Linux with Operating System Concepts

address
?
21. What is the difference between 
mail –b 
address
and 
mail –c 
address
?
22. By default, where are emails stored until they are read?
23. What must be true of networked Linux computers for you to be able to use r-utilities 
between them?
24. How does ssh differ from telnet?
25. When using ftp, how does cd differ from lcd?
26. When using ftp, what command allows you to upload a file to the remote computer?
27. How does wget differ from ftp?

---

200
◾
Linux with Operating System Concepts
28. Explain why you might want to use ping. Explain why you might want to use tracer-
oute instead of ping.
29. Explain why you might want to use nslookup.
30. How does nslookup differ from dig and host?
31. What is the equivalent in host of the command dig (i.e., the version of host that pro-
vides detailed information like dig)?
32. How would you specify using dig that you only want to obtain information for the 
listings of NSs?
33. What is an SOA?

---

201
C h a p t e r
6
Regular Expressions
T
his chapter’s learning objectives are
• To understand the use of regular expressions
• To understand the usage of the metacharacters of regular expressions
• To understand how to apply each of the metacharacters
• To be able to use grep/egrep to search for files for strings
• To be able to use the basic forms of sed and awk to solve search problems
6.1 INTRODUCTION
A regular expression (regex) is a string that expresses a pattern used to match against other 
strings. The pattern will either match some portion of another string or not. We can use 
regular expressions to define pattern matching rules which can then be used by programs. 
For instance, regular expressions are applied in software that performs spam filtering and 
natural language understanding.
Let us look at a concrete example to illustrate the problem that a regular expression can 
solve. You want to build a spam filter. You need to come up with a way to filter out any 
email message that contains the word Viagra. Searching for the literal string “Viagra” or 
“viagra” is easy enough. But, clever spammers will try to disguise the word by using sub-
stitution letters, adding characters or removing letters in the word. We might expect any 
of the following variations:
• v.i.a.g.r.a
• v1agra
• vi_ag_ra
• vi@gr@

---

202
◾
Linux with Operating System Concepts
• ViAgRa
• Viagr
To build our spam filter, we would not want to list every possible potential appearance of 
Viagra. For the six examples listed here, there are thousands of others that someone might 
come up with. Instead, we could define a single regular expression that could cover many 
or most of the possibilities.
Let us consider another example. We want to define a regular expression to match any 
person’s full name (we will assume just first and last name). A name starts with a capital 
letter and is followed by lower-case letters. This defines any capitalized word. To differenti-
ate, a name will consist of two of these with a space in between. We can define this regex as 
consisting of an upper-case letter followed by some number of lower-case letters followed 
by a space followed by a capital letter followed by some lower-case letters. Note that our 
regex does not ensure that the names found are actually people’s names. With the regular 
expression defined, we can use a Linux program, for instance grep, to scan through a file 
to locate all of the strings that are names.
To define a regular expression, you write a pattern that consists of