Syntax
Description
The wc (word count) command counts lines, words, and bytes in text input. It’s a fundamental tool for analyzing file sizes, counting records, and understanding text structure.
By default, wc displays all three counts. You can specify flags to show only specific counts.
Nash’s wc operates entirely in-memory and supports reading from files or standard input.
Options
Count lines only. Output the number of newline characters.
Count words only. Words are whitespace-separated tokens.
Count bytes only. Output the total number of bytes (characters).
When no options are specified, wc displays lines, words, and bytes (in that order).
When multiple options are specified, counts appear in the order: lines, words, bytes.
Examples
Default output (all counts)
echo -e "hello world\nfoo bar" | wc
lines words bytes
- 2 lines
- 4 words (hello, world, foo, bar)
- 20 bytes
Count lines only
echo -e "line 1\nline 2\nline 3" | wc -l
Count words only
echo "the quick brown fox" | wc -w
Count bytes only
Note: Includes the newline character added by echo.
Multiple flags
echo -e "one two\nthree four" | wc -l -w
Count file contents
echo -e "line 1\nline 2" > test.txt
wc test.txt
Pipeline Examples
Count matches
grep ERROR app.log | wc -l
Count unique values
cat data.txt | sort -u | wc -l
Words in a document
Total word count.
File size approximation
cat large-file.txt | wc -c
Count after filtering
ls | grep "\.txt$" | wc -l
.txt files in directory.
Average line length
BYTES=$(cat file.txt | wc -c)
LINES=$(cat file.txt | wc -l)
echo "Average: $((BYTES / LINES)) bytes per line"
Practical Use Cases
Verify CSV row count
Quick check of record count.
Check if file is empty
test $(cat file.txt | wc -l) -eq 0 && echo "Empty file"
Count log entries by level
echo "Errors: $(grep ERROR app.log | wc -l)"
echo "Warnings: $(grep WARN app.log | wc -l)"
echo "Info: $(grep INFO app.log | wc -l)"
Estimate reading time
WORDS=$(cat article.txt | wc -w)
MINS=$((WORDS / 200))
echo "Estimated reading time: $MINS minutes"
Monitor log growth
echo "Log size: $(cat app.log | wc -l) lines"
Count code lines
cat *.js | grep -v "^$" | grep -v "^[ ]*//" | wc -l
Data validation
EXPECTED=1000
ACTUAL=$(cat output.txt | wc -l)
test $ACTUAL -eq $EXPECTED && echo "OK" || echo "Mismatch: $ACTUAL rows"
Advanced Examples
Compare file sizes
echo "file1.txt: $(cat file1.txt | wc -c) bytes"
echo "file2.txt: $(cat file2.txt | wc -c) bytes"
Count fields in CSV
LINES=$(cat data.csv | wc -l)
FIELDS=$(head -1 data.csv | tr ',' '\n' | wc -l)
echo "$LINES rows × $FIELDS columns"
Average words per line
LINES=$(cat doc.txt | wc -l)
WORDS=$(cat doc.txt | wc -w)
echo "Average: $((WORDS / LINES)) words/line"
Multiple files
wc file1.txt file2.txt file3.txt
Conditional processing
if [ $(cat errors.log | wc -l) -gt 100 ]; then
echo "Too many errors!"
fi
%8d for each count.
Order is always: lines, words, bytes.
Understanding Counts
Number of newline characters (\n).echo -e "a\nb\nc" | wc -l
3Files without a trailing newline may count one fewer line than expected.
Number of whitespace-separated tokens.echo "one two three" | wc -w
3Multiple spaces count as one separator. Total character count including newlines.Output: 4 (3 letters + 1 newline) Common Patterns
Count pipeline results
Standard way to count output lines.
Check file emptiness
test $(cat file.txt | wc -l) -gt 0 && echo "Has content"
Sum counts across files
Shows individual and total counts.
Store count in variable
COUNT=$(cat data.txt | wc -l)
echo "Found $COUNT records"
Percentage calculation
TOTAL=$(cat all.txt | wc -l)
FILTERED=$(cat all.txt | grep pattern | wc -l)
PERCENT=$((FILTERED * 100 / TOTAL))
echo "$PERCENT% match"
Tips
Newline counting: wc -l counts newline characters. A file without a trailing newline may show one fewer line than you expect.
echo -n "" | wc -l # Output: 0 (no newline)
echo "" | wc -l # Output: 1 (one newline)
Word definition
Words are separated by any whitespace: spaces, tabs, newlines.
echo -e "one\ttwo\nthree" | wc -w # Output: 3
Byte vs character
For ASCII text, bytes = characters. For Unicode, may differ (not applicable in Nash’s VFS).
Combining with other commands
Always use wc at the end of pipelines:
grep pattern file | sort | uniq | wc -l
Nash’s wc loads entire input into memory for counting. This is efficient for typical text files in the VFS but consider the memory footprint for very large datasets.
- grep - Filter lines before counting
- sort - Organize before counting
- uniq - Remove duplicates before counting
- cut - Extract fields for counting
- head - Limit lines
- tail - View last lines
