การแก้ไขสตรีมในระดับขนาดใหญ่ด้วย sed และ tr

You have mastered regex matching with grep and find. The natural next step is transforming the text you match. sed and tr turn your regex skills into edits.

The s/pattern/replacement/ command replaces the first match per line. Add g for all matches.

sed -E enables extended regex so you can use +, ?, and groups without backslashes.

Parentheses capture parts of the match; \1, \2 reuse them in the replacement. Here we swap two words.

Use ^ and $ to act only at the start or end of a line, e.g. prefixing every line.

sed can target lines by pattern: /regex/d deletes matching lines, while -n '/regex/p' prints only matches.

sed -i edits a file directly. Always keep a backup with -i.bak until you trust the command.

tr translates or deletes characters (not regex). It reads stdin only. Here we uppercase text.

tr -d removes characters and tr -s squeezes repeats into one, great for cleaning whitespace.

Chain them: normalize a CSV by converting commas to newlines and stripping carriage returns.

Reach for sed when you need patterns and line logic; reach for tr for fast single-character translations and deletions.

You want to replace every occurrence of a word on each line with sed. What must you add?

You can now transform streams with regex:

• sed 's///g' substitutes; -E enables extended regex; \1 reuses captures
• /re/d and -n '/re/p' select lines
• sed -i.bak edits files safely
• tr translates, deletes (-d), and squeezes (-s) characters