tr — Character Translation

Character-level translation, deletion, and squeeze — the preprocessing step before sort, awk, or grep.

Character Translation

Translate lowercase to uppercase

echo "hello world" | tr 'a-z' 'A-Z'

Translate uppercase to lowercase

echo "HELLO WORLD" | tr 'A-Z' 'a-z'

Using POSIX character classes for locale-safe case conversion

echo "hello world" | tr '[:lower:]' '[:upper:]'

Replace spaces with underscores — filename sanitization

echo "my file name.txt" | tr ' ' '_'

Replace colons with newlines — break PATH into lines

echo "$PATH" | tr ':' '\n'

ROT13 — simple letter rotation cipher

echo "hello" | tr 'a-zA-Z' 'n-za-mN-ZA-M'

Delete Characters — -d

Delete all digits

echo "abc123def456" | tr -d '[:digit:]'

Delete all whitespace

echo "  h e l l o  " | tr -d '[:space:]'

Strip carriage returns — fix Windows line endings

tr -d '\r' < windows.txt > unix.txt

Remove all non-printable characters

tr -d '[:cntrl:]' < binary_mixed.txt

Delete specific characters — strip quotes

echo '"quoted"' | tr -d '"'

Squeeze Repeats — -s

Squeeze repeated spaces into one — normalize whitespace

echo "too    many     spaces" | tr -s ' '

Squeeze repeated newlines — collapse blank lines

tr -s '\n' < multi_blank.txt

Squeeze tabs to single tab

tr -s '\t' < messy.tsv

Complement — -c

Delete everything that is NOT a letter or newline — keep only alphabetic chars

echo "abc123!@#def" | tr -cd '[:alpha:]\n'

Keep only digits and newlines

echo "phone: (555) 123-4567" | tr -cd '[:digit:]\n'

Replace all non-alphanumeric chars with dashes — slug generation

echo "Hello, World! (2026)" | tr -cs '[:alnum:]' '-' | tr 'A-Z' 'a-z'

Delete everything except printable characters — sanitize input

tr -cd '[:print:]\n' < suspect.txt

Newline Handling

Join all lines into one — replace newlines with spaces

tr '\n' ' ' < multiline.txt && echo

Join lines with commas — then trim trailing comma

tr '\n' ',' < list.txt | sed 's/,$/\n/'

Split on commas to newlines

echo "a,b,c,d,e" | tr ',' '\n'

Replace null bytes with newlines — process find -print0 output without xargs

find /var/log -name "*.log" -print0 | tr '\0' '\n'

Character Classes Reference

Available POSIX character classes

# [:upper:]   Uppercase letters
# [:lower:]   Lowercase letters
# [:alpha:]   All letters
# [:digit:]   Digits 0-9
# [:alnum:]   Letters and digits
# [:space:]   Whitespace (space, tab, newline, etc.)
# [:blank:]   Space and tab only
# [:print:]   Printable characters (includes space)
# [:graph:]   Printable characters (excludes space)
# [:cntrl:]   Control characters
# [:punct:]   Punctuation characters
# [:xdigit:]  Hexadecimal digits (0-9, a-f, A-F)

Practical Pipelines

Normalize a file — lowercase, squeeze spaces, strip non-printable

tr 'A-Z' 'a-z' < input.txt | tr -s '[:space:]' | tr -cd '[:print:]\n'

Word frequency — split words, count

tr -cs '[:alpha:]' '\n' < document.txt | tr 'A-Z' 'a-z' | sort | uniq -c | sort -rn | head -20

Generate random password characters — use /dev/urandom with tr filter

tr -dc '[:alnum:]!@#$%' < /dev/urandom | head -c 24 && echo

Convert tabs to spaces — simple formatting

tr '\t' '    ' < tabbed.txt