grep

grep prints lines that match a pattern. It is the standard Unix pattern-matching tool, available in three regular-expression dialects: basic (BRE), extended (ERE), and Perl-compatible (PCRE, when compiled in).

Essential options

# Case-insensitive match
grep -i "pattern" file.txt
 
# Recursive search through a directory tree
grep -r "pattern" /path/to/dir
 
# Invert match (show lines that do NOT match)
grep -v "pattern" file.txt
 
# Show line numbers
grep -n "pattern" file.txt
 
# Count matches only
grep -c "pattern" file.txt
 
# Extended regex (unescaped +, ?, |, etc.)
grep -E "pattern+" file.txt

Regular expression notes

• In BRE (default), the metacharacters ?, +, {, |, (, and ) lose their special meaning; escape them with \?, \+, \{, \|, \(, \) to enable them.
• In ERE (-E), those metacharacters are special by default.
• PCRE (-P) enables lookaheads, lookbehinds, and non-greedy quantifiers where supported.
• \w matches word characters ([0-9a-zA-Z_]); see regex-metacharacters for the full class table.

Related: findstr (Windows equivalent), regex-metacharacters, sed-single-line-find-replace.

Sources

• [grep(1) — Linux manual pagegrep(1) — Linux manual page

See also

• findstr
• regex-metacharacters
• sed-single-line-find-replace