Hey there, let's break down each of your questions with practical, ready-to-use examples that you can test right away:

The core syntax for sed's search-and-replace operation is simple and flexible:

sed 's/search_term/replacement/flags' input_file

• s/: Marks the start of a substitute command
• flags: Optional modifiers to tweak behavior:
  • g: Replace all occurrences in a line (not just the first match)
  • i: Ignore case when searching
  • -i: Edit the file in-place (use with caution—always back up files first!)

Example: Replace every instance of "foo" with "bar" in example.txt:

sed 's/foo/bar/g' example.txt

To save changes directly to the file:

sed -i 's/foo/bar/g' example.txt

To target lines that begin with a specific string, use the ^ regex anchor (matches the start of a line) as a filter. The syntax looks like this:

sed '/^target_prefix/s/search/replace/flags' input_file

Example: Replace "ERROR" with "WARNING" only on lines that start with "LogEntry:":

sed '/^LogEntry:/s/ERROR/WARNING/g' system_logs.txt

If you want to replace the entire line that matches the prefix, use the c\ command:

sed '/^Deprecated:/c\// This line has been marked as deprecated' code_file.py

Using sed (best for large files or one-liners) #

Sed’s & character represents the entire matched string, making this task trivial. Just prepend three spaces to the match:

sed 's/abc\.xyz/   &/' input_file

Note the backslash before the dot (\.)—dots are special regex characters, so we escape it to match a literal dot. Add the g flag to replace every occurrence in a line:

sed 's/abc\.xyz/   &/g' input_file

Using Bash (for small files or inline processing) #

If you prefer a pure bash approach, use parameter expansion in a loop to modify each line:

while IFS= read -r line; do
  echo "${line//abc.xyz/   abc.xyz}"
done < input.txt > output.txt

This reads each line, replaces all instances of "abc.xyz" with the string prefixed by three spaces, and writes the result to output.txt.

内容的提问来源于stack exchange，提问作者preethy tulpi