4.8 grep and regular expressions

4.8 grep and regular expressions

Regular expressions are used in many UNIX tools. The string substitution used in grep, ex, and vi use regular expressions. The shell, in its file name expansion, uses a simplified regular expression format. awk and egrep use extended regular expressions.

a specific regular expression. It does wonders in finding lost functions and subroutines in a group of program sources, or checking to see if you have removed all references to an obsolete function or variable. An option of -n will cause the line number to be printed.

all three. Its options are -l, -w and -c, for lines, words and characters. The default is -lwc.

In this lab you will search, using the program grep. an old "C" program of mine named cman6.c. It is a Mandelbrot set program written for the Borland "C" compiler and some graphics library additions. The file cman6.c is available by anonymous ftp at ftp://lt.tucson.az.us/pub/cman6.c.

Pipes are the use of the "|" character to direct the output of one command into the input of another command. For example:

grep 'define' ~cis137/cman6.c | wc -l

Will extract all lines from ~cis137/cman6.c that have the string define and send them onto wc. wc will then count the number of lines. grep also has a -c option that will count the lines that match.

Another interesting option to grep is -v. This will cause grep to invert the condition. If the regular expression matches a line it will NOT be output. However, all the lines that didn't match will be output instead.

The shell, when it is processing a command line, first looks for pipe and redirection symbols. If it finds these it effectively removes these from the command line. The individual commands are unaware that they ever existed. Then it searches for and replaces variables and file names with wild cards. Afterwards it executes the individual commands.

Write a shell script called greplab that searches ~cis137/cman6.c for the various items below.

Use the echo command to print out your name and TABER CIS137. For each item also provide a short description of what you are printing out.

Start your shell script with #!/bin/bash to use the bash shell.

1. The line number and the line of all lines that have the
   string "closegraph".
2. The number of lines that have the string "include".
3. The number of lines with a period ".". (Be careful "." is a meta character.)
4. The number of lines with a greater than symbol ">". (Be careful ">" is a shell redirection character.)
5. The line number and the line where the string "main(" is. This function is where a "C" program starts running.
6. The number of lines that have the string "colormap".
7. The number of lines that have the string "/*". (Be careful "*" is a meta character.)

Watch out for the characters being interpreted by the shell with the use of single quotes. Escape special regular expression characters.

Turn in a copy of your shell script and its results. Make a printout of your output and shell script, and mark it with:

your name
TABER CIS137
Lab 4.8: grep & regular expressions

Please turn your lab to Louis Taber or to Pima Community College employee in room A-115 of the Santa Rita Building. Ask them to place it in the dark blue folder in Louis Taber's mailbox.

4.8.1 shell rules

See the sh(1) & csh(1) manual pages for a complete description.

• * Matches any string including a null string.
• ? Matches any single character.
• [ ] Matches any enclosed character. A range of characters can be specified with a "-". [a-d] == [abcd] If the first character following a "[" is a "^" then any character NOT enclosed is matched. bash also lets you use ! to negate the list.
• Periods at the start of file names must be matched explicitly.
• Forward slashes "/"must be matched explicitly.
• If no match is found, no substitution is done. (sh only).
• Matching filenames are alphabetized after a substitution is made.
• A "\" can be used prior to "*" & "?" to look for "*" & "?".
• A list "{i1,i2,...}" expands list - bash, csh, and tcsh.

4.8.2 regular expression rules

See the ed(1) manual pages for a complete description.  

• . Is a one character regular expression that matches any character.
• * Matches 0 or more of the preceding one character regular expression.
• Inside square brackets "[ ]" matches any enclosed character. A range can be specified with a "-". [a-d] == [abcd]. If the first character following a "[" is a "^" then any character not enclosed is matched.
• ^ at the beginning of the regular expression forces the regular expression to match at the beginning of a line.
• $ at the end of the regular expression forces the regular expression to match the final segment of a line.
• A "\" can be used prior to any special character if what is wanted is a match of that character. \ * ^ $ [ ] .

4.8.3 extended regular expression rules

See the egrep(1) manual pages for a complete description. egrep can run up to 10 times faster than grep. Its memory usage is less predictable. awk also uses extended regular expressions.

• All regular expression rules.
• + Matches 1 or more of the preceding regular expression.
• ? Matches 0 or 1 of the preceding regular expression.
• | Between two regular expressions | will match if either expression matches.
• ( ) Expressions may be enclosed in parentheses for grouping.

4.8 grep and regular expressions