grep : Used to grab strings. Use -i to ignore upper-/lower- casese, and -v for inverse match.Character Class:
[[:alpha:]] is [A-Za-z], [[:digit:]] is [0-9], [[:upper:]] is [A-Z], [[:lower:]] is [a-z], [[:alnum:]] is [0-9A-Za-z], [[:blank:]] is either blank space or tab, and [[:xdigit:]] is [0-9A-Fa-f].[[:punct:]] are ! " # $ % & ' ( ) * + , - . / : ; < = > ? @ [ \ ] ^ _ ` { | } ~[[:graph:]] matches [:alnum:] and [:punct:][^[:digit:]] matches anything except numbers, and is equivalent to [^0-9]
Repetition
A regular expression may be followed by one of several repetition
operators:
? The preceding item is optional and matched at most once.
* The preceding item will be matched zero or more times.
+ The preceding item will be matched one or more times.
{n} The preceding item is matched exactly n times.
{n,} The preceding item is matched n or more times.
{,m} The preceding item is matched at most m times. This is a GNU
extension.
{n,m} The preceding item is matched at least n times, but not more
than m times.
The -E switch enables extended RE (ERE), which is equivalent to egrep.The
-P switch enables PCRE, which works conveniently with shorthand classes.For legacy NCTU student IDs in the format of
u1234567 :
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd
grep "[ugd][[:digit:]a][[:digit:]]\{6\}" /etc/passwd
grep -E "[udg][0-9a][0-9]{6}" /etc/passwd
grep -E [udg][0-9a][0-9]{6} /etc/passwd
egrep [udg][0-9a][0-9]{6} /etc/passwd
grep -P "[udg][0-9a]\d{6}" /etc/passwd
grep -P "[udg][\da]\d{6}" /etc/passwd
All of the above commands give the same result.For new-style NYCU student IDs in the format of
u123456789 :
grep "[udg][0-9]\{9\}" /etc/passwd
grep "[ugd][[:digit:]]\{9\}" /etc/passwd
grep -E "[udg][0-9]{9}" /etc/passwd
grep -E [udg][0-9]{9} /etc/passwd
egrep [udg][0-9]{9} /etc/passwd
grep -P "[udg]\d{9}" /etc/passwd
All of the above commands give the same result.The
-e switch defines multiple patterns. This can be regarded as an equivalence to OR operation.To grab both of NCTU and NYCU student ID patterns in
grep -e "[udg][0-9a][0-9]\{6\}" -e "[udg][0-9]\{9\}" /etc/passwd
grep -e "[ugd][[:digit:]a][[:digit:]]\{6\}" -e "[ugd][[:digit:]]\{9\}" /etc/passwd
grep -E -e "[udg][0-9a][0-9]{6}" -e "[udg][0-9]{9}" /etc/passwd
grep -E -e [udg][0-9a][0-9]{6} -e [udg][0-9]{9} /etc/passwd
egrep -e [udg][0-9a][0-9]{6} -e [udg][0-9]{9} /etc/passwd
grep -P -e "[udg][0-9a]\d{6}" -e "[udg]\d{9}" /etc/passwd
or, in this special case, combine the two patterns together:
grep "[udg][0-9a][0-9]\{6\}\|\{8\}" /etc/passwd
sed : stream editor. Actions are enclosed within a pair of single quotes ' ', usually something like 's/PATTERN1/PATTERN2/g'Use
-e to combine multiple patterns.Practice: Write only one line commands to convert /etc/passwd into HTML file, using
sed.grep -E -e ^[udg][0-9a][0-9]{6} -e ^[udg][0-9]{9} /etc/passwd | sed -e 's/^/<a href=\"http:\/\/ukko\.life\.nctu\.edu\.tw\/~/' -e 's/:x.*s\//\">/' -e 's/:\/bin.*$/<\/a><br>/'
Question: What if single quote is one of the patterns?
wc : Word counter wc /etc/profileman : Check manual of the commands.bc : basic calculator , online manualecho "3+5" | bc echo "3-5" | bc echo "3*5" | bc echo "3%5" | bc echo "3/5" | bc -l echo "scale=8; 3/5" | bc -l echo "scale=8; l(2.71828)" | bc -l # calculate ln(2.71828) echo "scale=20; 4*a(1)" | bc -l # calculate π with 4×arctan(1) to the 20th digit. bc -l <<< "scale=20; 4*a(1)" # HERE-STRING requires string as stdin bc -l <<< echo "scale=20; 4*a(1)" # incorrect, because stdout cannot be used as input bc -l <<< `echo "scale=20; 4*a(1)"` # backquotes ` ` or $( ) converts stdout into string bc -l <<< $(echo "scale=20; 4*a(1)") # it is suggested to use $( ) instead of backquotes ` `Useful resource: http://x-bc.sourceforge.net .
bash with awk
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{print $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{print $3 $4}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{print $3, $4}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{print $4, $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | cut -d: -f3,4
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | cut -d: -f4,3
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{print $3 " " $4}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "\n" $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%d", $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%d\n", $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%4d\n", $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%5d\n", $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%05d\n", $3}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%05d,%05d\n", $3, $4}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%05d|%05d\n", $3, $4}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%05d\\%05d\n", $3, $4}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%05d%%%05d\n", $3, $4}'
grep "[udg][0-9a][0-9]\{6\}" /etc/passwd | awk -F: '{printf "%05d\x27%05d\n", $3, $4}'
bash code for the sum of infinite series 1+2+3+4+5+...+n , using a (stupid) loop:
#!/bin/bash counter=$1 # comments start with hashtag only sum=0 # initial value of $sum while [ $counter -gt 0 ] # -gt means grater than do sum=$(( $sum + $counter )) # $(( )) performs interger math under bash counter=$(( $counter - 1 )) done echo $sum # end of code |
|
If you save the bash script above as chmod +x ~/sum.bash before you can calculate 1+2+3+4+5+...+10 by ~/sum.bash 10 |
Question: How to make the code run faster?
Practice: Write a
bash script to calculate the factorial n! .seq : sequence generator for integersseq 1 10 # print integers 1 to 10, one integer per line seq -s " " 1 10 # print integers 1 to 10 in one single line separated by one space seq -s " " 1 2 11 # 1 3 5 7 9 11 seq -s " " 10 -1 1 # 10 9 8 7 6 5 4 3 2 1 seq 1 10 | tr -d "\n" # 12345678910 without newline
