diff
TLDR
Compare files
$ diff [file1.txt] [file2.txt]
Unified format$ diff -u [file1.txt] [file2.txt]
Compare directories$ diff -r [dir1] [dir2]
Side-by-side comparison$ diff -y [file1.txt] [file2.txt]
Ignore whitespace$ diff -w [file1.txt] [file2.txt]
SYNOPSIS
diff [options] file1 file2
DESCRIPTION
diff compares files line by line and outputs the differences. It's essential for version control, code review, and understanding changes between file versions.
The tool forms the basis of patch files and is used by version control systems.
PARAMETERS
-u, --unified
Unified diff format (recommended)-c, --context
Context diff format-y, --side-by-side
Side-by-side comparison-r, --recursive
Recursively compare directories-q, --brief
Report only if files differ-s, --report-identical-files
Report when files are identical-w, --ignore-all-space
Ignore whitespace changes-b, --ignore-space-change
Ignore whitespace amount changes-B, --ignore-blank-lines
Ignore blank line changes-i, --ignore-case
Case-insensitive comparison-N, --new-file
Treat missing files as empty
OUTPUT FORMATS
Normal format:
$ 3c3
< old line
---
> new line
Unified format (-u):< old line
---
> new line
$ --- file1.txt
+++ file2.txt
@@ -1,3 +1,3 @@
context
-old line
+new line
context
Context format (-c):+++ file2.txt
@@ -1,3 +1,3 @@
context
-old line
+new line
context
$ *** file1.txt
--- file2.txt
***************
*** 1,3 ****
context
! old line
context
--- 1,3 ----
context
! new line
context
--- file2.txt
***************
*** 1,3 ****
context
! old line
context
--- 1,3 ----
context
! new line
context
WORKFLOW
$ # Basic comparison
diff file1.txt file2.txt
# Unified diff (best for patches)
diff -u original.txt modified.txt
# Save as patch
diff -u original.txt modified.txt > changes.patch
# Compare directories
diff -r old_version/ new_version/
# Ignore whitespace
diff -w file1.txt file2.txt
# Brief output
diff -q file1.txt file2.txt
# Side by side
diff -y file1.txt file2.txt | less
diff file1.txt file2.txt
# Unified diff (best for patches)
diff -u original.txt modified.txt
# Save as patch
diff -u original.txt modified.txt > changes.patch
# Compare directories
diff -r old_version/ new_version/
# Ignore whitespace
diff -w file1.txt file2.txt
# Brief output
diff -q file1.txt file2.txt
# Side by side
diff -y file1.txt file2.txt | less
CREATING PATCHES
$ # Create patch
diff -u old.c new.c > fix.patch
# Apply patch
patch old.c < fix.patch
# Reverse patch
patch -R new.c < fix.patch
diff -u old.c new.c > fix.patch
# Apply patch
patch old.c < fix.patch
# Reverse patch
patch -R new.c < fix.patch
DIRECTORY COMPARISON
$ # Compare directories
diff -r dir1/ dir2/
# Exclude files
diff -r --exclude="*.log" dir1/ dir2/
# Brief directory diff
diff -rq dir1/ dir2/
diff -r dir1/ dir2/
# Exclude files
diff -r --exclude="*.log" dir1/ dir2/
# Brief directory diff
diff -rq dir1/ dir2/
EXIT STATUS
- 0 - Files are identical
- 1 - Files differ
- 2 - Error occurred
CAVEATS
Large files can be slow. Binary files show as "differ" without details. Whitespace handling varies. Line endings (CRLF vs LF) can cause issues. Not suitable for binary comparison (use cmp).
HISTORY
diff was created by Doug McIlroy and colleagues at Bell Labs around 1974, with the algorithm published by Hunt and McIlroy in 1976.
