What are common sed patterns?

In sed, patterns are usually defined using Regular Expressions (Regex). Here are some of the most common patterns you will use to manipulate text:

1. Matching the Start and End of a Line

• ^ (Caret): Matches the start of a line.
  • Example: sed 's/^/Starting: /' file (Adds "Starting: " to the beginning of every line).
• $ (Dollar): Matches the end of a line.
  • Example: sed 's/$/./' file (Adds a period to the end of every line).

2. Matching Characters

• . (Dot): Matches any single character.
  • Example: sed 's/h.t/hit/g' (Matches "hat", "hot", "hbt" and replaces them with "hit").
• * (Asterisk): Matches zero or more of the preceding character.
  • Example: sed 's/be*/bee/g' (Matches "b", "be", "bee" and replaces them).

3. Character Classes

• [ ] (Brackets): Matches any one character inside the brackets.
  • Example: sed 's/[0-9]/X/g' (Replaces every digit with "X").
  • Example: sed 's/[a-z]/L/g' (Replaces every lowercase letter with "L").

4. Deleting Lines

You can use patterns with the d command to remove lines:

• Delete blank lines: sed '/^$/d' filename
• Delete lines containing a word: sed '/temporary/d' filename

5. Using Alternate Delimiters

As mentioned in your lab, if you are working with paths (which contain many /), you can change the character used for the pattern:

• Instead of s/old/new/, use s#old#new# or s|old|new|.
  • Example: sed 's#/usr/bin#/usr/local/bin#g'

Summary Table

Would you like to try practicing one of these patterns on your sed_test.txt file?