2 min read



Head and Tail

it allows you to clip the data by line numbers (-n) or by bytes. The following example is to print the 1st 1K bytes of the last 2000000K bytes of a file.

tail --bytes=2000000K test.dbf | head --bytes=1k

Zip and unzip a folder

gzip can only zip a single file. If you want to zip a folder, need to use “tar”: example:

$ tar -zcvf tian.tar.gz /home/tian

to untar:

$ tar -zxvf tian.tar.gz

if it’s .tar instead of .tar.gz, then remove the -z

$ tar -xvf tian.tar

Change permission

$ chmod a=rwx ./*        #changes all to read, write, and executable
$ chmod -R go-w TMPAV7/    #removes the write permission from group or other for all files in TMPAV7 directory

Find string in a directory

$ grep -r "boo" /path/to/file  #find all file with "boo" 


ctrl+alt+F1 to exit x window startx to open x window ctrl+alt+* to kill the front process

File download

  • using wget to download files from http sites
    $ #download a single file
    $ wget http://hydrology.princeton.edu/data/pgf/0.5deg/daily/prcp_daily_1948-1948.nc
    $ #download a directory without including the index.html files
    $ wget -r --no-parent --reject "index.html*" ftp://gmaoftp.gsfc.nasa.gov/pub/data/sarith/TianZhou/
  • using ncftpget to download files from ftp sites
    $ ncftpget ftp://ftp.hydro.washington.edu/pub/tizhou/xxx.bin

Count number of files in a dir

$ ls -l | wc -l

Split an ascii file into several parts by line

$ split -l 2000 inflie.asc

Cut columns from a file with fixed delimiter (e.g. soil parameter)

-d: delimiter, default is tab -f: cut by field; -c: cut by characters; -b: cut by bytes

cut -d " " -f 1-53 soil.current > newsoil   #cut soil.current 1-53 columns by space

See the first and last 10 lines of a text file

$ (head;echo;tail) < file.txt


  • file I/O output 1st and 3rd columns

$ awk '{print $1, $3}' PNW.KW.flow.151dams > junk.txt

it can also read some columns from one file and write some of the columns to another file -F is to define the delimiter in the read in file.

$ awk -F',' '{printf("%.1f,%.1f,%.4f\n", $4, $5, $3)}' in.csv > out.txt

Shortcuts and tricks

  • Change directory to your previous location $ cd -
  • Last command argument:!$ example:
    $ mkdir temp
    $ cd !$
  • Last command argument: !! example:
    $ ping google.c
    $ ping !!om     # this will give you "ping google.com"
  • Combine multiple lines to one line
    $ less test.txt | - - -> newtest.txt   
    # every three lines in old file goes to one line in new file