: 336,776 x 19
#> plane year month day dep_time sched_dep_time dep_delay arr_time
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 N142… 2013 1 1 517 515 2 830
#> 2 N242… 2013 1 1 533 529 4 850
#> 3 N619… 2013 1 1 542 540 2 923
#> 4 N804… 2013 1 1 544 545 -1 1004
#> 5 N668… 2013 1 1 554 600 -6 812
#> 6 N394… 2013 1 1 554 558 -4 740
#> 7 N516… 2013 1 1 555 600 -5 913
#> 8 N829… 2013 1 1 557 600 -3 709
#> 9 N593… 2013 1 1 557 600 -3 838
#> 10 N3AL… 2013 1 1 558 600 -2 753
#> # … with 336,766 more rows, and 11 more variables: sched_arr_time <dbl>,
#> # arr_delay <dbl>, carrier <chr>, flight <dbl>, origin <chr>,
#> # dest <chr>, air_time <dbl>, distance <dbl>, hour <dbl>, minute <dbl>,
#> # time_hour <dttm>
Solution 5 - R
You do it like this:
df = read.table("file.txt", nrows=1, header=TRUE, sep="\t", stringsAsFactors=FALSE)
colClasses = as.list(apply(df, 2, class))
needCols = c("Year", "Jan", "Feb", "Mar", "Apr", "May", "Jun")
colClasses[!names(colClasses) %in% needCols] = list(NULL)
df = read.table("file.txt", header=TRUE, colClasses=colClasses, sep="\t", stringsAsFactors=FALSE)