> For the complete documentation index, see [llms.txt](https://www.shellspells.net/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://www.shellspells.net/linux/file-ops/sort-compare-count.md). # Sort / Compare / Count Read and insert file content ```bash sed -r ``` ## Sort

Command	Description
`sort [file]`	Alphabetical, ascending
`cat [file] \| sort`	Pipe output to `sort`.
`sort -n`	Sort numerically.
`sort -r`	Reverses the result of comparisons, producing a descending order sort.
`sort -k`	Specifies a sort key.
`sort -k2`	Sorts by second column.
`sort -k 2n,2`	Sorts numerically based on the second field, ignoring leading whitespace in that field.
`sort -k 1b`	Sorts by the first word in each line, ignoring case distinctions (`-b`).
`sort -u`	Outputs only the first of an equal run.
`sort -t`	Sets the field delimiter.
`sort -t ',' -k 2`	Sort lines by the second field

Sorts data.csv first by the second field numerically in descending order, then by the first field ignoring case: ```bash sort -k 2n,2r -k 1f data.csv ``` Sorts ps output by CPU% descending, then by PID (process ID): ```bash ps aux | sort -nrk 3,3 -k 8 ``` ## Compare ### Choosing the right command

Command	Description
`diff`	For line-by-line text comparison
`cmp`	For byte-level comparison
`comm`	For comparing sorted lists
`diffstat`	For quick statistics
`meld`	For visual comparison
`md5sum`	For verifying file integrity
`sha256sum`	For verifying file integrity

### Commands

Command	Description
`diff <file1> <file2>`	Compare two files line by line.
`diff -n <file1> <file2>`	Suppresses line numbers.
`diff -i <file1> <file2>`	Ignore case differences in file contents.
`diff -u <file1> <file2>`	Produce a unified format diff, showing several lines of context.
`diff -q <file1> <file2>`	Report only when files differ, without detailing the differences.
`diff -w <file1> <file2>`	Ignore all whitespace changes.
`diff -b <file1> <file2>`	Ignore changes in the amount of whitespace.
`diff --left-column <file1> <file2>`	Show line numbers for differences.
`cmp <file1> <file2>`	Byte-by-byte comparison.
`cmp -l <file1> <file2>`	Show all differences between the files.
`cmp -s <file1> <file2>`	Silent mode, do not write anything, just return an exit status.
`comm <file1> <file2>`	Compare two sorted files line by line.
`comm -1 -2 <file1> <file2>`	Suppress output of specified columns (1 and 2), showing only lines common to both files.
`diff3 <file1> <file2> file3.txt`	Compare three files line by line.
`diff3 -m <file1> <file2> file3.txt`	Show only conflicting changes among the three files.
`sdiff <file1> <file2>`	Side-by-side comparison of files.
`sdiff -i <file1> <file2>`	Ignore case differences in file contents.
`sdiff -s <file1> <file2>`	Suppress lines common to both files.
`diffstat diff_output`	Read a diff output file and display a histogram of insertions, deletions, and modifications per file.
`meld file1 file2`	Graphical tool to compare and merge files interactively.
`md5sum file`	Calculate MD5 checksum.
`sha256sum file`	Calculate SHA-256 checksum.

## Count

Command	Description
`wc [file]`	Count lines, words, and bytes.
`wc -l [file]`	Count lines.
`wc -w [file]`	Count words.
`wc -c [file]`	Count bytes.
`wc -m [file]`	Count characters.
`awk 'END {print NR}' [file]`	Count lines. `NR` represents the total number of records processed.
`awk '{count++} END {print count}' [file]`	Count lines.
`awk '/error/ {count++} END {print count}' [file]`	Count lines containing "error".
`sed '/^$/d' filename \| wc -l`	Counts non-empty lines.
`sed 's/[^a-zA-Z0-9 ]//g' filename \| wc -w`	Counts alphanumeric words
`sed -n '$=' [file]`	Count lines. `$=` command prints the total number of lines.
`grep pattern [file] \| wc -l`	Alternate way to count lines
`grep -c "[pattern]" [file]`	Lines with pattern match.
`grep -c "[pattern]" <file1> <file2>`	Miltiple files
`grep -o "[pattern]" [file] \| wc -l`	Count pattern matches, including multiple occurrences per line. `-o` outputs each match on a new line.
`grep -r -c "[pattern]" /directory`	Recursive counts lines with pattern match.
`find . -type f \| wc -l`	Count files in a directory and its subdirectories

Recursive by file extension: ```bash grep --include=\*. -ro "[pattern]" | wc -l ``` Recursive counts lines with pattern match: ```bash find / -type f -exec grep -o "[pattern]" {} + | wc -l ``` Uses `awk` to globally substitute (thus count) occurrences of the pattern in each line and sums them up: ```bash awk '{count += gsub(/[pattern]/, "[pattern]")} END {print count}' [file] ``` Count the total number of lines across all files in a directory and its subdirectories: ```bash find . -type f -exec wc -l {} + | awk '{total += $1} END {print total}' ```