Lniux-AWK use

AWK

The difference between AWK and sed

• AWK is used for more standardized text processing, used to count the number and output the specified field
• sed is used to process non-standard text into more standardized text

AWK field reference and separation

• AWK is also read by line. Each line is called a record. Words separated by spaces and tabs are called fields. You can specify the separator to separate fields.
• Use $ 1 $ 2 ... $ n for each field, and $ 0 for an entire line
  • awk '{print $1,$2,$3}' filename
  • awk -F "'"' / ^ menu / {print $ 2} '/boot/grub2/grub.cfg to split the field with a', remove the kernel information in the file
  • {print x ++, $ 2} x ++ can display the number of lines
• -F option is used to specify the separator
  • awk -F ',' '{print $1,$2,$3}' filename
  • Delimiters can be expressed using regular expressions

AWK expression

• System variables

  • FS and OFS field separators, OFS indicates the output field separators

    • head -5 / etc / passwd | awk 'BEGIN {FS = ":"} {print $ 1}'
      BEGING, pre-process before reading the file. Before reading the passwd file, specify the ":" as the delimiter through the FS option.
    • head -5 / etc / passwd | awk 'BEGIN {FS = ":"; OFS = "-"} {print $ 1, $ 2}' OFS = "-", OFS is after finding the field, the system uses spaces by default Separation, and OFS changed the default separation symbol.

  • RS line record separator, record line, RS is a newline by default, when multiple lines are combined into a single line, use RS to process

    • head -5 / etc / passwd | awk 'BEGIN {RS = ":"} {print $ 1}' The system default line record separator is a line break, RS = ":", replace the line record separator with:

  • NR and FNR rows

    • head -5 / etc / passwd | awk '{print NR, $ 0}' NR shows the line number, $ 0 shows the line content
    • awk '{ print NR,$0 }' /etc/hosts /etc/hosts
    • awk '{print FNR, $ 0}' / etc / hosts / etc / hosts
      For single files, there is no difference between using NR and FNR. If you connect multiple files, NR will always sort the content in order, without distinguishing files; FNR is based on Sort the files first, and then sort them individually.

  • The number of NF fields, the content of the last field can be taken out with $ NF

    • head -5 / etc / passwd | awk 'BEGIN {FS = ":"} {print NF}' outputs how many fields are in each line